feat: update dashboard to v0.11.9 (#7364) (#7371)

Signed-off-by: evenyag <realevenyag@gmail.com>
Co-authored-by: ZonaHe <zonahe@qq.com>
Co-authored-by: sunchanglong <sunchanglong@users.noreply.github.com>

.cargo/config.toml

@@ -12,3 +12,6 @@ fetch = true
 checkout = true
 list_files = true
 internal_use_git2 = false
+
+[env]
+CARGO_WORKSPACE_DIR = { value = "", relative = true }

.coderabbit.yaml

@@ -1,15 +0,0 @@
-# yaml-language-server: $schema=https://coderabbit.ai/integrations/schema.v2.json
-language: "en-US"
-early_access: false
-reviews:
-  profile: "chill"
-  request_changes_workflow: false
-  high_level_summary: true
-  poem: true
-  review_status: true
-  collapse_walkthrough: false
-  auto_review:
-    enabled: false
-    drafts: false
-chat:
-  auto_reply: true

.github/CODEOWNERS

@@ -4,7 +4,7 @@
 
 * @GreptimeTeam/db-approver
 
-## [Module] Databse Engine
+## [Module] Database Engine
 /src/index @zhongzc
 /src/mito2 @evenyag @v0y4g3r @waynexia
 /src/query @evenyag

.github/actions/build-linux-artifacts/action.yml

@@ -52,7 +52,7 @@ runs:
       uses: ./.github/actions/build-greptime-binary
       with:
         base-image: ubuntu
-        features: servers/dashboard,pg_kvbackend,mysql_kvbackend
+        features: servers/dashboard
         cargo-profile: ${{ inputs.cargo-profile }}
         artifacts-dir: greptime-linux-${{ inputs.arch }}-${{ inputs.version }}
         version: ${{ inputs.version }}
@@ -70,7 +70,7 @@ runs:
       if: ${{ inputs.arch == 'amd64' && inputs.dev-mode == 'false' }} # Builds greptime for centos if the host machine is amd64.
       with:
         base-image: centos
-        features: servers/dashboard,pg_kvbackend,mysql_kvbackend
+        features: servers/dashboard
         cargo-profile: ${{ inputs.cargo-profile }}
         artifacts-dir: greptime-linux-${{ inputs.arch }}-centos-${{ inputs.version }}
         version: ${{ inputs.version }}

.github/actions/release-cn-artifacts/action.yaml

@@ -64,11 +64,11 @@ inputs:
   upload-max-retry-times:
     description: Max retry times for uploading artifacts to S3
     required: false
-    default: "20"
+    default: "30"
   upload-retry-timeout:
     description: Timeout for uploading artifacts to S3
     required: false
-    default: "30" # minutes
+    default: "120" # minutes
 runs:
   using: composite
   steps:

.github/actions/setup-etcd-cluster/action.yml

@@ -12,7 +12,7 @@ runs:
   steps:
   - name: Install Etcd cluster
     shell: bash
-    run: | 
+    run: |
       helm upgrade \
         --install etcd oci://registry-1.docker.io/bitnamicharts/etcd \
         --set replicaCount=${{ inputs.etcd-replicas }} \
@@ -24,4 +24,9 @@ runs:
         --set auth.rbac.token.enabled=false \
         --set persistence.size=2Gi \
         --create-namespace \
+        --set global.security.allowInsecureImages=true \
+        --set image.registry=docker.io \
+        --set image.repository=greptime/etcd \
+        --set image.tag=3.6.1-debian-12-r3 \
+        --version 12.0.8 \
         -n ${{ inputs.namespace }}

.github/actions/setup-greptimedb-cluster/action.yml

@@ -10,13 +10,13 @@ inputs:
   meta-replicas:
     default: 2
     description: "Number of Metasrv replicas"
-  image-registry: 
+  image-registry:
     default: "docker.io"
     description: "Image registry"
-  image-repository: 
+  image-repository:
     default: "greptime/greptimedb"
     description: "Image repository"
-  image-tag: 
+  image-tag:
     default: "latest"
     description: 'Image tag'
   etcd-endpoints:
@@ -32,12 +32,12 @@ runs:
   steps:
   - name: Install GreptimeDB operator
     uses: nick-fields/retry@v3
-    with: 
+    with:
       timeout_minutes: 3
       max_attempts: 3
       shell: bash
       command: |
-        helm repo add greptime https://greptimeteam.github.io/helm-charts/ 
+        helm repo add greptime https://greptimeteam.github.io/helm-charts/
         helm repo update
         helm upgrade \
           --install \
@@ -48,10 +48,10 @@ runs:
           --wait-for-jobs
   - name: Install GreptimeDB cluster
     shell: bash
-    run: | 
+    run: |
       helm upgrade \
         --install my-greptimedb \
-        --set meta.etcdEndpoints=${{ inputs.etcd-endpoints }} \
+        --set meta.backendStorage.etcd.endpoints=${{ inputs.etcd-endpoints }} \
         --set meta.enableRegionFailover=${{ inputs.enable-region-failover }} \
         --set image.registry=${{ inputs.image-registry }} \
         --set image.repository=${{ inputs.image-repository }}  \
@@ -59,7 +59,7 @@ runs:
         --set base.podTemplate.main.resources.requests.cpu=50m \
         --set base.podTemplate.main.resources.requests.memory=256Mi \
         --set base.podTemplate.main.resources.limits.cpu=2000m \
-        --set base.podTemplate.main.resources.limits.memory=2Gi \
+        --set base.podTemplate.main.resources.limits.memory=3Gi \
         --set frontend.replicas=${{ inputs.frontend-replicas }} \
         --set datanode.replicas=${{ inputs.datanode-replicas }} \
         --set meta.replicas=${{ inputs.meta-replicas }} \
@@ -72,7 +72,7 @@ runs:
   - name: Wait for GreptimeDB
     shell: bash
     run: |
-      while true; do 
+      while true; do
         PHASE=$(kubectl -n my-greptimedb get gtc my-greptimedb -o jsonpath='{.status.clusterPhase}')
         if [ "$PHASE" == "Running" ]; then
           echo "Cluster is ready"
@@ -86,10 +86,10 @@ runs:
   - name: Print GreptimeDB info
     if: always()
     shell: bash
-    run: | 
+    run: |
       kubectl get all --show-labels -n my-greptimedb
   - name: Describe Nodes
     if: always()
     shell: bash
-    run: | 
+    run: |
       kubectl describe nodes

.github/actions/setup-greptimedb-cluster/with-remote-wal.yaml

@@ -7,7 +7,8 @@ meta:
     provider = "kafka"
     broker_endpoints = ["kafka.kafka-cluster.svc.cluster.local:9092"]
     num_topics = 3
-    auto_prune_topic_records = true
+    auto_prune_interval = "30s"
+    trigger_flush_threshold = 100
 
     [datanode]
     [datanode.client]
@@ -21,7 +22,7 @@ datanode:
     [wal]
     provider = "kafka"
     broker_endpoints = ["kafka.kafka-cluster.svc.cluster.local:9092"]
-    linger = "2ms"
+    overwrite_entry_start_id = true
 frontend:
   configData: |-
     [runtime]

.github/actions/setup-kafka-cluster/action.yml

@@ -12,7 +12,7 @@ runs:
   steps:
   - name: Install Kafka cluster
     shell: bash
-    run: | 
+    run: |
       helm upgrade \
         --install kafka oci://registry-1.docker.io/bitnamicharts/kafka \
         --set controller.replicaCount=${{ inputs.controller-replicas }} \
@@ -23,4 +23,8 @@ runs:
         --set listeners.controller.protocol=PLAINTEXT \
         --set listeners.client.protocol=PLAINTEXT \
         --create-namespace \
+        --set image.registry=docker.io \
+        --set image.repository=greptime/kafka \
+        --set image.tag=3.9.0-debian-12-r1 \
+        --version 31.0.0 \
         -n ${{ inputs.namespace }}

.github/actions/setup-postgres-cluster/action.yml

@@ -6,9 +6,7 @@ inputs:
     description: "Number of PostgreSQL replicas"
   namespace:
     default: "postgres-namespace"
-  postgres-version:
-    default: "14.2"
-    description: "PostgreSQL version"
+    description: "The PostgreSQL namespace"
   storage-size:
     default: "1Gi"
     description: "Storage size for PostgreSQL"
@@ -22,7 +20,11 @@ runs:
       helm upgrade \
         --install postgresql oci://registry-1.docker.io/bitnamicharts/postgresql \
         --set replicaCount=${{ inputs.postgres-replicas }} \
-        --set image.tag=${{ inputs.postgres-version }} \
+        --set global.security.allowInsecureImages=true \
+        --set image.registry=docker.io \
+        --set image.repository=greptime/postgresql \
+        --set image.tag=17.5.0-debian-12-r3 \
+        --version 16.7.4 \
         --set persistence.size=${{ inputs.storage-size }} \
         --set postgresql.username=greptimedb \
         --set postgresql.password=admin \

.github/labeler.yaml

@@ -0,0 +1,15 @@
+ci:
+  - changed-files:
+      - any-glob-to-any-file: .github/**
+
+docker:
+  - changed-files:
+      - any-glob-to-any-file: docker/**
+
+documentation:
+  - changed-files:
+      - any-glob-to-any-file: docs/**
+
+dashboard:
+  - changed-files:
+      - any-glob-to-any-file: grafana/**

.github/scripts/check-version.sh

@@ -0,0 +1,42 @@
+#!/bin/bash
+
+# Get current version
+CURRENT_VERSION=$1
+if [ -z "$CURRENT_VERSION" ]; then
+  echo "Error: Failed to get current version"
+  exit 1
+fi
+
+# Get the latest version from GitHub Releases
+API_RESPONSE=$(curl -s "https://api.github.com/repos/GreptimeTeam/greptimedb/releases/latest")
+
+if [ -z "$API_RESPONSE" ] || [ "$(echo "$API_RESPONSE" | jq -r '.message')" = "Not Found" ]; then
+  echo "Error: Failed to fetch latest version from GitHub"
+  exit 1
+fi
+
+# Get the latest version
+LATEST_VERSION=$(echo "$API_RESPONSE" | jq -r '.tag_name')
+
+if [ -z "$LATEST_VERSION" ] || [ "$LATEST_VERSION" = "null" ]; then
+  echo "Error: No valid version found in GitHub releases"
+  exit 1
+fi
+
+# Cleaned up version number format (removed possible 'v' prefix and -nightly suffix)
+CLEAN_CURRENT=$(echo "$CURRENT_VERSION" | sed 's/^v//' | sed 's/-nightly-.*//')
+CLEAN_LATEST=$(echo "$LATEST_VERSION" | sed 's/^v//' | sed 's/-nightly-.*//')
+
+echo "Current version: $CLEAN_CURRENT"
+echo "Latest release version: $CLEAN_LATEST"
+
+# Use sort -V to compare versions
+HIGHER_VERSION=$(printf "%s\n%s" "$CLEAN_CURRENT" "$CLEAN_LATEST" | sort -V | tail -n1)
+
+if [ "$HIGHER_VERSION" = "$CLEAN_CURRENT" ]; then
+  echo "Current version ($CLEAN_CURRENT) is NEWER than or EQUAL to latest ($CLEAN_LATEST)"
+  echo "should-push-latest-tag=true" >> $GITHUB_OUTPUT
+else
+  echo "Current version ($CLEAN_CURRENT) is OLDER than latest ($CLEAN_LATEST)"
+  echo "should-push-latest-tag=false" >> $GITHUB_OUTPUT
+fi

.github/scripts/create-version.sh

@@ -8,19 +8,20 @@ set -e
 # - If it's a nightly build, the version is 'nightly-YYYYMMDD-$(git rev-parse --short HEAD)', like 'nightly-20230712-e5b243c'.
 # create_version ${GIHUB_EVENT_NAME} ${NEXT_RELEASE_VERSION} ${NIGHTLY_RELEASE_PREFIX}
 function create_version() {
-  # Read from envrionment variables.
+  # Read from environment variables.
   if [ -z "$GITHUB_EVENT_NAME" ]; then
-      echo "GITHUB_EVENT_NAME is empty"
+      echo "GITHUB_EVENT_NAME is empty" >&2
       exit 1
   fi
 
   if [ -z "$NEXT_RELEASE_VERSION" ]; then
-      echo "NEXT_RELEASE_VERSION is empty"
-      exit 1
+      echo "NEXT_RELEASE_VERSION is empty, use version from Cargo.toml" >&2
+      # NOTE: Need a `v` prefix for the version string.
+      export NEXT_RELEASE_VERSION=v$(grep '^version = ' Cargo.toml | cut -d '"' -f 2 | head -n 1)
   fi
 
   if [ -z "$NIGHTLY_RELEASE_PREFIX" ]; then
-      echo "NIGHTLY_RELEASE_PREFIX is empty"
+      echo "NIGHTLY_RELEASE_PREFIX is empty" >&2
       exit 1
   fi
 
@@ -35,7 +36,7 @@ function create_version() {
   # It will be like 'dev-2023080819-f0e7216c'.
   if [ "$NEXT_RELEASE_VERSION" = dev ]; then
     if [ -z "$COMMIT_SHA" ]; then
-      echo "COMMIT_SHA is empty in dev build"
+      echo "COMMIT_SHA is empty in dev build" >&2
       exit 1
     fi
     echo "dev-$(date "+%Y%m%d-%s")-$(echo "$COMMIT_SHA" | cut -c1-8)"
@@ -45,7 +46,7 @@ function create_version() {
   # Note: Only output 'version=xxx' to stdout when everything is ok, so that it can be used in GitHub Actions Outputs.
   if [ "$GITHUB_EVENT_NAME" = push ]; then
     if [ -z "$GITHUB_REF_NAME" ]; then
-      echo "GITHUB_REF_NAME is empty in push event"
+      echo "GITHUB_REF_NAME is empty in push event" >&2
       exit 1
     fi
     echo "$GITHUB_REF_NAME"
@@ -54,7 +55,7 @@ function create_version() {
   elif [ "$GITHUB_EVENT_NAME" = schedule ]; then
     echo "$NEXT_RELEASE_VERSION-$NIGHTLY_RELEASE_PREFIX-$(date "+%Y%m%d")"
   else
-    echo "Unsupported GITHUB_EVENT_NAME: $GITHUB_EVENT_NAME"
+    echo "Unsupported GITHUB_EVENT_NAME: $GITHUB_EVENT_NAME" >&2
     exit 1
   fi
 }

.github/scripts/deploy-greptimedb.sh

@@ -10,7 +10,7 @@ GREPTIMEDB_IMAGE_TAG=${GREPTIMEDB_IMAGE_TAG:-latest}
 ETCD_CHART="oci://registry-1.docker.io/bitnamicharts/etcd"
 GREPTIME_CHART="https://greptimeteam.github.io/helm-charts/"
 
-# Ceate a cluster with 1 control-plane node and 5 workers.
+# Create a cluster with 1 control-plane node and 5 workers.
 function create_kind_cluster() {
   cat <<EOF | kind create cluster --name "${CLUSTER}" --image kindest/node:"$KUBERNETES_VERSION" --config=-
 kind: Cluster
@@ -68,7 +68,7 @@ function deploy_greptimedb_cluster() {
 
   helm install "$cluster_name" greptime/greptimedb-cluster \
     --set image.tag="$GREPTIMEDB_IMAGE_TAG" \
-    --set meta.etcdEndpoints="etcd.$install_namespace:2379" \
+    --set meta.backendStorage.etcd.endpoints="etcd.$install_namespace:2379" \
     -n "$install_namespace"
 
   # Wait for greptimedb cluster to be ready.
@@ -103,7 +103,7 @@ function deploy_greptimedb_cluster_with_s3_storage() {
 
   helm install "$cluster_name" greptime/greptimedb-cluster -n "$install_namespace" \
     --set image.tag="$GREPTIMEDB_IMAGE_TAG" \
-    --set meta.etcdEndpoints="etcd.$install_namespace:2379" \
+    --set meta.backendStorage.etcd.endpoints="etcd.$install_namespace:2379" \
     --set storage.s3.bucket="$AWS_CI_TEST_BUCKET" \
     --set storage.s3.region="$AWS_REGION" \
     --set storage.s3.root="$DATA_ROOT" \

.github/scripts/pull-test-deps-images.sh

@@ -0,0 +1,34 @@
+#!/bin/bash
+
+# This script is used to pull the test dependency images that are stored in public ECR one by one to avoid rate limiting.
+
+set -e
+
+MAX_RETRIES=3
+
+IMAGES=(
+  "greptime/zookeeper:3.7"
+  "greptime/kafka:3.9.0-debian-12-r1"
+  "greptime/etcd:3.6.1-debian-12-r3"
+  "greptime/minio:2024"
+  "greptime/mysql:5.7"
+)
+
+for image in "${IMAGES[@]}"; do
+  for ((attempt=1; attempt<=MAX_RETRIES; attempt++)); do
+    if docker pull "$image"; then
+      # Successfully pulled the image.
+      break
+    else
+      # Use some simple exponential backoff to avoid rate limiting.
+      if [ $attempt -lt $MAX_RETRIES ]; then
+        sleep_seconds=$((attempt * 5))
+        echo "Attempt $attempt failed for $image, waiting $sleep_seconds seconds"
+        sleep $sleep_seconds  # 5s, 10s delays
+      else
+        echo "Failed to pull $image after $MAX_RETRIES attempts"
+        exit 1
+      fi
+    fi
+  done
+done

.github/scripts/update-dev-builder-version.sh

@@ -0,0 +1,37 @@
+#!/bin/bash
+
+DEV_BUILDER_IMAGE_TAG=$1
+
+update_dev_builder_version() {
+  if [ -z "$DEV_BUILDER_IMAGE_TAG" ]; then
+    echo "Error: Should specify the dev-builder image tag"
+    exit 1
+  fi
+
+  # Configure Git configs.
+  git config --global user.email greptimedb-ci@greptime.com
+  git config --global user.name greptimedb-ci
+
+  # Checkout a new branch.
+  BRANCH_NAME="ci/update-dev-builder-$(date +%Y%m%d%H%M%S)"
+  git checkout -b $BRANCH_NAME
+
+  # Update the dev-builder image tag in the Makefile.
+  sed -i "s/DEV_BUILDER_IMAGE_TAG ?=.*/DEV_BUILDER_IMAGE_TAG ?= ${DEV_BUILDER_IMAGE_TAG}/g" Makefile
+
+  # Commit the changes.
+  git add Makefile
+  git commit -m "ci: update dev-builder image tag"
+  git push origin $BRANCH_NAME
+
+  # Create a Pull Request.
+  gh pr create \
+      --title "ci: update dev-builder image tag" \
+      --body "This PR updates the dev-builder image tag" \
+      --base main \
+      --head $BRANCH_NAME \
+      --reviewer zyy17 \
+      --reviewer daviderli614
+}
+
+update_dev_builder_version

.github/scripts/update-helm-charts-version.sh

@@ -0,0 +1,46 @@
+#!/bin/bash
+
+set -e
+
+VERSION=${VERSION}
+GITHUB_TOKEN=${GITHUB_TOKEN}
+
+update_helm_charts_version() {
+  # Configure Git configs.
+  git config --global user.email update-helm-charts-version@greptime.com
+  git config --global user.name update-helm-charts-version
+
+  # Clone helm-charts repository.
+  git clone "https://x-access-token:${GITHUB_TOKEN}@github.com/GreptimeTeam/helm-charts.git"
+  cd helm-charts
+
+  # Set default remote for gh CLI
+  gh repo set-default GreptimeTeam/helm-charts
+
+  # Checkout a new branch.
+  BRANCH_NAME="chore/greptimedb-${VERSION}"
+  git checkout -b $BRANCH_NAME
+
+  # Update version.
+  make update-version CHART=greptimedb-cluster VERSION=${VERSION}
+  make update-version CHART=greptimedb-standalone VERSION=${VERSION}
+
+  # Update docs.
+  make docs
+
+  # Commit the changes.
+  git add .
+  git commit -s -m "chore: Update GreptimeDB version to ${VERSION}"
+  git push origin $BRANCH_NAME
+
+  # Create a Pull Request.
+  gh pr create \
+    --title "chore: Update GreptimeDB version to ${VERSION}" \
+    --body "This PR updates the GreptimeDB version." \
+    --base main \
+    --head $BRANCH_NAME \
+    --reviewer zyy17 \
+    --reviewer daviderli614
+}
+
+update_helm_charts_version

.github/scripts/update-homebrew-greptme-version.sh

@@ -0,0 +1,42 @@
+#!/bin/bash
+
+set -e
+
+VERSION=${VERSION}
+GITHUB_TOKEN=${GITHUB_TOKEN}
+
+update_homebrew_greptime_version() {
+  # Configure Git configs.
+  git config --global user.email update-greptime-version@greptime.com
+  git config --global user.name update-greptime-version
+
+  # Clone helm-charts repository.
+  git clone "https://x-access-token:${GITHUB_TOKEN}@github.com/GreptimeTeam/homebrew-greptime.git"
+  cd homebrew-greptime
+
+  # Set default remote for gh CLI
+  gh repo set-default GreptimeTeam/homebrew-greptime
+
+  # Checkout a new branch.
+  BRANCH_NAME="chore/greptimedb-${VERSION}"
+  git checkout -b $BRANCH_NAME
+
+  # Update version.
+  make update-greptime-version VERSION=${VERSION}
+
+  # Commit the changes.
+  git add .
+  git commit -s -m "chore: Update GreptimeDB version to ${VERSION}"
+  git push origin $BRANCH_NAME
+
+  # Create a Pull Request.
+  gh pr create \
+    --title "chore: Update GreptimeDB version to ${VERSION}" \
+    --body "This PR updates the GreptimeDB version." \
+    --base main \
+    --head $BRANCH_NAME \
+    --reviewer zyy17 \
+    --reviewer daviderli614
+}
+
+update_homebrew_greptime_version

.github/scripts/upload-artifacts-to-s3.sh

@@ -41,7 +41,7 @@ function upload_artifacts() {
 # Updates the latest version information in AWS S3 if UPDATE_VERSION_INFO is true.
 function update_version_info() {
   if [ "$UPDATE_VERSION_INFO" == "true" ]; then
-    # If it's the officail release(like v1.0.0, v1.0.1, v1.0.2, etc.), update latest-version.txt.
+    # If it's the official release(like v1.0.0, v1.0.1, v1.0.2, etc.), update latest-version.txt.
     if [[ "$VERSION" =~ ^v[0-9]+\.[0-9]+\.[0-9]+$ ]]; then
       echo "Updating latest-version.txt"
       echo "$VERSION" > latest-version.txt

.github/workflows/dev-build.yml

@@ -55,6 +55,11 @@ on:
         description: Build and push images to DockerHub and ACR
         required: false
         default: true
+      upload_artifacts_to_s3:
+        type: boolean
+        description: Whether upload artifacts to s3
+        required: false
+        default: false
       cargo_profile:
         type: choice
         description: The cargo profile to use in building GreptimeDB.
@@ -238,7 +243,7 @@ jobs:
           version: ${{ needs.allocate-runners.outputs.version }}
           push-latest-tag: false # Don't push the latest tag to registry.
           dev-mode: true # Only build the standard images.
-          
+
       - name: Echo Docker image tag to step summary
         run: |
           echo "## Docker Image Tag" >> $GITHUB_STEP_SUMMARY
@@ -281,7 +286,7 @@ jobs:
           aws-cn-access-key-id: ${{ secrets.AWS_CN_ACCESS_KEY_ID }}
           aws-cn-secret-access-key: ${{ secrets.AWS_CN_SECRET_ACCESS_KEY }}
           aws-cn-region: ${{ vars.AWS_RELEASE_BUCKET_REGION }}
-          upload-to-s3: false
+          upload-to-s3: ${{ inputs.upload_artifacts_to_s3 }}
           dev-mode: true                     # Only build the standard images(exclude centos images).
           push-latest-tag: false             # Don't push the latest tag to registry.
           update-version-info: false         # Don't update the version info in S3.

.github/workflows/develop.yml

@@ -22,6 +22,7 @@ concurrency:
 
 jobs:
   check-typos-and-docs:
+    if: ${{ github.repository == 'GreptimeTeam/greptimedb' }}
     name: Check typos and docs
     runs-on: ubuntu-latest
     steps:
@@ -36,6 +37,7 @@ jobs:
           || (echo "'config/config.md' is not up-to-date, please run 'make config-docs'." && exit 1)
 
   license-header-check:
+    if: ${{ github.repository == 'GreptimeTeam/greptimedb' }}
     runs-on: ubuntu-latest
     name: Check License Header
     steps:
@@ -45,6 +47,7 @@ jobs:
       - uses: korandoru/hawkeye@v5
 
   check:
+    if: ${{ github.repository == 'GreptimeTeam/greptimedb' }}
     name: Check
     runs-on: ${{ matrix.os }}
     strategy:
@@ -71,6 +74,7 @@ jobs:
         run: cargo check --locked --workspace --all-targets
 
   toml:
+    if: ${{ github.repository == 'GreptimeTeam/greptimedb' }}
     name: Toml Check
     runs-on: ubuntu-latest
     timeout-minutes: 60
@@ -85,6 +89,7 @@ jobs:
         run: taplo format --check
 
   build:
+    if: ${{ github.repository == 'GreptimeTeam/greptimedb' }}
     name: Build GreptimeDB binaries
     runs-on: ${{ matrix.os }}
     strategy:
@@ -127,6 +132,7 @@ jobs:
           version: current
 
   fuzztest:
+    if: ${{ github.repository == 'GreptimeTeam/greptimedb' }}
     name: Fuzz Test
     needs: build
     runs-on: ubuntu-latest
@@ -183,11 +189,13 @@ jobs:
           max-total-time: 120
 
   unstable-fuzztest:
+    if: ${{ github.repository == 'GreptimeTeam/greptimedb' }}
     name: Unstable Fuzz Test
     needs: build-greptime-ci
     runs-on: ubuntu-latest
     timeout-minutes: 60
     strategy:
+      fail-fast: false
       matrix:
         target: [ "unstable_fuzz_create_table_standalone" ]
     steps:
@@ -215,12 +223,12 @@ jobs:
         run: |
           sudo apt update && sudo apt install -y libfuzzer-14-dev
           cargo install cargo-fuzz cargo-gc-bin --force
-      - name: Download pre-built binariy
+      - name: Download pre-built binary
         uses: actions/download-artifact@v4
         with:
           name: bin
           path: .
-      - name: Unzip bianry
+      - name: Unzip binary
         run: |
           tar -xvf ./bin.tar.gz
           rm ./bin.tar.gz
@@ -242,8 +250,14 @@ jobs:
           name: unstable-fuzz-logs
           path: /tmp/unstable-greptime/
           retention-days: 3
+      - name: Describe pods
+        if: failure()
+        shell: bash
+        run: |
+          kubectl describe pod -n my-greptimedb
 
   build-greptime-ci:
+    if: ${{ github.repository == 'GreptimeTeam/greptimedb' }}
     name: Build GreptimeDB binary (profile-CI)
     runs-on: ${{ matrix.os }}
     strategy:
@@ -267,7 +281,7 @@ jobs:
       - name: Install cargo-gc-bin
         shell: bash
         run: cargo install cargo-gc-bin --force
-      - name: Build greptime bianry
+      - name: Build greptime binary
         shell: bash
         # `cargo gc` will invoke `cargo build` with specified args
         run: cargo gc --profile ci -- --bin greptime --features "pg_kvbackend,mysql_kvbackend"
@@ -285,11 +299,13 @@ jobs:
           version: current
 
   distributed-fuzztest:
+    if: ${{ github.repository == 'GreptimeTeam/greptimedb' }}
     name: Fuzz Test (Distributed, ${{ matrix.mode.name }}, ${{ matrix.target }})
     runs-on: ubuntu-latest
     needs:  build-greptime-ci
     timeout-minutes: 60
     strategy:
+      fail-fast: false
       matrix:
         target: [ "fuzz_create_table", "fuzz_alter_table", "fuzz_create_database", "fuzz_create_logical_table", "fuzz_alter_logical_table", "fuzz_insert", "fuzz_insert_logical_table" ]
         mode:
@@ -319,9 +335,9 @@ jobs:
         name: Setup Minio
         uses: ./.github/actions/setup-minio
       - if: matrix.mode.kafka
-        name: Setup Kafka cluser
+        name: Setup Kafka cluster
         uses: ./.github/actions/setup-kafka-cluster
-      - name: Setup Etcd cluser
+      - name: Setup Etcd cluster
         uses: ./.github/actions/setup-etcd-cluster
       # Prepares for fuzz tests
       - uses: arduino/setup-protoc@v3
@@ -394,6 +410,11 @@ jobs:
         shell: bash
         run: |
           kubectl describe nodes
+      - name: Describe pod
+        if: failure()
+        shell: bash
+        run: |
+          kubectl describe pod -n my-greptimedb
       - name: Export kind logs
         if: failure()
         shell: bash
@@ -416,11 +437,13 @@ jobs:
           docker system prune -f
 
   distributed-fuzztest-with-chaos:
+    if: ${{ github.repository == 'GreptimeTeam/greptimedb' }}
     name: Fuzz Test with Chaos (Distributed, ${{ matrix.mode.name }}, ${{ matrix.target }})
     runs-on: ubuntu-latest
     needs:  build-greptime-ci
     timeout-minutes: 60
     strategy:
+      fail-fast: false
       matrix:
         target: ["fuzz_migrate_mito_regions", "fuzz_migrate_metric_regions", "fuzz_failover_mito_regions", "fuzz_failover_metric_regions"]
         mode:
@@ -465,9 +488,9 @@ jobs:
         name: Setup Minio
         uses: ./.github/actions/setup-minio
       - if: matrix.mode.kafka
-        name: Setup Kafka cluser
+        name: Setup Kafka cluster
         uses: ./.github/actions/setup-kafka-cluster
-      - name: Setup Etcd cluser
+      - name: Setup Etcd cluster
         uses: ./.github/actions/setup-etcd-cluster
       # Prepares for fuzz tests
       - uses: arduino/setup-protoc@v3
@@ -541,6 +564,11 @@ jobs:
         shell: bash
         run: |
           kubectl describe nodes
+      - name: Describe pods
+        if: failure()
+        shell: bash
+        run: |
+          kubectl describe pod -n my-greptimedb
       - name: Export kind logs
         if: failure()
         shell: bash
@@ -563,10 +591,12 @@ jobs:
           docker system prune -f
 
   sqlness:
+    if: ${{ github.repository == 'GreptimeTeam/greptimedb' }}
     name: Sqlness Test (${{ matrix.mode.name }})
     needs: build
     runs-on: ${{ matrix.os }}
     strategy:
+      fail-fast: false
       matrix:
         os: [ ubuntu-latest ]
         mode:
@@ -609,6 +639,7 @@ jobs:
           retention-days: 3
 
   fmt:
+    if: ${{ github.repository == 'GreptimeTeam/greptimedb' }}
     name: Rustfmt
     runs-on: ubuntu-latest
     timeout-minutes: 60
@@ -626,6 +657,7 @@ jobs:
         run: make fmt-check
 
   clippy:
+    if: ${{ github.repository == 'GreptimeTeam/greptimedb' }}
     name: Clippy
     runs-on: ubuntu-latest
     timeout-minutes: 60
@@ -651,6 +683,7 @@ jobs:
         run: make clippy
 
   conflict-check:
+    if: ${{ github.repository == 'GreptimeTeam/greptimedb' }}
     name: Check for conflict
     runs-on: ubuntu-latest
     steps:
@@ -661,7 +694,7 @@ jobs:
         uses: olivernybroe/action-conflict-finder@v4.0
 
   test:
-    if: github.event_name != 'merge_group'
+    if: ${{ github.repository == 'GreptimeTeam/greptimedb' && github.event_name != 'merge_group' }}
     runs-on: ubuntu-22.04-arm
     timeout-minutes: 60
     needs:  [conflict-check, clippy, fmt]
@@ -686,6 +719,10 @@ jobs:
           save-if: ${{ github.ref == 'refs/heads/main' }}
       - name: Install latest nextest release
         uses: taiki-e/install-action@nextest
+
+      - name: Pull test dependencies images
+        run: ./.github/scripts/pull-test-deps-images.sh
+
       - name: Setup external services
         working-directory: tests-integration/fixtures
         run: docker compose up -d --wait
@@ -713,7 +750,7 @@ jobs:
           UNITTEST_LOG_DIR: "__unittest_logs"
 
   coverage:
-    if: github.event_name == 'merge_group'
+    if: ${{ github.repository == 'GreptimeTeam/greptimedb' && github.event_name == 'merge_group' }}
     runs-on: ubuntu-22.04-8-cores
     timeout-minutes: 60
     steps:
@@ -773,6 +810,7 @@ jobs:
           verbose: true
 
   # compat:
+  #   if: ${{ github.repository == 'GreptimeTeam/greptimedb' }}
   #   name: Compatibility Test
   #   needs: build
   #   runs-on: ubuntu-22.04

.github/workflows/grafana.yml

@@ -21,32 +21,6 @@ jobs:
         run: sudo apt-get install -y jq
 
       # Make the check.sh script executable
-      - name: Make check.sh executable
-        run: chmod +x grafana/check.sh
-
-      # Run the check.sh script
-      - name: Run check.sh
-        run: ./grafana/check.sh
-
-      # Only run summary.sh for pull_request events (not for merge queues or final pushes)
-      - name: Check if this is a pull request
-        id: check-pr
+      - name: Check grafana dashboards
         run: |
-          if [[ "${{ github.event_name }}" == "pull_request" ]]; then
-            echo "is_pull_request=true" >> $GITHUB_OUTPUT
-          else
-            echo "is_pull_request=false" >> $GITHUB_OUTPUT
-          fi
-
-      # Make the summary.sh script executable
-      - name: Make summary.sh executable
-        if: steps.check-pr.outputs.is_pull_request == 'true'
-        run: chmod +x grafana/summary.sh
-
-      # Run the summary.sh script and add its output to the GitHub Job Summary
-      - name: Run summary.sh and add to Job Summary
-        if: steps.check-pr.outputs.is_pull_request == 'true'
-        run: |
-          SUMMARY=$(./grafana/summary.sh)
-          echo "### Summary of Grafana Panels" >> $GITHUB_STEP_SUMMARY
-          echo "$SUMMARY" >> $GITHUB_STEP_SUMMARY
+          make check-dashboards

.github/workflows/nightly-ci.yml

@@ -117,16 +117,16 @@ jobs:
     name: Run clean build on Linux
     runs-on: ubuntu-latest
     if: ${{ github.repository == 'GreptimeTeam/greptimedb' }}
-    timeout-minutes: 60
+    timeout-minutes: 45
     steps:
       - uses: actions/checkout@v4
         with:
           fetch-depth: 0
           persist-credentials: false
-      - uses: cachix/install-nix-action@v27
-        with:
-          nix_path: nixpkgs=channel:nixos-24.11
-      - run: nix develop --command cargo build
+      - uses: cachix/install-nix-action@v31
+      - run: nix develop --command cargo check --bin greptime
+        env:
+          CARGO_BUILD_RUSTFLAGS: "-C link-arg=-fuse-ld=mold"
 
   check-status:
     name: Check status

.github/workflows/pr-labeling.yaml

@@ -0,0 +1,42 @@
+name: 'PR Labeling'
+
+on:
+  pull_request_target:
+    types:
+      - opened
+      - synchronize
+      - reopened
+
+permissions:
+  contents: read
+  pull-requests: write
+  issues: write
+
+jobs:
+  labeler:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout sources
+        uses: actions/checkout@v4
+
+      - uses: actions/labeler@v5
+        with:
+          configuration-path: ".github/labeler.yaml"
+          repo-token: "${{ secrets.GITHUB_TOKEN }}"
+
+  size-label:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: pascalgn/size-label-action@v0.5.5
+        env:
+          GITHUB_TOKEN: "${{ secrets.GITHUB_TOKEN }}"
+        with:
+          sizes: >
+            {
+              "0": "XS",
+              "100": "S",
+              "300": "M",
+              "1000": "L",
+              "1500": "XL",
+              "2000": "XXL"
+            }

.github/workflows/release-dev-builder-images.yaml

@@ -24,11 +24,19 @@ on:
         description: Release dev-builder-android image
         required: false
         default: false
+      update_dev_builder_image_tag:
+        type: boolean
+        description: Update the DEV_BUILDER_IMAGE_TAG in Makefile and create a PR
+        required: false
+        default: false
 
 jobs:
   release-dev-builder-images:
     name: Release dev builder images
-    if: ${{ inputs.release_dev_builder_ubuntu_image || inputs.release_dev_builder_centos_image || inputs.release_dev_builder_android_image }} # Only manually trigger this job.
+    # The jobs are triggered by the following events:
+    # 1. Manually triggered workflow_dispatch event
+    # 2. Push event when the PR that modifies the `rust-toolchain.toml` or `docker/dev-builder/**` is merged to main
+    if: ${{ github.event_name == 'push' || inputs.release_dev_builder_ubuntu_image || inputs.release_dev_builder_centos_image || inputs.release_dev_builder_android_image }}
     runs-on: ubuntu-latest
     outputs:
       version: ${{ steps.set-version.outputs.version }}
@@ -57,9 +65,9 @@ jobs:
           version: ${{ env.VERSION }}
           dockerhub-image-registry-username: ${{ secrets.DOCKERHUB_USERNAME }}
           dockerhub-image-registry-token: ${{ secrets.DOCKERHUB_TOKEN }}
-          build-dev-builder-ubuntu: ${{ inputs.release_dev_builder_ubuntu_image }}
-          build-dev-builder-centos: ${{ inputs.release_dev_builder_centos_image }}
-          build-dev-builder-android: ${{ inputs.release_dev_builder_android_image }}
+          build-dev-builder-ubuntu: ${{ inputs.release_dev_builder_ubuntu_image || github.event_name == 'push' }}
+          build-dev-builder-centos: ${{ inputs.release_dev_builder_centos_image || github.event_name == 'push' }}
+          build-dev-builder-android: ${{ inputs.release_dev_builder_android_image || github.event_name == 'push' }}
 
   release-dev-builder-images-ecr:
     name: Release dev builder images to AWS ECR
@@ -85,7 +93,7 @@ jobs:
 
       - name: Push dev-builder-ubuntu image
         shell: bash
-        if: ${{ inputs.release_dev_builder_ubuntu_image }}
+        if: ${{ inputs.release_dev_builder_ubuntu_image || github.event_name == 'push' }}
         env:
           IMAGE_VERSION: ${{ needs.release-dev-builder-images.outputs.version }}
           IMAGE_NAMESPACE: ${{ vars.IMAGE_NAMESPACE }}
@@ -106,7 +114,7 @@ jobs:
 
       - name: Push dev-builder-centos image
         shell: bash
-        if: ${{ inputs.release_dev_builder_centos_image }}
+        if: ${{ inputs.release_dev_builder_centos_image || github.event_name == 'push' }}
         env:
           IMAGE_VERSION: ${{ needs.release-dev-builder-images.outputs.version }}
           IMAGE_NAMESPACE: ${{ vars.IMAGE_NAMESPACE }}
@@ -127,7 +135,7 @@ jobs:
 
       - name: Push dev-builder-android image
         shell: bash
-        if: ${{ inputs.release_dev_builder_android_image }}
+        if: ${{ inputs.release_dev_builder_android_image || github.event_name == 'push' }}
         env:
           IMAGE_VERSION: ${{ needs.release-dev-builder-images.outputs.version }}
           IMAGE_NAMESPACE: ${{ vars.IMAGE_NAMESPACE }}
@@ -162,7 +170,7 @@ jobs:
 
       - name: Push dev-builder-ubuntu image
         shell: bash
-        if: ${{ inputs.release_dev_builder_ubuntu_image }}
+        if: ${{ inputs.release_dev_builder_ubuntu_image || github.event_name == 'push' }}
         env:
           IMAGE_VERSION: ${{ needs.release-dev-builder-images.outputs.version }}
           IMAGE_NAMESPACE: ${{ vars.IMAGE_NAMESPACE }}
@@ -176,7 +184,7 @@ jobs:
 
       - name: Push dev-builder-centos image
         shell: bash
-        if: ${{ inputs.release_dev_builder_centos_image }}
+        if: ${{ inputs.release_dev_builder_centos_image || github.event_name == 'push' }}
         env:
           IMAGE_VERSION: ${{ needs.release-dev-builder-images.outputs.version }}
           IMAGE_NAMESPACE: ${{ vars.IMAGE_NAMESPACE }}
@@ -190,7 +198,7 @@ jobs:
 
       - name: Push dev-builder-android image
         shell: bash
-        if: ${{ inputs.release_dev_builder_android_image }}
+        if: ${{ inputs.release_dev_builder_android_image || github.event_name == 'push' }}
         env:
           IMAGE_VERSION: ${{ needs.release-dev-builder-images.outputs.version }}
           IMAGE_NAMESPACE: ${{ vars.IMAGE_NAMESPACE }}
@@ -201,3 +209,24 @@ jobs:
             quay.io/skopeo/stable:latest \
             copy -a docker://docker.io/$IMAGE_NAMESPACE/dev-builder-android:$IMAGE_VERSION \
             docker://$ACR_IMAGE_REGISTRY/$IMAGE_NAMESPACE/dev-builder-android:$IMAGE_VERSION
+  
+  update-dev-builder-image-tag:
+    name: Update dev-builder image tag
+    runs-on: ubuntu-latest
+    permissions:
+      contents: write
+      pull-requests: write
+    if: ${{ github.event_name == 'push' || inputs.update_dev_builder_image_tag }}
+    needs: [
+      release-dev-builder-images
+    ]
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+
+      - name: Update dev-builder image tag
+        shell: bash
+        env:
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+        run: |
+          ./.github/scripts/update-dev-builder-version.sh ${{ needs.release-dev-builder-images.outputs.version }}

.github/workflows/release.yml

@@ -88,10 +88,8 @@ env:
   # Controls whether to run tests, include unit-test, integration-test and sqlness.
   DISABLE_RUN_TESTS: ${{ inputs.skip_test || vars.DEFAULT_SKIP_TEST }}
 
-  # The scheduled version is '${{ env.NEXT_RELEASE_VERSION }}-nightly-YYYYMMDD', like v0.2.0-nigthly-20230313;
+  # The scheduled version is '${{ env.NEXT_RELEASE_VERSION }}-nightly-YYYYMMDD', like v0.2.0-nightly-20230313;
   NIGHTLY_RELEASE_PREFIX: nightly
-  # Note: The NEXT_RELEASE_VERSION should be modified manually by every formal release.
-  NEXT_RELEASE_VERSION: v0.14.0
 
 jobs:
   allocate-runners:
@@ -112,6 +110,8 @@ jobs:
 
       # The 'version' use as the global tag name of the release workflow.
       version: ${{ steps.create-version.outputs.version }}
+
+      should-push-latest-tag: ${{ steps.check-version.outputs.should-push-latest-tag }}
     steps:
       - name: Checkout
         uses: actions/checkout@v4
@@ -126,7 +126,7 @@ jobs:
 
       # The create-version will create a global variable named 'version' in the global workflows.
       # - If it's a tag push release, the version is the tag name(${{ github.ref_name }});
-      # - If it's a scheduled release, the version is '${{ env.NEXT_RELEASE_VERSION }}-nightly-$buildTime', like v0.2.0-nigthly-20230313;
+      # - If it's a scheduled release, the version is '${{ env.NEXT_RELEASE_VERSION }}-nightly-$buildTime', like v0.2.0-nightly-20230313;
       # - If it's a manual release, the version is '${{ env.NEXT_RELEASE_VERSION }}-<short-git-sha>-YYYYMMDDSS', like v0.2.0-e5b243c-2023071245;
       - name: Create version
         id: create-version
@@ -135,9 +135,13 @@ jobs:
         env:
           GITHUB_EVENT_NAME: ${{ github.event_name }}
           GITHUB_REF_NAME: ${{ github.ref_name }}
-          NEXT_RELEASE_VERSION: ${{ env.NEXT_RELEASE_VERSION }}
           NIGHTLY_RELEASE_PREFIX: ${{ env.NIGHTLY_RELEASE_PREFIX }}
 
+      - name: Check version
+        id: check-version
+        run: |
+          ./.github/scripts/check-version.sh "${{ steps.create-version.outputs.version }}"
+
       - name: Allocate linux-amd64 runner
         if: ${{ inputs.build_linux_amd64_artifacts || github.event_name == 'push' || github.event_name == 'schedule' }}
         uses: ./.github/actions/start-runner
@@ -317,7 +321,7 @@ jobs:
           image-registry-username: ${{ secrets.DOCKERHUB_USERNAME }}
           image-registry-password: ${{ secrets.DOCKERHUB_TOKEN }}
           version: ${{ needs.allocate-runners.outputs.version }}
-          push-latest-tag: ${{ github.ref_type == 'tag' && !contains(github.ref_name, 'nightly') && github.event_name != 'schedule' }}
+          push-latest-tag: ${{ needs.allocate-runners.outputs.should-push-latest-tag == 'true' && github.ref_type == 'tag' && !contains(github.ref_name, 'nightly') && github.event_name != 'schedule' }}
 
       - name: Set build image result
         id: set-build-image-result
@@ -364,7 +368,7 @@ jobs:
           dev-mode: false
           upload-to-s3: true
           update-version-info: true
-          push-latest-tag: ${{ github.ref_type == 'tag' && !contains(github.ref_name, 'nightly') && github.event_name != 'schedule' }}
+          push-latest-tag: ${{ needs.allocate-runners.outputs.should-push-latest-tag == 'true' && github.ref_type == 'tag' && !contains(github.ref_name, 'nightly') && github.event_name != 'schedule' }}
 
   publish-github-release:
     name: Create GitHub release and upload artifacts
@@ -391,7 +395,7 @@ jobs:
 
   ### Stop runners ###
   # It's very necessary to split the job of releasing runners into 'stop-linux-amd64-runner' and 'stop-linux-arm64-runner'.
-  # Because we can terminate the specified EC2 instance immediately after the job is finished without uncessary waiting.
+  # Because we can terminate the specified EC2 instance immediately after the job is finished without unnecessary waiting.
   stop-linux-amd64-runner: # It's always run as the last job in the workflow to make sure that the runner is released.
     name: Stop linux-amd64 runner
     # Only run this job when the runner is allocated.
@@ -444,10 +448,10 @@ jobs:
           aws-region: ${{ vars.EC2_RUNNER_REGION }}
           github-token: ${{ secrets.GH_PERSONAL_ACCESS_TOKEN }}
 
-  bump-doc-version:
-    name: Bump doc version
+  bump-downstream-repo-versions:
+    name: Bump downstream repo versions
     if: ${{ github.event_name == 'push' || github.event_name == 'schedule' }}
-    needs: [allocate-runners]
+    needs: [allocate-runners, publish-github-release]
     runs-on: ubuntu-latest
     # Permission reference: https://docs.github.com/en/actions/using-jobs/assigning-permissions-to-jobs
     permissions:
@@ -459,13 +463,58 @@ jobs:
           fetch-depth: 0
           persist-credentials: false
       - uses: ./.github/actions/setup-cyborg
-      - name: Bump doc version
+      - name: Bump downstream repo versions
         working-directory: cyborg
-        run: pnpm tsx bin/bump-doc-version.ts
+        run: pnpm tsx bin/bump-versions.ts
         env:
+          TARGET_REPOS: website,docs,demo
           VERSION: ${{ needs.allocate-runners.outputs.version }}
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          WEBSITE_REPO_TOKEN: ${{ secrets.WEBSITE_REPO_TOKEN }}
           DOCS_REPO_TOKEN: ${{ secrets.DOCS_REPO_TOKEN }}
+          DEMO_REPO_TOKEN: ${{ secrets.DEMO_REPO_TOKEN }}
+
+  bump-helm-charts-version:
+    name: Bump helm charts version
+    if: ${{ github.ref_type == 'tag' && !contains(github.ref_name, 'nightly') && github.event_name != 'schedule' }}
+    needs: [allocate-runners, publish-github-release]
+    runs-on: ubuntu-latest
+    permissions:
+      contents: write
+      pull-requests: write
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Bump helm charts version
+        env:
+          GITHUB_TOKEN: ${{ secrets.HELM_CHARTS_REPO_TOKEN }}
+          VERSION: ${{ needs.allocate-runners.outputs.version }}
+        run: |
+          ./.github/scripts/update-helm-charts-version.sh
+
+  bump-homebrew-greptime-version:
+    name: Bump homebrew greptime version
+    if: ${{ github.ref_type == 'tag' && !contains(github.ref_name, 'nightly') && github.event_name != 'schedule' }}
+    needs: [allocate-runners, publish-github-release]
+    runs-on: ubuntu-latest
+    permissions:
+      contents: write
+      pull-requests: write
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Bump homebrew greptime version
+        env:
+          GITHUB_TOKEN: ${{ secrets.HOMEBREW_GREPTIME_REPO_TOKEN }}
+          VERSION: ${{ needs.allocate-runners.outputs.version }}
+        run: |
+          ./.github/scripts/update-homebrew-greptme-version.sh
 
   notification:
     if: ${{ github.repository == 'GreptimeTeam/greptimedb' && (github.event_name == 'push' || github.event_name == 'schedule') && always() }}

.github/workflows/semantic-pull-request.yml

@@ -14,6 +14,9 @@ concurrency:
 jobs:
   check:
     runs-on: ubuntu-latest
+    permissions:
+      pull-requests: write  # Add permissions to modify PRs
+      issues: write
     timeout-minutes: 10
     steps:
       - uses: actions/checkout@v4

.gitignore

@@ -28,6 +28,7 @@ debug/
 # Logs
 **/__unittest_logs
 logs/
+!grafana/dashboards/logs/
 
 # cpython's generated python byte code
 **/__pycache__/
@@ -57,3 +58,6 @@ tests-fuzz/corpus/
 
 ## default data home
 greptimedb_data
+
+# github
+!/.github

CONTRIBUTING.md

@@ -108,7 +108,7 @@ of what you were trying to do and what went wrong. You can also reach for help i
 The core team will be thrilled if you would like to participate in any way you like. When you are stuck, try to ask for help by filing an issue, with a detailed description of what you were trying to do and what went wrong. If you have any questions or if you would like to get involved in our community, please check out:
 
 - [GreptimeDB Community Slack](https://greptime.com/slack)
-- [GreptimeDB Github Discussions](https://github.com/GreptimeTeam/greptimedb/discussions)
+- [GreptimeDB GitHub Discussions](https://github.com/GreptimeTeam/greptimedb/discussions)
 
 Also, see some extra GreptimeDB content:
 

Cargo.toml

@@ -30,12 +30,14 @@ members = [
     "src/common/recordbatch",
     "src/common/runtime",
     "src/common/session",
+    "src/common/stat",
     "src/common/substrait",
     "src/common/telemetry",
     "src/common/test-util",
     "src/common/time",
     "src/common/version",
     "src/common/wal",
+    "src/common/workload",
     "src/datanode",
     "src/datatypes",
     "src/file-engine",
@@ -47,6 +49,7 @@ members = [
     "src/meta-client",
     "src/meta-srv",
     "src/metric-engine",
+    "src/mito-codec",
     "src/mito2",
     "src/object-store",
     "src/operator",
@@ -68,15 +71,17 @@ members = [
 resolver = "2"
 
 [workspace.package]
-version = "0.14.0"
+version = "0.15.5"
 edition = "2021"
 license = "Apache-2.0"
 
 [workspace.lints]
-clippy.print_stdout = "warn"
-clippy.print_stderr = "warn"
 clippy.dbg_macro = "warn"
 clippy.implicit_clone = "warn"
+clippy.result_large_err = "allow"
+clippy.large_enum_variant = "allow"
+clippy.doc_overindented_list_items = "allow"
+clippy.uninlined_format_args = "allow"
 rust.unknown_lints = "deny"
 rust.unexpected_cfgs = { level = "warn", check-cfg = ['cfg(tokio_unstable)'] }
 
@@ -112,24 +117,25 @@ clap = { version = "4.4", features = ["derive"] }
 config = "0.13.0"
 crossbeam-utils = "0.8"
 dashmap = "6.1"
-datafusion = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "5bbedc6704162afb03478f56ffb629405a4e1220" }
-datafusion-common = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "5bbedc6704162afb03478f56ffb629405a4e1220" }
-datafusion-expr = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "5bbedc6704162afb03478f56ffb629405a4e1220" }
-datafusion-functions = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "5bbedc6704162afb03478f56ffb629405a4e1220" }
-datafusion-optimizer = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "5bbedc6704162afb03478f56ffb629405a4e1220" }
-datafusion-physical-expr = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "5bbedc6704162afb03478f56ffb629405a4e1220" }
-datafusion-physical-plan = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "5bbedc6704162afb03478f56ffb629405a4e1220" }
-datafusion-sql = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "5bbedc6704162afb03478f56ffb629405a4e1220" }
-datafusion-substrait = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "5bbedc6704162afb03478f56ffb629405a4e1220" }
+datafusion = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "12c0381babd52c681043957e9d6ee083a03f7646" }
+datafusion-common = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "12c0381babd52c681043957e9d6ee083a03f7646" }
+datafusion-expr = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "12c0381babd52c681043957e9d6ee083a03f7646" }
+datafusion-functions = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "12c0381babd52c681043957e9d6ee083a03f7646" }
+datafusion-optimizer = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "12c0381babd52c681043957e9d6ee083a03f7646" }
+datafusion-physical-expr = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "12c0381babd52c681043957e9d6ee083a03f7646" }
+datafusion-physical-plan = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "12c0381babd52c681043957e9d6ee083a03f7646" }
+datafusion-sql = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "12c0381babd52c681043957e9d6ee083a03f7646" }
+datafusion-substrait = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "12c0381babd52c681043957e9d6ee083a03f7646" }
 deadpool = "0.12"
 deadpool-postgres = "0.14"
 derive_builder = "0.20"
 dotenv = "0.15"
+either = "1.15"
 etcd-client = "0.14"
 fst = "0.4.7"
 futures = "0.3"
 futures-util = "0.3"
-greptime-proto = { git = "https://github.com/GreptimeTeam/greptime-proto.git", rev = "b6d9cffd43c4e6358805a798f17e03e232994b82" }
+greptime-proto = { git = "https://github.com/GreptimeTeam/greptime-proto.git", rev = "f3103a8c9b8ce162457d0a3e3ca00d53d1a8bd06" }
 hex = "0.4"
 http = "1"
 humantime = "2.1"
@@ -145,6 +151,7 @@ meter-core = { git = "https://github.com/GreptimeTeam/greptime-meter.git", rev =
 mockall = "0.13"
 moka = "0.12"
 nalgebra = "0.33"
+nix = { version = "0.30.1", default-features = false, features = ["event", "fs", "process"] }
 notify = "8.0"
 num_cpus = "1.16"
 object_store_opendal = "0.50"
@@ -161,8 +168,10 @@ parquet = { version = "54.2", default-features = false, features = ["arrow", "as
 paste = "1.0"
 pin-project = "1.0"
 prometheus = { version = "0.13.3", features = ["process"] }
-promql-parser = { version = "0.5.1", features = ["ser"] }
-prost = "0.13"
+promql-parser = { git = "https://github.com/GreptimeTeam/promql-parser.git", rev = "0410e8b459dda7cb222ce9596f8bf3971bd07bd2", features = [
+    "ser",
+] }
+prost = { version = "0.13", features = ["no-recursion-limit"] }
 raft-engine = { version = "0.4.1", default-features = false }
 rand = "0.9"
 ratelimit = "0.10"
@@ -174,7 +183,7 @@ reqwest = { version = "0.12", default-features = false, features = [
     "stream",
     "multipart",
 ] }
-rskafka = { git = "https://github.com/influxdata/rskafka.git", rev = "75535b5ad9bae4a5dbb582c82e44dfd81ec10105", features = [
+rskafka = { git = "https://github.com/influxdata/rskafka.git", rev = "8dbd01ed809f5a791833a594e85b144e36e45820", features = [
     "transport-tls",
 ] }
 rstest = "0.25"
@@ -191,7 +200,7 @@ simd-json = "0.15"
 similar-asserts = "1.6.0"
 smallvec = { version = "1", features = ["serde"] }
 snafu = "0.8"
-sqlparser = { git = "https://github.com/GreptimeTeam/sqlparser-rs.git", rev = "e98e6b322426a9d397a71efef17075966223c089", features = [
+sqlparser = { git = "https://github.com/GreptimeTeam/sqlparser-rs.git", rev = "0cf6c04490d59435ee965edd2078e8855bd8471e", features = [
     "visitor",
     "serde",
 ] } # branch = "v0.54.x"
@@ -212,6 +221,8 @@ tokio-util = { version = "0.7", features = ["io-util", "compat"] }
 toml = "0.8.8"
 tonic = { version = "0.12", features = ["tls", "gzip", "zstd"] }
 tower = "0.5"
+tower-http = "0.6"
+tracing = "0.1"
 tracing-appender = "0.2"
 tracing-subscriber = { version = "0.3", features = ["env-filter", "json", "fmt"] }
 typetag = "0.2"
@@ -255,6 +266,7 @@ common-test-util = { path = "src/common/test-util" }
 common-time = { path = "src/common/time" }
 common-version = { path = "src/common/version" }
 common-wal = { path = "src/common/wal" }
+common-workload = { path = "src/common/workload" }
 datanode = { path = "src/datanode" }
 datatypes = { path = "src/datatypes" }
 file-engine = { path = "src/file-engine" }
@@ -266,9 +278,13 @@ log-store = { path = "src/log-store" }
 meta-client = { path = "src/meta-client" }
 meta-srv = { path = "src/meta-srv" }
 metric-engine = { path = "src/metric-engine" }
+mito-codec = { path = "src/mito-codec" }
 mito2 = { path = "src/mito2" }
 object-store = { path = "src/object-store" }
 operator = { path = "src/operator" }
+otel-arrow-rust = { git = "https://github.com/open-telemetry/otel-arrow", rev = "5d551412d2a12e689cde4d84c14ef29e36784e51", features = [
+    "server",
+] }
 partition = { path = "src/partition" }
 pipeline = { path = "src/pipeline" }
 plugins = { path = "src/plugins" }
@@ -278,6 +294,7 @@ query = { path = "src/query" }
 servers = { path = "src/servers" }
 session = { path = "src/session" }
 sql = { path = "src/sql" }
+stat = { path = "src/common/stat" }
 store-api = { path = "src/store-api" }
 substrait = { path = "src/common/substrait" }
 table = { path = "src/table" }

Makefile

@@ -8,7 +8,7 @@ CARGO_BUILD_OPTS := --locked
 IMAGE_REGISTRY ?= docker.io
 IMAGE_NAMESPACE ?= greptime
 IMAGE_TAG ?= latest
-DEV_BUILDER_IMAGE_TAG ?= 2024-12-25-a71b93dd-20250305072908
+DEV_BUILDER_IMAGE_TAG ?= 2025-05-19-f55023f3-20250829091211
 BUILDX_MULTI_PLATFORM_BUILD ?= false
 BUILDX_BUILDER_NAME ?= gtbuilder
 BASE_IMAGE ?= ubuntu
@@ -222,6 +222,16 @@ start-cluster: ## Start the greptimedb cluster with etcd by using docker compose
 stop-cluster: ## Stop the greptimedb cluster that created by docker compose.
 	docker compose -f ./docker/docker-compose/cluster-with-etcd.yaml stop
 
+##@ Grafana
+
+.PHONY: check-dashboards
+check-dashboards: ## Check the Grafana dashboards.
+	@./grafana/scripts/check.sh
+
+.PHONY: dashboards
+dashboards: ## Generate the Grafana dashboards for standalone mode and intermediate dashboards.
+	@./grafana/scripts/gen-dashboards.sh
+
 ##@ Docs
 config-docs: ## Generate configuration documentation from toml files.
 	docker run --rm \

README.md

@@ -6,7 +6,9 @@
   </picture>
 </p>
 
-<h2 align="center">Unified & Cost-Effective Observability  Database for Metrics, Logs, and Events</h2>
+<h2 align="center">Real-Time & Cloud-Native Observability  Database<br/>for metrics, logs, and traces</h2>
+
+>  Delivers sub-second querying at PB scale and exceptional cost efficiency from edge to cloud.
 
 <div align="center">
 <h3 align="center">
@@ -49,74 +51,77 @@
 </div>
 
 - [Introduction](#introduction)
-- [**Features: Why GreptimeDB**](#why-greptimedb)
-- [Architecture](https://docs.greptime.com/contributor-guide/overview/#architecture)
-- [Try it for free](#try-greptimedb)
+- [⭐ Key Features](#features)
+- [Quick Comparison](#quick-comparison)
+- [Architecture](#architecture)
+- [Try GreptimeDB](#try-greptimedb)
 - [Getting Started](#getting-started)
-- [Project Status](#project-status)
-- [Join the community](#community)
-  - [Contributing](#contributing)
+- [Build From Source](#build-from-source)
 - [Tools & Extensions](#tools--extensions)
+- [Project Status](#project-status)
+- [Community](#community)
 - [License](#license)
+- [Commercial Support](#commercial-support)
+- [Contributing](#contributing)
 - [Acknowledgement](#acknowledgement)
 
 ## Introduction
 
-**GreptimeDB** is an open-source, cloud-native, unified & cost-effective observability database for **Metrics**, **Logs**, and **Traces**. You can gain real-time insights from Edge to Cloud at Any Scale.
+**GreptimeDB** is an open-source, cloud-native database purpose-built for the unified collection and analysis of observability data (metrics, logs, and traces). Whether you’re operating on the edge, in the cloud, or across hybrid environments, GreptimeDB empowers real-time insights at massive scale — all in one system.
 
-## News
+## Features
 
-**[GreptimeDB tops JSONBench's billion-record cold run test!](https://greptime.com/blogs/2025-03-18-jsonbench-greptimedb-performance)**
+|   Feature  | Description |
+| --------- | ----------- |
+| [Unified Observability Data](https://docs.greptime.com/user-guide/concepts/why-greptimedb) | Store metrics, logs, and traces as timestamped, contextual wide events. Query via [SQL](https://docs.greptime.com/user-guide/query-data/sql), [PromQL](https://docs.greptime.com/user-guide/query-data/promql), and [streaming](https://docs.greptime.com/user-guide/flow-computation/overview). |
+| [High Performance & Cost Effective](https://docs.greptime.com/user-guide/manage-data/data-index) | Written in Rust, with a distributed query engine, [rich indexing](https://docs.greptime.com/user-guide/manage-data/data-index), and optimized columnar storage, delivering sub-second responses at PB scale. |
+| [Cloud-Native Architecture](https://docs.greptime.com/user-guide/concepts/architecture) | Designed for [Kubernetes](https://docs.greptime.com/user-guide/deployments/deploy-on-kubernetes/greptimedb-operator-management), with compute/storage separation, native object storage (AWS S3, Azure Blob, etc.) and seamless cross-cloud access. |
+| [Developer-Friendly](https://docs.greptime.com/user-guide/protocols/overview) | Access via SQL/PromQL interfaces, REST API, MySQL/PostgreSQL protocols, and popular ingestion [protocols](https://docs.greptime.com/user-guide/protocols/overview). |
+| [Flexible Deployment](https://docs.greptime.com/user-guide/deployments/overview) | Deploy anywhere: edge (including ARM/[Android](https://docs.greptime.com/user-guide/deployments/run-on-android)) or cloud, with unified APIs and efficient data sync. |
 
-## Why GreptimeDB
+Learn more in [Why GreptimeDB](https://docs.greptime.com/user-guide/concepts/why-greptimedb) and [Observability 2.0 and the Database for It](https://greptime.com/blogs/2025-04-25-greptimedb-observability2-new-database).
 
-Our core developers have been building observability data platforms for years. Based on our best practices, GreptimeDB was born to give you:
+## Quick Comparison
 
-* **Unified Processing of Observability Data**
+| Feature                         | GreptimeDB            | Traditional TSDB   | Log Stores      |
+|----------------------------------|-----------------------|--------------------|-----------------|
+| Data Types                      | Metrics, Logs, Traces | Metrics only       | Logs only       |
+| Query Language                  | SQL, PromQL, Streaming|  Custom/PromQL     | Custom/DSL      |
+| Deployment                      | Edge + Cloud          | Cloud/On-prem      | Mostly central  |
+| Indexing & Performance          | PB-Scale, Sub-second  | Varies             | Varies          |
+| Integration                     | REST, SQL, Common protocols | Varies     | Varies          |
 
-  A unified database that treats metrics, logs, and traces as timestamped wide events with context, supporting [SQL](https://docs.greptime.com/user-guide/query-data/sql)/[PromQL](https://docs.greptime.com/user-guide/query-data/promql) queries and [stream processing](https://docs.greptime.com/user-guide/flow-computation/overview) to simplify complex data stacks.
+**Performance:**
+* [GreptimeDB tops JSONBench's billion-record cold run test!](https://greptime.com/blogs/2025-03-18-jsonbench-greptimedb-performance)
+* [TSBS Benchmark](https://github.com/GreptimeTeam/greptimedb/tree/main/docs/benchmarks/tsbs)
 
-* **High Performance and Cost-effective**
+Read [more benchmark reports](https://docs.greptime.com/user-guide/concepts/features-that-you-concern#how-is-greptimedbs-performance-compared-to-other-solutions).
 
-   Written in Rust, combines a distributed query engine with [rich indexing](https://docs.greptime.com/user-guide/manage-data/data-index) (inverted, fulltext, skip data, and vector) and optimized columnar storage to deliver sub-second responses on petabyte-scale data and high-cost efficiency.
+## Architecture
 
-* **Cloud-native Distributed Database**
-
-  Built for [Kubernetes](https://docs.greptime.com/user-guide/deployments/deploy-on-kubernetes/greptimedb-operator-management). GreptimeDB achieves seamless scalability with its [cloud-native architecture](https://docs.greptime.com/user-guide/concepts/architecture) of separated compute and storage, built on object storage (AWS S3, Azure Blob Storage, etc.) while enabling cross-cloud deployment through a unified data access layer.
-
-* **Developer-Friendly**
-
-  Access standardized SQL/PromQL interfaces through built-in web dashboard, REST API, and MySQL/PostgreSQL protocols. Supports widely adopted data ingestion [protocols](https://docs.greptime.com/user-guide/protocols/overview) for seamless migration and integration.
-
-* **Flexible Deployment Options**
-
-  Deploy GreptimeDB anywhere from ARM-based edge devices to cloud environments with unified APIs and bandwidth-efficient data synchronization. Query edge and cloud data seamlessly through identical APIs. [Learn how to run on Android](https://docs.greptime.com/user-guide/deployments/run-on-android/).
-
-For more detailed info please read  [Why GreptimeDB](https://docs.greptime.com/user-guide/concepts/why-greptimedb).
+* Read the [architecture](https://docs.greptime.com/contributor-guide/overview/#architecture) document.
+* [DeepWiki](https://deepwiki.com/GreptimeTeam/greptimedb/1-overview) provides an in-depth look at GreptimeDB:
+  <img alt="GreptimeDB System Overview" src="docs/architecture.png">
 
 ## Try GreptimeDB
 
 ### 1. [Live Demo](https://greptime.com/playground)
 
-Try out the features of GreptimeDB right from your browser.
+Experience GreptimeDB directly in your browser.
 
 ### 2. [GreptimeCloud](https://console.greptime.cloud/)
 
 Start instantly with a free cluster.
 
-### 3. Docker Image
-
-To install GreptimeDB locally, the recommended way is via Docker:
+### 3. Docker (Local Quickstart)
 
 ```shell
 docker pull greptime/greptimedb
 ```
 
-Start a GreptimeDB container with:
-
 ```shell
 docker run -p 127.0.0.1:4000-4003:4000-4003 \
-  -v "$(pwd)/greptimedb:./greptimedb_data" \
+  -v "$(pwd)/greptimedb_data:/greptimedb_data" \
   --name greptime --rm \
   greptime/greptimedb:latest standalone start \
   --http-addr 0.0.0.0:4000 \
@@ -124,114 +129,90 @@ docker run -p 127.0.0.1:4000-4003:4000-4003 \
   --mysql-addr 0.0.0.0:4002 \
   --postgres-addr 0.0.0.0:4003
 ```
+Dashboard: [http://localhost:4000/dashboard](http://localhost:4000/dashboard)
+[Full Install Guide](https://docs.greptime.com/getting-started/installation/overview)
 
-Access the dashboard via `http://localhost:4000/dashboard`.
-
-Read more about [Installation](https://docs.greptime.com/getting-started/installation/overview) on docs.
+**Troubleshooting:**
+* Cannot connect to the database? Ensure that ports `4000`, `4001`, `4002`, and `4003` are not blocked by a firewall or used by other services.
+* Failed to start? Check the container logs with `docker logs greptime` for further details.
 
 ## Getting Started
 
-* [Quickstart](https://docs.greptime.com/getting-started/quick-start)
-* [User Guide](https://docs.greptime.com/user-guide/overview)
-* [Demos](https://github.com/GreptimeTeam/demo-scene)
-* [FAQ](https://docs.greptime.com/faq-and-others/faq)
+- [Quickstart](https://docs.greptime.com/getting-started/quick-start)
+- [User Guide](https://docs.greptime.com/user-guide/overview)
+- [Demo Scenes](https://github.com/GreptimeTeam/demo-scene)
+- [FAQ](https://docs.greptime.com/faq-and-others/faq)
 
-## Build
-
-Check the prerequisite:
+## Build From Source
 
+**Prerequisites:**
 * [Rust toolchain](https://www.rust-lang.org/tools/install) (nightly)
 * [Protobuf compiler](https://grpc.io/docs/protoc-installation/) (>= 3.15)
 * C/C++ building essentials, including `gcc`/`g++`/`autoconf` and glibc library (eg. `libc6-dev` on Ubuntu and `glibc-devel` on Fedora)
 * Python toolchain (optional): Required only if using some test scripts.
 
-Build GreptimeDB binary:
-
-```shell
+**Build and Run:**
+```bash
 make
-```
-
-Run a standalone server:
-
-```shell
 cargo run -- standalone start
 ```
 
 ## Tools & Extensions
 
-### Kubernetes
-
-- [GreptimeDB Operator](https://github.com/GrepTimeTeam/greptimedb-operator)
-
-### Dashboard
-
-- [The dashboard UI for GreptimeDB](https://github.com/GreptimeTeam/dashboard)
-
-### SDK
-
-- [GreptimeDB Go Ingester](https://github.com/GreptimeTeam/greptimedb-ingester-go)
-- [GreptimeDB Java Ingester](https://github.com/GreptimeTeam/greptimedb-ingester-java)
-- [GreptimeDB C++ Ingester](https://github.com/GreptimeTeam/greptimedb-ingester-cpp)
-- [GreptimeDB Erlang Ingester](https://github.com/GreptimeTeam/greptimedb-ingester-erl)
-- [GreptimeDB Rust Ingester](https://github.com/GreptimeTeam/greptimedb-ingester-rust)
-- [GreptimeDB JavaScript Ingester](https://github.com/GreptimeTeam/greptimedb-ingester-js)
-
-### Grafana Dashboard
-
-Our official Grafana dashboard for monitoring GreptimeDB is available at [grafana](grafana/README.md) directory.
+- **Kubernetes:** [GreptimeDB Operator](https://github.com/GrepTimeTeam/greptimedb-operator)
+- **Helm Charts:** [Greptime Helm Charts](https://github.com/GreptimeTeam/helm-charts)
+- **Dashboard:** [Web UI](https://github.com/GreptimeTeam/dashboard)
+- **SDKs/Ingester:** [Go](https://github.com/GreptimeTeam/greptimedb-ingester-go), [Java](https://github.com/GreptimeTeam/greptimedb-ingester-java), [C++](https://github.com/GreptimeTeam/greptimedb-ingester-cpp), [Erlang](https://github.com/GreptimeTeam/greptimedb-ingester-erl), [Rust](https://github.com/GreptimeTeam/greptimedb-ingester-rust), [JS](https://github.com/GreptimeTeam/greptimedb-ingester-js)
+- **Grafana**: [Official Dashboard](https://github.com/GreptimeTeam/greptimedb/blob/main/grafana/README.md)
 
 ## Project Status
 
-GreptimeDB is currently in Beta. We are targeting GA (General Availability) with v1.0 release by Early 2025.
+> **Status:** Beta.
+> **GA (v1.0):** Targeted for mid 2025.
 
-While in Beta, GreptimeDB is already:
-
-* Being used in production by early adopters
-* Actively maintained with regular releases, [about version number](https://docs.greptime.com/nightly/reference/about-greptimedb-version)
-* Suitable for testing and evaluation
+- Being used in production by early adopters
+- Stable, actively maintained, with regular releases ([version info](https://docs.greptime.com/nightly/reference/about-greptimedb-version))
+- Suitable for evaluation and pilot deployments
 
 For production use, we recommend using the latest stable release.
+[![Star History Chart](https://api.star-history.com/svg?repos=GreptimeTeam/GreptimeDB&type=Date)](https://www.star-history.com/#GreptimeTeam/GreptimeDB&Date)
+
+If you find this project useful, a ⭐ would mean a lot to us!
+<img alt="Known Users" src="https://greptime.com/logo/img/users.png"/>
 
 ## Community
 
-Our core team is thrilled to see you participate in any ways you like. When you are stuck, try to
-ask for help by filling an issue with a detailed description of what you were trying to do
-and what went wrong. If you have any questions or if you would like to get involved in our
-community, please check out:
+We invite you to engage and contribute!
 
-- GreptimeDB Community on [Slack](https://greptime.com/slack)
-- GreptimeDB [GitHub Discussions forum](https://github.com/GreptimeTeam/greptimedb/discussions)
-- Greptime official [website](https://greptime.com)
-
-In addition, you may:
-
-- View our official [Blog](https://greptime.com/blogs/)
-- Connect us with [Linkedin](https://www.linkedin.com/company/greptime/)
-- Follow us on [Twitter](https://twitter.com/greptime)
-
-## Commercial Support
-
-If you are running GreptimeDB OSS in your organization, we offer additional
-enterprise add-ons, installation services, training, and consulting. [Contact
-us](https://greptime.com/contactus) and we will reach out to you with more
-detail of our commercial license.
+- [Slack](https://greptime.com/slack)
+- [Discussions](https://github.com/GreptimeTeam/greptimedb/discussions)
+- [Official Website](https://greptime.com/)
+- [Blog](https://greptime.com/blogs/)
+- [LinkedIn](https://www.linkedin.com/company/greptime/)
+- [X (Twitter)](https://X.com/greptime)
+- [YouTube](https://www.youtube.com/@greptime)
 
 ## License
 
-GreptimeDB uses the [Apache License 2.0](https://apache.org/licenses/LICENSE-2.0.txt) to strike a balance between
-open contributions and allowing you to use the software however you want.
+GreptimeDB is licensed under the [Apache License 2.0](https://apache.org/licenses/LICENSE-2.0.txt).
+
+## Commercial Support
+
+Running GreptimeDB in your organization?
+We offer enterprise add-ons, services, training, and consulting.
+[Contact us](https://greptime.com/contactus) for details.
 
 ## Contributing
 
-Please refer to [contribution guidelines](CONTRIBUTING.md) and [internal concepts docs](https://docs.greptime.com/contributor-guide/overview.html) for more information.
+- Read our [Contribution Guidelines](https://github.com/GreptimeTeam/greptimedb/blob/main/CONTRIBUTING.md).
+- Explore [Internal Concepts](https://docs.greptime.com/contributor-guide/overview.html) and [DeepWiki](https://deepwiki.com/GreptimeTeam/greptimedb).
+- Pick up a [good first issue](https://github.com/GreptimeTeam/greptimedb/issues?q=is%3Aissue+is%3Aopen+label%3A%22good+first+issue%22) and join the #contributors [Slack](https://greptime.com/slack) channel.
 
 ## Acknowledgement
 
-Special thanks to all the contributors who have propelled GreptimeDB forward. For a complete list of contributors, please refer to [AUTHOR.md](AUTHOR.md).
+Special thanks to all contributors! See [AUTHORS.md](https://github.com/GreptimeTeam/greptimedb/blob/main/AUTHOR.md).
 
-- GreptimeDB uses [Apache Arrow™](https://arrow.apache.org/) as the memory model and [Apache Parquet™](https://parquet.apache.org/) as the persistent file format.
-- GreptimeDB's query engine is powered by [Apache Arrow DataFusion™](https://arrow.apache.org/datafusion/).
-- [Apache OpenDAL™](https://opendal.apache.org) gives GreptimeDB a very general and elegant data access abstraction layer.
-- GreptimeDB's meta service is based on [etcd](https://etcd.io/).
-
-<img alt="Known Users" src="https://greptime.com/logo/img/users.png"/>
+- Uses [Apache Arrow™](https://arrow.apache.org/) (memory model)
+- [Apache Parquet™](https://parquet.apache.org/) (file storage)
+- [Apache Arrow DataFusion™](https://arrow.apache.org/datafusion/) (query engine)
+- [Apache OpenDAL™](https://opendal.apache.org/) (data access abstraction)

config/config.md

@@ -27,6 +27,7 @@
 | `http.body_limit` | String | `64MB` | HTTP request body limit.<br/>The following units are supported: `B`, `KB`, `KiB`, `MB`, `MiB`, `GB`, `GiB`, `TB`, `TiB`, `PB`, `PiB`.<br/>Set to 0 to disable limit. |
 | `http.enable_cors` | Bool | `true` | HTTP CORS support, it's turned on by default<br/>This allows browser to access http APIs without CORS restrictions |
 | `http.cors_allowed_origins` | Array | Unset | Customize allowed origins for HTTP CORS. |
+| `http.prom_validation_mode` | String | `strict` | Whether to enable validation for Prometheus remote write requests.<br/>Available options:<br/>- strict: deny invalid UTF-8 strings (default).<br/>- lossy: allow invalid UTF-8 strings, replace invalid characters with REPLACEMENT_CHARACTER(U+FFFD).<br/>- unchecked: do not valid strings. |
 | `grpc` | -- | -- | The gRPC server options. |
 | `grpc.bind_addr` | String | `127.0.0.1:4001` | The address to bind the gRPC server. |
 | `grpc.runtime_size` | Integer | `8` | The number of server worker threads. |
@@ -99,7 +100,7 @@
 | `query` | -- | -- | The query engine options. |
 | `query.parallelism` | Integer | `0` | Parallelism of the query engine.<br/>Default to 0, which means the number of CPU cores. |
 | `storage` | -- | -- | The data storage options. |
-| `storage.data_home` | String | `./greptimedb_data/` | The working home directory. |
+| `storage.data_home` | String | `./greptimedb_data` | The working home directory. |
 | `storage.type` | String | `File` | The storage type used to store the data.<br/>- `File`: the data is stored in the local file system.<br/>- `S3`: the data is stored in the S3 object storage.<br/>- `Gcs`: the data is stored in the Google Cloud Storage.<br/>- `Azblob`: the data is stored in the Azure Blob Storage.<br/>- `Oss`: the data is stored in the Aliyun OSS. |
 | `storage.cache_path` | String | Unset | Read cache configuration for object storage such as 'S3' etc, it's configured by default when using object storage. It is recommended to configure it when using object storage for better performance.<br/>A local file directory, defaults to `{data_home}`. An empty string means disabling. |
 | `storage.cache_capacity` | String | Unset | The local file cache capacity in bytes. If your disk space is sufficient, it is recommended to set it larger. |
@@ -122,6 +123,7 @@
 | `storage.http_client.connect_timeout` | String | `30s` | The timeout for only the connect phase of a http client. |
 | `storage.http_client.timeout` | String | `30s` | The total request timeout, applied from when the request starts connecting until the response body has finished.<br/>Also considered a total deadline. |
 | `storage.http_client.pool_idle_timeout` | String | `90s` | The timeout for idle sockets being kept-alive. |
+| `storage.http_client.skip_ssl_validation` | Bool | `false` | To skip the ssl verification<br/>**Security Notice**: Setting `skip_ssl_validation = true` disables certificate verification, making connections vulnerable to man-in-the-middle attacks. Only use this in development or trusted private networks. |
 | `[[region_engine]]` | -- | -- | The region engine options. You can configure multiple region engines. |
 | `region_engine.mito` | -- | -- | The Mito engine options. |
 | `region_engine.mito.num_workers` | Integer | `8` | Number of region workers. |
@@ -145,6 +147,7 @@
 | `region_engine.mito.write_cache_ttl` | String | Unset | TTL for write cache. |
 | `region_engine.mito.sst_write_buffer_size` | String | `8MB` | Buffer size for SST writing. |
 | `region_engine.mito.parallel_scan_channel_size` | Integer | `32` | Capacity of the channel to send data from parallel scan tasks to the main task. |
+| `region_engine.mito.max_concurrent_scan_files` | Integer | `128` | Maximum number of SST files to scan concurrently. |
 | `region_engine.mito.allow_stale_entries` | Bool | `false` | Whether to allow stale WAL entries read during replay. |
 | `region_engine.mito.min_compaction_interval` | String | `0m` | Minimum time interval between two compactions.<br/>To align with the old behavior, the default value is 0 (no restrictions). |
 | `region_engine.mito.index` | -- | -- | The options for index in Mito engine. |
@@ -154,6 +157,7 @@
 | `region_engine.mito.index.metadata_cache_size` | String | `64MiB` | Cache size for inverted index metadata. |
 | `region_engine.mito.index.content_cache_size` | String | `128MiB` | Cache size for inverted index content. |
 | `region_engine.mito.index.content_cache_page_size` | String | `64KiB` | Page size for inverted index content cache. |
+| `region_engine.mito.index.result_cache_size` | String | `128MiB` | Cache size for index result. |
 | `region_engine.mito.inverted_index` | -- | -- | The options for inverted index in Mito engine. |
 | `region_engine.mito.inverted_index.create_on_flush` | String | `auto` | Whether to create the index on flush.<br/>- `auto`: automatically (default)<br/>- `disable`: never |
 | `region_engine.mito.inverted_index.create_on_compaction` | String | `auto` | Whether to create the index on compaction.<br/>- `auto`: automatically (default)<br/>- `disable`: never |
@@ -188,17 +192,18 @@
 | `logging.max_log_files` | Integer | `720` | The maximum amount of log files. |
 | `logging.tracing_sample_ratio` | -- | -- | The percentage of tracing will be sampled and exported.<br/>Valid range `[0, 1]`, 1 means all traces are sampled, 0 means all traces are not sampled, the default value is 1.<br/>ratio > 1 are treated as 1. Fractions < 0 are treated as 0 |
 | `logging.tracing_sample_ratio.default_ratio` | Float | `1.0` | -- |
-| `logging.slow_query` | -- | -- | The slow query log options. |
-| `logging.slow_query.enable` | Bool | `false` | Whether to enable slow query log. |
-| `logging.slow_query.threshold` | String | Unset | The threshold of slow query. |
-| `logging.slow_query.sample_ratio` | Float | Unset | The sampling ratio of slow query log. The value should be in the range of (0, 1]. |
-| `export_metrics` | -- | -- | The datanode can export its metrics and send to Prometheus compatible service (e.g. send to `greptimedb` itself) from remote-write API.<br/>This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape. |
+| `slow_query` | -- | -- | The slow query log options. |
+| `slow_query.enable` | Bool | `false` | Whether to enable slow query log. |
+| `slow_query.record_type` | String | Unset | The record type of slow queries. It can be `system_table` or `log`. |
+| `slow_query.threshold` | String | Unset | The threshold of slow query. |
+| `slow_query.sample_ratio` | Float | Unset | The sampling ratio of slow query log. The value should be in the range of (0, 1]. |
+| `export_metrics` | -- | -- | The standalone can export its metrics and send to Prometheus compatible service (e.g. `greptimedb`) from remote-write API.<br/>This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape. |
 | `export_metrics.enable` | Bool | `false` | whether enable export metrics. |
 | `export_metrics.write_interval` | String | `30s` | The interval of export metrics. |
 | `export_metrics.self_import` | -- | -- | For `standalone` mode, `self_import` is recommended to collect metrics generated by itself<br/>You must create the database before enabling it. |
 | `export_metrics.self_import.db` | String | Unset | -- |
 | `export_metrics.remote_write` | -- | -- | -- |
-| `export_metrics.remote_write.url` | String | `""` | The url the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`. |
+| `export_metrics.remote_write.url` | String | `""` | The prometheus remote write endpoint that the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`. |
 | `export_metrics.remote_write.headers` | InlineTable | -- | HTTP headers of Prometheus remote-write carry. |
 | `tracing` | -- | -- | The tracing options. Only effect when compiled with `tokio-console` feature. |
 | `tracing.tokio_console_addr` | String | Unset | The tokio console address. |
@@ -224,10 +229,12 @@
 | `http.body_limit` | String | `64MB` | HTTP request body limit.<br/>The following units are supported: `B`, `KB`, `KiB`, `MB`, `MiB`, `GB`, `GiB`, `TB`, `TiB`, `PB`, `PiB`.<br/>Set to 0 to disable limit. |
 | `http.enable_cors` | Bool | `true` | HTTP CORS support, it's turned on by default<br/>This allows browser to access http APIs without CORS restrictions |
 | `http.cors_allowed_origins` | Array | Unset | Customize allowed origins for HTTP CORS. |
+| `http.prom_validation_mode` | String | `strict` | Whether to enable validation for Prometheus remote write requests.<br/>Available options:<br/>- strict: deny invalid UTF-8 strings (default).<br/>- lossy: allow invalid UTF-8 strings, replace invalid characters with REPLACEMENT_CHARACTER(U+FFFD).<br/>- unchecked: do not valid strings. |
 | `grpc` | -- | -- | The gRPC server options. |
 | `grpc.bind_addr` | String | `127.0.0.1:4001` | The address to bind the gRPC server. |
 | `grpc.server_addr` | String | `127.0.0.1:4001` | The address advertised to the metasrv, and used for connections from outside the host.<br/>If left empty or unset, the server will automatically use the IP address of the first network interface<br/>on the host, with the same port number as the one specified in `grpc.bind_addr`. |
 | `grpc.runtime_size` | Integer | `8` | The number of server worker threads. |
+| `grpc.flight_compression` | String | `arrow_ipc` | Compression mode for frontend side Arrow IPC service. Available options:<br/>- `none`: disable all compression<br/>- `transport`: only enable gRPC transport compression (zstd)<br/>- `arrow_ipc`: only enable Arrow IPC compression (lz4)<br/>- `all`: enable all compression.<br/>Default to `none` |
 | `grpc.tls` | -- | -- | gRPC server TLS options, see `mysql.tls` section. |
 | `grpc.tls.mode` | String | `disable` | TLS mode. |
 | `grpc.tls.cert_path` | String | Unset | Certificate file path. |
@@ -288,17 +295,17 @@
 | `logging.max_log_files` | Integer | `720` | The maximum amount of log files. |
 | `logging.tracing_sample_ratio` | -- | -- | The percentage of tracing will be sampled and exported.<br/>Valid range `[0, 1]`, 1 means all traces are sampled, 0 means all traces are not sampled, the default value is 1.<br/>ratio > 1 are treated as 1. Fractions < 0 are treated as 0 |
 | `logging.tracing_sample_ratio.default_ratio` | Float | `1.0` | -- |
-| `logging.slow_query` | -- | -- | The slow query log options. |
-| `logging.slow_query.enable` | Bool | `false` | Whether to enable slow query log. |
-| `logging.slow_query.threshold` | String | Unset | The threshold of slow query. |
-| `logging.slow_query.sample_ratio` | Float | Unset | The sampling ratio of slow query log. The value should be in the range of (0, 1]. |
-| `export_metrics` | -- | -- | The datanode can export its metrics and send to Prometheus compatible service (e.g. send to `greptimedb` itself) from remote-write API.<br/>This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape. |
+| `slow_query` | -- | -- | The slow query log options. |
+| `slow_query.enable` | Bool | `true` | Whether to enable slow query log. |
+| `slow_query.record_type` | String | `system_table` | The record type of slow queries. It can be `system_table` or `log`.<br/>If `system_table` is selected, the slow queries will be recorded in a system table `greptime_private.slow_queries`.<br/>If `log` is selected, the slow queries will be logged in a log file `greptimedb-slow-queries.*`. |
+| `slow_query.threshold` | String | `30s` | The threshold of slow query. It can be human readable time string, for example: `10s`, `100ms`, `1s`. |
+| `slow_query.sample_ratio` | Float | `1.0` | The sampling ratio of slow query log. The value should be in the range of (0, 1]. For example, `0.1` means 10% of the slow queries will be logged and `1.0` means all slow queries will be logged. |
+| `slow_query.ttl` | String | `30d` | The TTL of the `slow_queries` system table. Default is `30d` when `record_type` is `system_table`. |
+| `export_metrics` | -- | -- | The frontend can export its metrics and send to Prometheus compatible service (e.g. `greptimedb` itself) from remote-write API.<br/>This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape. |
 | `export_metrics.enable` | Bool | `false` | whether enable export metrics. |
 | `export_metrics.write_interval` | String | `30s` | The interval of export metrics. |
-| `export_metrics.self_import` | -- | -- | For `standalone` mode, `self_import` is recommend to collect metrics generated by itself<br/>You must create the database before enabling it. |
-| `export_metrics.self_import.db` | String | Unset | -- |
 | `export_metrics.remote_write` | -- | -- | -- |
-| `export_metrics.remote_write.url` | String | `""` | The url the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`. |
+| `export_metrics.remote_write.url` | String | `""` | The prometheus remote write endpoint that the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`. |
 | `export_metrics.remote_write.headers` | InlineTable | -- | HTTP headers of Prometheus remote-write carry. |
 | `tracing` | -- | -- | The tracing options. Only effect when compiled with `tokio-console` feature. |
 | `tracing.tokio_console_addr` | String | Unset | The tokio console address. |
@@ -308,22 +315,31 @@
 
 | Key | Type | Default | Descriptions |
 | --- | -----| ------- | ----------- |
-| `data_home` | String | `./greptimedb_data/metasrv/` | The working home directory. |
-| `bind_addr` | String | `127.0.0.1:3002` | The bind address of metasrv. |
-| `server_addr` | String | `127.0.0.1:3002` | The communication server address for the frontend and datanode to connect to metasrv.<br/>If left empty or unset, the server will automatically use the IP address of the first network interface<br/>on the host, with the same port number as the one specified in `bind_addr`. |
+| `data_home` | String | `./greptimedb_data` | The working home directory. |
 | `store_addrs` | Array | -- | Store server address default to etcd store.<br/>For postgres store, the format is:<br/>"password=password dbname=postgres user=postgres host=localhost port=5432"<br/>For etcd store, the format is:<br/>"127.0.0.1:2379" |
 | `store_key_prefix` | String | `""` | If it's not empty, the metasrv will store all data with this key prefix. |
-| `backend` | String | `etcd_store` | The datastore for meta server.<br/>Available values:<br/>- `etcd_store` (default value)<br/>- `memory_store`<br/>- `postgres_store` |
+| `backend` | String | `etcd_store` | The datastore for meta server.<br/>Available values:<br/>- `etcd_store` (default value)<br/>- `memory_store`<br/>- `postgres_store`<br/>- `mysql_store` |
 | `meta_table_name` | String | `greptime_metakv` | Table name in RDS to store metadata. Effect when using a RDS kvbackend.<br/>**Only used when backend is `postgres_store`.** |
 | `meta_election_lock_id` | Integer | `1` | Advisory lock id in PostgreSQL for election. Effect when using PostgreSQL as kvbackend<br/>Only used when backend is `postgres_store`. |
 | `selector` | String | `round_robin` | Datanode selector type.<br/>- `round_robin` (default value)<br/>- `lease_based`<br/>- `load_based`<br/>For details, please see "https://docs.greptime.com/developer-guide/metasrv/selector". |
 | `use_memory_store` | Bool | `false` | Store data in memory. |
 | `enable_region_failover` | Bool | `false` | Whether to enable region failover.<br/>This feature is only available on GreptimeDB running on cluster mode and<br/>- Using Remote WAL<br/>- Using shared storage (e.g., s3). |
+| `allow_region_failover_on_local_wal` | Bool | `false` | Whether to allow region failover on local WAL.<br/>**This option is not recommended to be set to true, because it may lead to data loss during failover.** |
 | `node_max_idle_time` | String | `24hours` | Max allowed idle time before removing node info from metasrv memory. |
 | `enable_telemetry` | Bool | `true` | Whether to enable greptimedb telemetry. Enabled by default. |
 | `runtime` | -- | -- | The runtime options. |
 | `runtime.global_rt_size` | Integer | `8` | The number of threads to execute the runtime for global read operations. |
 | `runtime.compact_rt_size` | Integer | `4` | The number of threads to execute the runtime for global write operations. |
+| `grpc` | -- | -- | The gRPC server options. |
+| `grpc.bind_addr` | String | `127.0.0.1:3002` | The address to bind the gRPC server. |
+| `grpc.server_addr` | String | `127.0.0.1:3002` | The communication server address for the frontend and datanode to connect to metasrv.<br/>If left empty or unset, the server will automatically use the IP address of the first network interface<br/>on the host, with the same port number as the one specified in `bind_addr`. |
+| `grpc.runtime_size` | Integer | `8` | The number of server worker threads. |
+| `grpc.max_recv_message_size` | String | `512MB` | The maximum receive message size for gRPC server. |
+| `grpc.max_send_message_size` | String | `512MB` | The maximum send message size for gRPC server. |
+| `http` | -- | -- | The HTTP server options. |
+| `http.addr` | String | `127.0.0.1:4000` | The address to bind the HTTP server. |
+| `http.timeout` | String | `0s` | HTTP request timeout. Set to 0 to disable timeout. |
+| `http.body_limit` | String | `64MB` | HTTP request body limit.<br/>The following units are supported: `B`, `KB`, `KiB`, `MB`, `MiB`, `GB`, `GiB`, `TB`, `TiB`, `PB`, `PiB`.<br/>Set to 0 to disable limit. |
 | `procedure` | -- | -- | Procedure storage options. |
 | `procedure.max_retry_times` | Integer | `12` | Procedure max retry time. |
 | `procedure.retry_delay` | String | `500ms` | Initial retry delay of procedures, increases exponentially |
@@ -361,17 +377,11 @@
 | `logging.max_log_files` | Integer | `720` | The maximum amount of log files. |
 | `logging.tracing_sample_ratio` | -- | -- | The percentage of tracing will be sampled and exported.<br/>Valid range `[0, 1]`, 1 means all traces are sampled, 0 means all traces are not sampled, the default value is 1.<br/>ratio > 1 are treated as 1. Fractions < 0 are treated as 0 |
 | `logging.tracing_sample_ratio.default_ratio` | Float | `1.0` | -- |
-| `logging.slow_query` | -- | -- | The slow query log options. |
-| `logging.slow_query.enable` | Bool | `false` | Whether to enable slow query log. |
-| `logging.slow_query.threshold` | String | Unset | The threshold of slow query. |
-| `logging.slow_query.sample_ratio` | Float | Unset | The sampling ratio of slow query log. The value should be in the range of (0, 1]. |
-| `export_metrics` | -- | -- | The datanode can export its metrics and send to Prometheus compatible service (e.g. send to `greptimedb` itself) from remote-write API.<br/>This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape. |
+| `export_metrics` | -- | -- | The metasrv can export its metrics and send to Prometheus compatible service (e.g. `greptimedb` itself) from remote-write API.<br/>This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape. |
 | `export_metrics.enable` | Bool | `false` | whether enable export metrics. |
 | `export_metrics.write_interval` | String | `30s` | The interval of export metrics. |
-| `export_metrics.self_import` | -- | -- | For `standalone` mode, `self_import` is recommend to collect metrics generated by itself<br/>You must create the database before enabling it. |
-| `export_metrics.self_import.db` | String | Unset | -- |
 | `export_metrics.remote_write` | -- | -- | -- |
-| `export_metrics.remote_write.url` | String | `""` | The url the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`. |
+| `export_metrics.remote_write.url` | String | `""` | The prometheus remote write endpoint that the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`. |
 | `export_metrics.remote_write.headers` | InlineTable | -- | HTTP headers of Prometheus remote-write carry. |
 | `tracing` | -- | -- | The tracing options. Only effect when compiled with `tokio-console` feature. |
 | `tracing.tokio_console_addr` | String | Unset | The tokio console address. |
@@ -397,6 +407,7 @@
 | `grpc.runtime_size` | Integer | `8` | The number of server worker threads. |
 | `grpc.max_recv_message_size` | String | `512MB` | The maximum receive message size for gRPC server. |
 | `grpc.max_send_message_size` | String | `512MB` | The maximum send message size for gRPC server. |
+| `grpc.flight_compression` | String | `arrow_ipc` | Compression mode for datanode side Arrow IPC service. Available options:<br/>- `none`: disable all compression<br/>- `transport`: only enable gRPC transport compression (zstd)<br/>- `arrow_ipc`: only enable Arrow IPC compression (lz4)<br/>- `all`: enable all compression.<br/>Default to `none` |
 | `grpc.tls` | -- | -- | gRPC server TLS options, see `mysql.tls` section. |
 | `grpc.tls.mode` | String | `disable` | TLS mode. |
 | `grpc.tls.cert_path` | String | Unset | Certificate file path. |
@@ -439,7 +450,7 @@
 | `query` | -- | -- | The query engine options. |
 | `query.parallelism` | Integer | `0` | Parallelism of the query engine.<br/>Default to 0, which means the number of CPU cores. |
 | `storage` | -- | -- | The data storage options. |
-| `storage.data_home` | String | `./greptimedb_data/` | The working home directory. |
+| `storage.data_home` | String | `./greptimedb_data` | The working home directory. |
 | `storage.type` | String | `File` | The storage type used to store the data.<br/>- `File`: the data is stored in the local file system.<br/>- `S3`: the data is stored in the S3 object storage.<br/>- `Gcs`: the data is stored in the Google Cloud Storage.<br/>- `Azblob`: the data is stored in the Azure Blob Storage.<br/>- `Oss`: the data is stored in the Aliyun OSS. |
 | `storage.cache_path` | String | Unset | Read cache configuration for object storage such as 'S3' etc, it's configured by default when using object storage. It is recommended to configure it when using object storage for better performance.<br/>A local file directory, defaults to `{data_home}`. An empty string means disabling. |
 | `storage.cache_capacity` | String | Unset | The local file cache capacity in bytes. If your disk space is sufficient, it is recommended to set it larger. |
@@ -462,6 +473,7 @@
 | `storage.http_client.connect_timeout` | String | `30s` | The timeout for only the connect phase of a http client. |
 | `storage.http_client.timeout` | String | `30s` | The total request timeout, applied from when the request starts connecting until the response body has finished.<br/>Also considered a total deadline. |
 | `storage.http_client.pool_idle_timeout` | String | `90s` | The timeout for idle sockets being kept-alive. |
+| `storage.http_client.skip_ssl_validation` | Bool | `false` | To skip the ssl verification<br/>**Security Notice**: Setting `skip_ssl_validation = true` disables certificate verification, making connections vulnerable to man-in-the-middle attacks. Only use this in development or trusted private networks. |
 | `[[region_engine]]` | -- | -- | The region engine options. You can configure multiple region engines. |
 | `region_engine.mito` | -- | -- | The Mito engine options. |
 | `region_engine.mito.num_workers` | Integer | `8` | Number of region workers. |
@@ -485,6 +497,7 @@
 | `region_engine.mito.write_cache_ttl` | String | Unset | TTL for write cache. |
 | `region_engine.mito.sst_write_buffer_size` | String | `8MB` | Buffer size for SST writing. |
 | `region_engine.mito.parallel_scan_channel_size` | Integer | `32` | Capacity of the channel to send data from parallel scan tasks to the main task. |
+| `region_engine.mito.max_concurrent_scan_files` | Integer | `128` | Maximum number of SST files to scan concurrently. |
 | `region_engine.mito.allow_stale_entries` | Bool | `false` | Whether to allow stale WAL entries read during replay. |
 | `region_engine.mito.min_compaction_interval` | String | `0m` | Minimum time interval between two compactions.<br/>To align with the old behavior, the default value is 0 (no restrictions). |
 | `region_engine.mito.index` | -- | -- | The options for index in Mito engine. |
@@ -494,6 +507,7 @@
 | `region_engine.mito.index.metadata_cache_size` | String | `64MiB` | Cache size for inverted index metadata. |
 | `region_engine.mito.index.content_cache_size` | String | `128MiB` | Cache size for inverted index content. |
 | `region_engine.mito.index.content_cache_page_size` | String | `64KiB` | Page size for inverted index content cache. |
+| `region_engine.mito.index.result_cache_size` | String | `128MiB` | Cache size for index result. |
 | `region_engine.mito.inverted_index` | -- | -- | The options for inverted index in Mito engine. |
 | `region_engine.mito.inverted_index.create_on_flush` | String | `auto` | Whether to create the index on flush.<br/>- `auto`: automatically (default)<br/>- `disable`: never |
 | `region_engine.mito.inverted_index.create_on_compaction` | String | `auto` | Whether to create the index on compaction.<br/>- `auto`: automatically (default)<br/>- `disable`: never |
@@ -528,17 +542,11 @@
 | `logging.max_log_files` | Integer | `720` | The maximum amount of log files. |
 | `logging.tracing_sample_ratio` | -- | -- | The percentage of tracing will be sampled and exported.<br/>Valid range `[0, 1]`, 1 means all traces are sampled, 0 means all traces are not sampled, the default value is 1.<br/>ratio > 1 are treated as 1. Fractions < 0 are treated as 0 |
 | `logging.tracing_sample_ratio.default_ratio` | Float | `1.0` | -- |
-| `logging.slow_query` | -- | -- | The slow query log options. |
-| `logging.slow_query.enable` | Bool | `false` | Whether to enable slow query log. |
-| `logging.slow_query.threshold` | String | Unset | The threshold of slow query. |
-| `logging.slow_query.sample_ratio` | Float | Unset | The sampling ratio of slow query log. The value should be in the range of (0, 1]. |
-| `export_metrics` | -- | -- | The datanode can export its metrics and send to Prometheus compatible service (e.g. send to `greptimedb` itself) from remote-write API.<br/>This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape. |
+| `export_metrics` | -- | -- | The datanode can export its metrics and send to Prometheus compatible service (e.g. `greptimedb` itself) from remote-write API.<br/>This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape. |
 | `export_metrics.enable` | Bool | `false` | whether enable export metrics. |
 | `export_metrics.write_interval` | String | `30s` | The interval of export metrics. |
-| `export_metrics.self_import` | -- | -- | For `standalone` mode, `self_import` is recommend to collect metrics generated by itself<br/>You must create the database before enabling it. |
-| `export_metrics.self_import.db` | String | Unset | -- |
 | `export_metrics.remote_write` | -- | -- | -- |
-| `export_metrics.remote_write.url` | String | `""` | The url the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`. |
+| `export_metrics.remote_write.url` | String | `""` | The prometheus remote write endpoint that the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`. |
 | `export_metrics.remote_write.headers` | InlineTable | -- | HTTP headers of Prometheus remote-write carry. |
 | `tracing` | -- | -- | The tracing options. Only effect when compiled with `tokio-console` feature. |
 | `tracing.tokio_console_addr` | String | Unset | The tokio console address. |
@@ -584,9 +592,5 @@
 | `logging.max_log_files` | Integer | `720` | The maximum amount of log files. |
 | `logging.tracing_sample_ratio` | -- | -- | The percentage of tracing will be sampled and exported.<br/>Valid range `[0, 1]`, 1 means all traces are sampled, 0 means all traces are not sampled, the default value is 1.<br/>ratio > 1 are treated as 1. Fractions < 0 are treated as 0 |
 | `logging.tracing_sample_ratio.default_ratio` | Float | `1.0` | -- |
-| `logging.slow_query` | -- | -- | The slow query log options. |
-| `logging.slow_query.enable` | Bool | `false` | Whether to enable slow query log. |
-| `logging.slow_query.threshold` | String | Unset | The threshold of slow query. |
-| `logging.slow_query.sample_ratio` | Float | Unset | The sampling ratio of slow query log. The value should be in the range of (0, 1]. |
 | `tracing` | -- | -- | The tracing options. Only effect when compiled with `tokio-console` feature. |
 | `tracing.tokio_console_addr` | String | Unset | The tokio console address. |

config/datanode.example.toml

@@ -44,6 +44,13 @@ runtime_size = 8
 max_recv_message_size = "512MB"
 ## The maximum send message size for gRPC server.
 max_send_message_size = "512MB"
+## Compression mode for datanode side Arrow IPC service. Available options:
+## - `none`: disable all compression
+## - `transport`: only enable gRPC transport compression (zstd)
+## - `arrow_ipc`: only enable Arrow IPC compression (lz4)
+## - `all`: enable all compression.
+## Default to `none`
+flight_compression = "arrow_ipc"
 
 ## gRPC server TLS options, see `mysql.tls` section.
 [grpc.tls]
@@ -252,7 +259,7 @@ parallelism = 0
 ## The data storage options.
 [storage]
 ## The working home directory.
-data_home = "./greptimedb_data/"
+data_home = "./greptimedb_data"
 
 ## The storage type used to store the data.
 ## - `File`: the data is stored in the local file system.
@@ -360,6 +367,10 @@ timeout = "30s"
 ## The timeout for idle sockets being kept-alive.
 pool_idle_timeout = "90s"
 
+## To skip the ssl verification
+## **Security Notice**: Setting `skip_ssl_validation = true` disables certificate verification, making connections vulnerable to man-in-the-middle attacks. Only use this in development or trusted private networks.
+skip_ssl_validation = false
+
 # Custom storage options
 # [[storage.providers]]
 # name = "S3"
@@ -463,6 +474,9 @@ sst_write_buffer_size = "8MB"
 ## Capacity of the channel to send data from parallel scan tasks to the main task.
 parallel_scan_channel_size = 32
 
+## Maximum number of SST files to scan concurrently.
+max_concurrent_scan_files = 128
+
 ## Whether to allow stale WAL entries read during replay.
 allow_stale_entries = false
 
@@ -499,6 +513,9 @@ content_cache_size = "128MiB"
 ## Page size for inverted index content cache.
 content_cache_page_size = "64KiB"
 
+## Cache size for index result.
+result_cache_size = "128MiB"
+
 ## The options for inverted index in Mito engine.
 [region_engine.mito.inverted_index]
 
@@ -632,37 +649,16 @@ max_log_files = 720
 [logging.tracing_sample_ratio]
 default_ratio = 1.0
 
-## The slow query log options.
-[logging.slow_query]
-## Whether to enable slow query log.
-enable = false
-
-## The threshold of slow query.
-## @toml2docs:none-default
-threshold = "10s"
-
-## The sampling ratio of slow query log. The value should be in the range of (0, 1].
-## @toml2docs:none-default
-sample_ratio = 1.0
-
-## The datanode can export its metrics and send to Prometheus compatible service (e.g. send to `greptimedb` itself) from remote-write API.
+## The datanode can export its metrics and send to Prometheus compatible service (e.g. `greptimedb` itself) from remote-write API.
 ## This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape.
 [export_metrics]
-
 ## whether enable export metrics.
 enable = false
-
 ## The interval of export metrics.
 write_interval = "30s"
 
-## For `standalone` mode, `self_import` is recommend to collect metrics generated by itself
-## You must create the database before enabling it.
-[export_metrics.self_import]
-## @toml2docs:none-default
-db = "greptime_metrics"
-
 [export_metrics.remote_write]
-## The url the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`.
+## The prometheus remote write endpoint that the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`.
 url = ""
 
 ## HTTP headers of Prometheus remote-write carry.

config/flownode.example.toml

@@ -100,19 +100,6 @@ max_log_files = 720
 [logging.tracing_sample_ratio]
 default_ratio = 1.0
 
-## The slow query log options.
-[logging.slow_query]
-## Whether to enable slow query log.
-enable = false
-
-## The threshold of slow query.
-## @toml2docs:none-default
-threshold = "10s"
-
-## The sampling ratio of slow query log. The value should be in the range of (0, 1].
-## @toml2docs:none-default
-sample_ratio = 1.0
-
 ## The tracing options. Only effect when compiled with `tokio-console` feature.
 #+ [tracing]
 ## The tokio console address.

config/frontend.example.toml

@@ -37,6 +37,12 @@ enable_cors = true
 ## Customize allowed origins for HTTP CORS.
 ## @toml2docs:none-default
 cors_allowed_origins = ["https://example.com"]
+## Whether to enable validation for Prometheus remote write requests.
+## Available options:
+## - strict: deny invalid UTF-8 strings (default).
+## - lossy: allow invalid UTF-8 strings, replace invalid characters with REPLACEMENT_CHARACTER(U+FFFD).
+## - unchecked: do not valid strings.
+prom_validation_mode = "strict"
 
 ## The gRPC server options.
 [grpc]
@@ -48,6 +54,13 @@ bind_addr = "127.0.0.1:4001"
 server_addr = "127.0.0.1:4001"
 ## The number of server worker threads.
 runtime_size = 8
+## Compression mode for frontend side Arrow IPC service. Available options:
+## - `none`: disable all compression
+## - `transport`: only enable gRPC transport compression (zstd)
+## - `arrow_ipc`: only enable Arrow IPC compression (lz4)
+## - `all`: enable all compression.
+## Default to `none`
+flight_compression = "arrow_ipc"
 
 ## gRPC server TLS options, see `mysql.tls` section.
 [grpc.tls]
@@ -223,36 +236,34 @@ max_log_files = 720
 default_ratio = 1.0
 
 ## The slow query log options.
-[logging.slow_query]
+[slow_query]
 ## Whether to enable slow query log.
-enable = false
+enable = true
 
-## The threshold of slow query.
-## @toml2docs:none-default
-threshold = "10s"
+## The record type of slow queries. It can be `system_table` or `log`.
+## If `system_table` is selected, the slow queries will be recorded in a system table `greptime_private.slow_queries`.
+## If `log` is selected, the slow queries will be logged in a log file `greptimedb-slow-queries.*`.
+record_type = "system_table"
 
-## The sampling ratio of slow query log. The value should be in the range of (0, 1].
-## @toml2docs:none-default
+## The threshold of slow query. It can be human readable time string, for example: `10s`, `100ms`, `1s`.
+threshold = "30s"
+
+## The sampling ratio of slow query log. The value should be in the range of (0, 1]. For example, `0.1` means 10% of the slow queries will be logged and `1.0` means all slow queries will be logged.
 sample_ratio = 1.0
 
-## The datanode can export its metrics and send to Prometheus compatible service (e.g. send to `greptimedb` itself) from remote-write API.
+## The TTL of the `slow_queries` system table. Default is `30d` when `record_type` is `system_table`.
+ttl = "30d"
+
+## The frontend can export its metrics and send to Prometheus compatible service (e.g. `greptimedb` itself) from remote-write API.
 ## This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape.
 [export_metrics]
-
 ## whether enable export metrics.
 enable = false
-
 ## The interval of export metrics.
 write_interval = "30s"
 
-## For `standalone` mode, `self_import` is recommend to collect metrics generated by itself
-## You must create the database before enabling it.
-[export_metrics.self_import]
-## @toml2docs:none-default
-db = "greptime_metrics"
-
 [export_metrics.remote_write]
-## The url the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`.
+## The prometheus remote write endpoint that the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`.
 url = ""
 
 ## HTTP headers of Prometheus remote-write carry.

config/metasrv.example.toml

@@ -1,13 +1,5 @@
 ## The working home directory.
-data_home = "./greptimedb_data/metasrv/"
-
-## The bind address of metasrv.
-bind_addr = "127.0.0.1:3002"
-
-## The communication server address for the frontend and datanode to connect to metasrv.
-## If left empty or unset, the server will automatically use the IP address of the first network interface
-## on the host, with the same port number as the one specified in `bind_addr`.
-server_addr = "127.0.0.1:3002"
+data_home = "./greptimedb_data"
 
 ## Store server address default to etcd store.
 ## For postgres store, the format is:
@@ -24,6 +16,7 @@ store_key_prefix = ""
 ## - `etcd_store` (default value)
 ## - `memory_store`
 ## - `postgres_store`
+## - `mysql_store`
 backend = "etcd_store"
 
 ## Table name in RDS to store metadata. Effect when using a RDS kvbackend.
@@ -50,6 +43,10 @@ use_memory_store = false
 ## - Using shared storage (e.g., s3).
 enable_region_failover = false
 
+## Whether to allow region failover on local WAL.
+## **This option is not recommended to be set to true, because it may lead to data loss during failover.**
+allow_region_failover_on_local_wal = false
+
 ## Max allowed idle time before removing node info from metasrv memory.
 node_max_idle_time = "24hours"
 
@@ -63,6 +60,32 @@ node_max_idle_time = "24hours"
 ## The number of threads to execute the runtime for global write operations.
 #+ compact_rt_size = 4
 
+## The gRPC server options.
+[grpc]
+## The address to bind the gRPC server.
+bind_addr = "127.0.0.1:3002"
+## The communication server address for the frontend and datanode to connect to metasrv.
+## If left empty or unset, the server will automatically use the IP address of the first network interface
+## on the host, with the same port number as the one specified in `bind_addr`.
+server_addr = "127.0.0.1:3002"
+## The number of server worker threads.
+runtime_size = 8
+## The maximum receive message size for gRPC server.
+max_recv_message_size = "512MB"
+## The maximum send message size for gRPC server.
+max_send_message_size = "512MB"
+
+## The HTTP server options.
+[http]
+## The address to bind the HTTP server.
+addr = "127.0.0.1:4000"
+## HTTP request timeout. Set to 0 to disable timeout.
+timeout = "0s"
+## HTTP request body limit.
+## The following units are supported: `B`, `KB`, `KiB`, `MB`, `MiB`, `GB`, `GiB`, `TB`, `TiB`, `PB`, `PiB`.
+## Set to 0 to disable limit.
+body_limit = "64MB"
+
 ## Procedure storage options.
 [procedure]
 
@@ -214,37 +237,16 @@ max_log_files = 720
 [logging.tracing_sample_ratio]
 default_ratio = 1.0
 
-## The slow query log options.
-[logging.slow_query]
-## Whether to enable slow query log.
-enable = false
-
-## The threshold of slow query.
-## @toml2docs:none-default
-threshold = "10s"
-
-## The sampling ratio of slow query log. The value should be in the range of (0, 1].
-## @toml2docs:none-default
-sample_ratio = 1.0
-
-## The datanode can export its metrics and send to Prometheus compatible service (e.g. send to `greptimedb` itself) from remote-write API.
+## The metasrv can export its metrics and send to Prometheus compatible service (e.g. `greptimedb` itself) from remote-write API.
 ## This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape.
 [export_metrics]
-
 ## whether enable export metrics.
 enable = false
-
 ## The interval of export metrics.
 write_interval = "30s"
 
-## For `standalone` mode, `self_import` is recommend to collect metrics generated by itself
-## You must create the database before enabling it.
-[export_metrics.self_import]
-## @toml2docs:none-default
-db = "greptime_metrics"
-
 [export_metrics.remote_write]
-## The url the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`.
+## The prometheus remote write endpoint that the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`.
 url = ""
 
 ## HTTP headers of Prometheus remote-write carry.

config/standalone.example.toml

@@ -43,6 +43,13 @@ enable_cors = true
 ## @toml2docs:none-default
 cors_allowed_origins = ["https://example.com"]
 
+## Whether to enable validation for Prometheus remote write requests.
+## Available options:
+## - strict: deny invalid UTF-8 strings (default).
+## - lossy: allow invalid UTF-8 strings, replace invalid characters with REPLACEMENT_CHARACTER(U+FFFD).
+## - unchecked: do not valid strings.
+prom_validation_mode = "strict"
+
 ## The gRPC server options.
 [grpc]
 ## The address to bind the gRPC server.
@@ -343,7 +350,7 @@ parallelism = 0
 ## The data storage options.
 [storage]
 ## The working home directory.
-data_home = "./greptimedb_data/"
+data_home = "./greptimedb_data"
 
 ## The storage type used to store the data.
 ## - `File`: the data is stored in the local file system.
@@ -451,6 +458,10 @@ timeout = "30s"
 ## The timeout for idle sockets being kept-alive.
 pool_idle_timeout = "90s"
 
+## To skip the ssl verification
+## **Security Notice**: Setting `skip_ssl_validation = true` disables certificate verification, making connections vulnerable to man-in-the-middle attacks. Only use this in development or trusted private networks.
+skip_ssl_validation = false
+
 # Custom storage options
 # [[storage.providers]]
 # name = "S3"
@@ -554,6 +565,9 @@ sst_write_buffer_size = "8MB"
 ## Capacity of the channel to send data from parallel scan tasks to the main task.
 parallel_scan_channel_size = 32
 
+## Maximum number of SST files to scan concurrently.
+max_concurrent_scan_files = 128
+
 ## Whether to allow stale WAL entries read during replay.
 allow_stale_entries = false
 
@@ -590,6 +604,9 @@ content_cache_size = "128MiB"
 ## Page size for inverted index content cache.
 content_cache_page_size = "64KiB"
 
+## Cache size for index result.
+result_cache_size = "128MiB"
+
 ## The options for inverted index in Mito engine.
 [region_engine.mito.inverted_index]
 
@@ -724,25 +741,27 @@ max_log_files = 720
 default_ratio = 1.0
 
 ## The slow query log options.
-[logging.slow_query]
+[slow_query]
 ## Whether to enable slow query log.
-enable = false
+#+ enable = false
+
+## The record type of slow queries. It can be `system_table` or `log`.
+## @toml2docs:none-default
+#+ record_type = "system_table"
 
 ## The threshold of slow query.
 ## @toml2docs:none-default
-threshold = "10s"
+#+ threshold = "10s"
 
 ## The sampling ratio of slow query log. The value should be in the range of (0, 1].
 ## @toml2docs:none-default
-sample_ratio = 1.0
+#+ sample_ratio = 1.0
 
-## The datanode can export its metrics and send to Prometheus compatible service (e.g. send to `greptimedb` itself) from remote-write API.
+## The standalone can export its metrics and send to Prometheus compatible service (e.g. `greptimedb`) from remote-write API.
 ## This is only used for `greptimedb` to export its own metrics internally. It's different from prometheus scrape.
 [export_metrics]
-
 ## whether enable export metrics.
 enable = false
-
 ## The interval of export metrics.
 write_interval = "30s"
 
@@ -753,7 +772,7 @@ write_interval = "30s"
 db = "greptime_metrics"
 
 [export_metrics.remote_write]
-## The url the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`.
+## The prometheus remote write endpoint that the metrics send to. The url example can be: `http://127.0.0.1:4000/v1/prometheus/write?db=greptime_metrics`.
 url = ""
 
 ## HTTP headers of Prometheus remote-write carry.

cyborg/bin/bump-doc-version.ts

@@ -1,75 +0,0 @@
-/*
- * Copyright 2023 Greptime Team
- *
- * Licensed under the Apache License, Version 2.0 (the "License");
- * you may not use this file except in compliance with the License.
- * You may obtain a copy of the License at
- *
- *     http://www.apache.org/licenses/LICENSE-2.0
- *
- * Unless required by applicable law or agreed to in writing, software
- * distributed under the License is distributed on an "AS IS" BASIS,
- * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
- * See the License for the specific language governing permissions and
- * limitations under the License.
- */
-
-import * as core from "@actions/core";
-import {obtainClient} from "@/common";
-
-async function triggerWorkflow(workflowId: string, version: string) {
-  const docsClient = obtainClient("DOCS_REPO_TOKEN")
-  try {
-    await docsClient.rest.actions.createWorkflowDispatch({
-      owner: "GreptimeTeam",
-      repo: "docs",
-      workflow_id: workflowId,
-      ref: "main",
-      inputs: {
-        version,
-      },
-    });
-    console.log(`Successfully triggered ${workflowId} workflow with version ${version}`);
-  } catch (error) {
-    core.setFailed(`Failed to trigger workflow: ${error.message}`);
-  }
-}
-
-function determineWorkflow(version: string): [string, string] {
-  // Check if it's a nightly version
-  if (version.includes('nightly')) {
-    return ['bump-nightly-version.yml', version];
-  }
-
-  const parts = version.split('.');
-
-  if (parts.length !== 3) {
-    throw new Error('Invalid version format');
-  }
-
-  // If patch version (last number) is 0, it's a major version
-  // Return only major.minor version
-  if (parts[2] === '0') {
-    return ['bump-version.yml', `${parts[0]}.${parts[1]}`];
-  }
-
-  // Otherwise it's a patch version, use full version
-  return ['bump-patch-version.yml', version];
-}
-
-const version = process.env.VERSION;
-if (!version) {
-  core.setFailed("VERSION environment variable is required");
-  process.exit(1);
-}
-
-// Remove 'v' prefix if exists
-const cleanVersion = version.startsWith('v') ? version.slice(1) : version;
-
-try {
-  const [workflowId, apiVersion] = determineWorkflow(cleanVersion);
-  triggerWorkflow(workflowId, apiVersion);
-} catch (error) {
-  core.setFailed(`Error processing version: ${error.message}`);
-  process.exit(1);
-}

cyborg/bin/bump-versions.ts

@@ -0,0 +1,156 @@
+/*
+ * Copyright 2023 Greptime Team
+ *
+ * Licensed under the Apache License, Version 2.0 (the "License");
+ * you may not use this file except in compliance with the License.
+ * You may obtain a copy of the License at
+ *
+ *     http://www.apache.org/licenses/LICENSE-2.0
+ *
+ * Unless required by applicable law or agreed to in writing, software
+ * distributed under the License is distributed on an "AS IS" BASIS,
+ * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+ * See the License for the specific language governing permissions and
+ * limitations under the License.
+ */
+
+import * as core from "@actions/core";
+import {obtainClient} from "@/common";
+
+interface RepoConfig {
+  tokenEnv: string;
+  repo: string;
+  workflowLogic: (version: string) => [string, string] | null;
+}
+
+const REPO_CONFIGS: Record<string, RepoConfig> = {
+  website: {
+    tokenEnv: "WEBSITE_REPO_TOKEN",
+    repo: "website",
+    workflowLogic: (version: string) => {
+      // Skip nightly versions for website
+      if (version.includes('nightly')) {
+        console.log('Nightly version detected for website, skipping workflow trigger.');
+        return null;
+      }
+      return ['bump-patch-version.yml', version];
+    }
+  },
+  demo: {
+    tokenEnv: "DEMO_REPO_TOKEN",
+    repo: "demo-scene",
+    workflowLogic: (version: string) => {
+      // Skip nightly versions for demo
+      if (version.includes('nightly')) {
+        console.log('Nightly version detected for demo, skipping workflow trigger.');
+        return null;
+      }
+      return ['bump-patch-version.yml', version];
+    }
+  },
+  docs: {
+    tokenEnv: "DOCS_REPO_TOKEN",
+    repo: "docs",
+    workflowLogic: (version: string) => {
+      // Check if it's a nightly version
+      if (version.includes('nightly')) {
+        return ['bump-nightly-version.yml', version];
+      }
+
+      const parts = version.split('.');
+      if (parts.length !== 3) {
+        throw new Error('Invalid version format');
+      }
+
+      // If patch version (last number) is 0, it's a major version
+      // Return only major.minor version
+      if (parts[2] === '0') {
+        return ['bump-version.yml', `${parts[0]}.${parts[1]}`];
+      }
+
+      // Otherwise it's a patch version, use full version
+      return ['bump-patch-version.yml', version];
+    }
+  }
+};
+
+async function triggerWorkflow(repoConfig: RepoConfig, workflowId: string, version: string) {
+  const client = obtainClient(repoConfig.tokenEnv);
+  try {
+    await client.rest.actions.createWorkflowDispatch({
+      owner: "GreptimeTeam",
+      repo: repoConfig.repo,
+      workflow_id: workflowId,
+      ref: "main",
+      inputs: {
+        version,
+      },
+    });
+    console.log(`Successfully triggered ${workflowId} workflow for ${repoConfig.repo} with version ${version}`);
+  } catch (error) {
+    core.setFailed(`Failed to trigger workflow for ${repoConfig.repo}: ${error.message}`);
+    throw error;
+  }
+}
+
+async function processRepo(repoName: string, version: string) {
+  const repoConfig = REPO_CONFIGS[repoName];
+  if (!repoConfig) {
+    throw new Error(`Unknown repository: ${repoName}`);
+  }
+
+  try {
+    const workflowResult = repoConfig.workflowLogic(version);
+    if (workflowResult === null) {
+      // Skip this repo (e.g., nightly version for website)
+      return;
+    }
+
+    const [workflowId, apiVersion] = workflowResult;
+    await triggerWorkflow(repoConfig, workflowId, apiVersion);
+  } catch (error) {
+    core.setFailed(`Error processing ${repoName} with version ${version}: ${error.message}`);
+    throw error;
+  }
+}
+
+async function main() {
+  const version = process.env.VERSION;
+  if (!version) {
+    core.setFailed("VERSION environment variable is required");
+    process.exit(1);
+  }
+
+  // Remove 'v' prefix if exists
+  const cleanVersion = version.startsWith('v') ? version.slice(1) : version;
+
+  // Get target repositories from environment variable
+  // Default to both if not specified
+  const targetRepos = process.env.TARGET_REPOS?.split(',').map(repo => repo.trim()) || ['website', 'docs'];
+
+  console.log(`Processing version ${cleanVersion} for repositories: ${targetRepos.join(', ')}`);
+
+  const errors: string[] = [];
+
+  // Process each repository
+  for (const repo of targetRepos) {
+    try {
+      await processRepo(repo, cleanVersion);
+    } catch (error) {
+      errors.push(`${repo}: ${error.message}`);
+    }
+  }
+
+  if (errors.length > 0) {
+    core.setFailed(`Failed to process some repositories: ${errors.join('; ')}`);
+    process.exit(1);
+  }
+
+  console.log('All repositories processed successfully');
+}
+
+// Execute main function
+main().catch((error) => {
+  core.setFailed(`Unexpected error: ${error.message}`);
+  process.exit(1);
+});

docs/how-to/how-to-change-log-level-on-the-fly.md

@@ -11,6 +11,6 @@ And database will reply with something like:
 Log Level changed from Some("info") to "trace,flow=debug"%
 ```
 
-The data is a string in the format of `global_level,module1=level1,module2=level2,...` that follow the same rule of `RUST_LOG`. 
+The data is a string in the format of `global_level,module1=level1,module2=level2,...` that follows the same rule of `RUST_LOG`. 
 
 The module is the module name of the log, and the level is the log level. The log level can be one of the following: `trace`, `debug`, `info`, `warn`, `error`, `off`(case insensitive).

docs/how-to/how-to-implement-sql-statement.md

@@ -14,7 +14,7 @@ impl SqlQueryHandler for Instance {
 ```
 
 Normally, when a SQL query arrives at GreptimeDB, the `do_query` method will be called. After some parsing work, the SQL
-will be feed into `StatementExecutor`:
+will be fed into `StatementExecutor`:
 
 ```rust
 // in Frontend Instance:
@@ -27,7 +27,7 @@ an example.
 
 Now, what if the statements should be handled differently for GreptimeDB Standalone and Cluster? You can see there's
 a `SqlStatementExecutor` field in `StatementExecutor`. Each GreptimeDB Standalone and Cluster has its own implementation
-of `SqlStatementExecutor`. If you are going to implement the statements differently in the two mode (
+of `SqlStatementExecutor`. If you are going to implement the statements differently in the two modes (
 like `CREATE TABLE`), you have to implement them in their own `SqlStatementExecutor`s.
 
 Summarize as the diagram below:

docs/how-to/how-to-profile-memory.md

@@ -1,6 +1,6 @@
 # Profile memory usage of GreptimeDB
 
-This crate provides an easy approach to dump memory profiling info. A set of ready to use scripts is provided in [docs/how-to/memory-profile-scripts](docs/how-to/memory-profile-scripts).
+This crate provides an easy approach to dump memory profiling info. A set of ready to use scripts is provided in [docs/how-to/memory-profile-scripts](./memory-profile-scripts/scripts).
 
 ## Prerequisites
 ### jemalloc
@@ -44,6 +44,10 @@ Dump memory profiling data through HTTP API:
 
 ```bash
 curl -X POST localhost:4000/debug/prof/mem > greptime.hprof
+# or output flamegraph directly
+curl -X POST "localhost:4000/debug/prof/mem?output=flamegraph" > greptime.svg
+# or output pprof format
+curl -X POST "localhost:4000/debug/prof/mem?output=proto" > greptime.pprof
 ```
 
 You can periodically dump profiling data and compare them to find the delta memory usage.

docs/how-to/how-to-write-aggregate-function.md

@@ -1,8 +1,8 @@
 Currently, our query engine is based on DataFusion, so all aggregate function is executed by DataFusion, through its UDAF interface. You can find DataFusion's UDAF example [here](https://github.com/apache/arrow-datafusion/blob/arrow2/datafusion-examples/examples/simple_udaf.rs). Basically, we provide the same way as DataFusion to write aggregate functions: both are centered in a struct called "Accumulator" to accumulates states along the way in aggregation.
 
-However, DataFusion's UDAF implementation has a huge restriction, that it requires user to provide a concrete "Accumulator". Take `Median` aggregate function for example, to aggregate a `u32` datatype column, you have to write a `MedianU32`, and use `SELECT MEDIANU32(x)` in SQL. `MedianU32` cannot be used to aggregate a `i32` datatype column. Or, there's another way: you can use a special type that can hold all kinds of data (like our `Value` enum or Arrow's `ScalarValue`), and `match` all the way up to do aggregate calculations. It might work, though rather tedious. (But I think it's DataFusion's prefer way to write UDAF.)
+However, DataFusion's UDAF implementation has a huge restriction, that it requires user to provide a concrete "Accumulator". Take `Median` aggregate function for example, to aggregate a `u32` datatype column, you have to write a `MedianU32`, and use `SELECT MEDIANU32(x)` in SQL. `MedianU32` cannot be used to aggregate a `i32` datatype column. Or, there's another way: you can use a special type that can hold all kinds of data (like our `Value` enum or Arrow's `ScalarValue`), and `match` all the way up to do aggregate calculations. It might work, though rather tedious. (But I think it's DataFusion's preferred way to write UDAF.)
 
-So is there a way we can make an aggregate function that automatically match the input data's type? For example, a `Median` aggregator that can work on both `u32` column and `i32`? The answer is yes until we found a way to bypassing DataFusion's restriction, a restriction that DataFusion simply don't pass the input data's type when creating an Accumulator.
+So is there a way we can make an aggregate function that automatically match the input data's type? For example, a `Median` aggregator that can work on both `u32` column and `i32`? The answer is yes until we find a way to bypass DataFusion's restriction, a restriction that DataFusion simply doesn't pass the input data's type when creating an Accumulator.
 
 > There's an example in `my_sum_udaf_example.rs`, take that as quick start.
 
@@ -16,7 +16,7 @@ You must first define a struct that will be used to create your accumulator. For
 struct MySumAccumulatorCreator {}
 ```
 
-Attribute macro `#[as_aggr_func_creator]` and derive macro `#[derive(Debug, AggrFuncTypeStore)]` must both annotated on the struct. They work together to provide a storage of aggregate function's input data types, which are needed for creating generic accumulator later.  
+Attribute macro `#[as_aggr_func_creator]` and derive macro `#[derive(Debug, AggrFuncTypeStore)]` must both be annotated on the struct. They work together to provide a storage of aggregate function's input data types, which are needed for creating generic accumulator later.  
 
 > Note that the `as_aggr_func_creator` macro will add fields to the struct, so the struct cannot be defined as an empty struct without field like `struct Foo;`, neither as a new type like `struct Foo(bar)`.
 
@@ -32,11 +32,11 @@ pub trait AggregateFunctionCreator: Send + Sync + Debug {
 
 You can use input data's type in methods that return output type and state types (just invoke `input_types()`).
 
-The output type is aggregate function's output data's type. For example, `SUM` aggregate function's output type is `u64` for a `u32` datatype column. The state types are accumulator's internal states' types. Take `AVG` aggregate function on a `i32` column as example, it's state types are `i64` (for sum) and `u64` (for count).
+The output type is aggregate function's output data's type. For example, `SUM` aggregate function's output type is `u64` for a `u32` datatype column. The state types are accumulator's internal states' types. Take `AVG` aggregate function on a `i32` column as example, its state types are `i64` (for sum) and `u64` (for count).
 
 The `creator` function is where you define how an accumulator (that will be used in DataFusion) is created. You define "how" to create the accumulator (instead of "what" to create), using the input data's type as arguments. With input datatype known, you can create accumulator generically.
 
-# 2. Impl `Accumulator` trait for you accumulator.
+# 2. Impl `Accumulator` trait for your accumulator.
 
 The accumulator is where you store the aggregate calculation states and evaluate a result. You must impl `Accumulator` trait for it. The trait's definition is:
 
@@ -49,7 +49,7 @@ pub trait Accumulator: Send + Sync + Debug {
 }
 ```
 
-The DataFusion basically execute aggregate like this:
+The DataFusion basically executes aggregate like this:
 
 1. Partitioning all input data for aggregate. Create an accumulator for each part.
 2. Call `update_batch` on each accumulator with partitioned data, to let you update your aggregate calculation.
@@ -57,16 +57,16 @@ The DataFusion basically execute aggregate like this:
 4. Call `merge_batch` to merge all accumulator's internal state to one.
 5. Execute `evaluate` on the chosen one to get the final calculation result.
 
-Once you know the meaning of each method, you can easily write your accumulator. You can refer to `Median` accumulator or `SUM` accumulator defined in  file `my_sum_udaf_example.rs` for more details.
+Once you know the meaning of each method, you can easily write your accumulator. You can refer to `Median` accumulator or `SUM` accumulator defined in file `my_sum_udaf_example.rs` for more details.
 
 # 3. Register your aggregate function to our query engine.
 
 You can call `register_aggregate_function` method in query engine to register your aggregate function. To do that, you have to new an instance of struct `AggregateFunctionMeta`. The struct has three fields, first is the name of your aggregate function's name. The function name is case-sensitive due to DataFusion's restriction. We strongly recommend using lowercase for your name. If you have to use uppercase name, wrap your aggregate function with quotation marks. For example, if you define an aggregate function named "my_aggr", you can use "`SELECT MY_AGGR(x)`"; if you define "my_AGGR", you have to use "`SELECT "my_AGGR"(x)`".
 
-The second field is arg_counts ,the count of the arguments. Like accumulator `percentile`, calculating the p_number of the column. We need to input the value of column and the value of p to cacalate, and so the count of the arguments is two.
+The second field is arg_counts ,the count of the arguments. Like accumulator `percentile`, calculating the p_number of the column. We need to input the value of column and the value of p to calculate, and so the count of the arguments is two.
 
 The third field is a function about how to create your accumulator creator that you defined in step 1 above. Create creator, that's a bit intertwined, but it is how we make DataFusion use a newly created aggregate function each time it executes a SQL, preventing the stored input types from affecting each other. The key detail can be starting looking at our `DfContextProviderAdapter` struct's `get_aggregate_meta` method.
 
 # (Optional) 4. Make your aggregate function automatically registered.
 
-If you've written a great aggregate function that want to let everyone use it, you can make it automatically registered to our query engine at start time. It's quick simple, just refer to the `AggregateFunctions::register` function in `common/function/src/scalars/aggregate/mod.rs`.
+If you've written a great aggregate function that wants to let everyone use it, you can make it automatically register to our query engine at start time. It's quick and simple, just refer to the `AggregateFunctions::register` function in `common/function/src/scalars/aggregate/mod.rs`.

docs/how-to/how-to-write-fuzz-tests.md

@@ -3,7 +3,7 @@
 This document introduces how to write fuzz tests in GreptimeDB.
 
 ## What is a fuzz test
-Fuzz test is tool that leverage deterministic random generation to assist in finding bugs. The goal of fuzz tests is to identify inputs generated by the fuzzer that cause system panics, crashes, or unexpected behaviors to occur. And we are using the [cargo-fuzz](https://github.com/rust-fuzz/cargo-fuzz) to run our fuzz test targets.
+Fuzz test is tool that leverages deterministic random generation to assist in finding bugs. The goal of fuzz tests is to identify inputs generated by the fuzzer that cause system panics, crashes, or unexpected behaviors to occur. And we are using the [cargo-fuzz](https://github.com/rust-fuzz/cargo-fuzz) to run our fuzz test targets.
 
 ## Why we need them
 - Find bugs by leveraging random generation

flake.lock

@@ -8,11 +8,11 @@
         "rust-analyzer-src": "rust-analyzer-src"
       },
       "locked": {
-        "lastModified": 1737613896,
-        "narHash": "sha256-ldqXIglq74C7yKMFUzrS9xMT/EVs26vZpOD68Sh7OcU=",
+        "lastModified": 1745735608,
+        "narHash": "sha256-L0jzm815XBFfF2wCFmR+M1CF+beIEFj6SxlqVKF59Ec=",
         "owner": "nix-community",
         "repo": "fenix",
-        "rev": "303a062fdd8e89f233db05868468975d17855d80",
+        "rev": "c39a78eba6ed2a022cc3218db90d485077101496",
         "type": "github"
       },
       "original": {
@@ -41,16 +41,16 @@
     },
     "nixpkgs": {
       "locked": {
-        "lastModified": 1737569578,
-        "narHash": "sha256-6qY0pk2QmUtBT9Mywdvif0i/CLVgpCjMUn6g9vB+f3M=",
+        "lastModified": 1748162331,
+        "narHash": "sha256-rqc2RKYTxP3tbjA+PB3VMRQNnjesrT0pEofXQTrMsS8=",
         "owner": "NixOS",
         "repo": "nixpkgs",
-        "rev": "47addd76727f42d351590c905d9d1905ca895b82",
+        "rev": "7c43f080a7f28b2774f3b3f43234ca11661bf334",
         "type": "github"
       },
       "original": {
         "owner": "NixOS",
-        "ref": "nixos-24.11",
+        "ref": "nixos-25.05",
         "repo": "nixpkgs",
         "type": "github"
       }
@@ -65,11 +65,11 @@
     "rust-analyzer-src": {
       "flake": false,
       "locked": {
-        "lastModified": 1737581772,
-        "narHash": "sha256-t1P2Pe3FAX9TlJsCZbmJ3wn+C4qr6aSMypAOu8WNsN0=",
+        "lastModified": 1745694049,
+        "narHash": "sha256-fxvRYH/tS7hGQeg9zCVh5RBcSWT+JGJet7RA8Ss+rC0=",
         "owner": "rust-lang",
         "repo": "rust-analyzer",
-        "rev": "582af7ee9c8d84f5d534272fc7de9f292bd849be",
+        "rev": "d8887c0758bbd2d5f752d5bd405d4491e90e7ed6",
         "type": "github"
       },
       "original": {

flake.nix

@@ -2,7 +2,7 @@
   description = "Development environment flake";
 
   inputs = {
-    nixpkgs.url = "github:NixOS/nixpkgs/nixos-24.11";
+    nixpkgs.url = "github:NixOS/nixpkgs/nixos-25.05";
     fenix = {
       url = "github:nix-community/fenix";
       inputs.nixpkgs.follows = "nixpkgs";
@@ -21,7 +21,7 @@
         lib = nixpkgs.lib;
         rustToolchain = fenix.packages.${system}.fromToolchainName {
           name = (lib.importTOML ./rust-toolchain.toml).toolchain.channel;
-          sha256 = "sha256-f/CVA1EC61EWbh0SjaRNhLL0Ypx2ObupbzigZp8NmL4=";
+          sha256 = "sha256-tJJr8oqX3YD+ohhPK7jlt/7kvKBnBqJVjYtoFr520d4=";
         };
       in
       {
@@ -51,6 +51,7 @@
           ];
 
           LD_LIBRARY_PATH = pkgs.lib.makeLibraryPath buildInputs;
+          NIX_HARDENING_ENABLE = "";
         };
       });
 }

grafana/README.md

@@ -1,61 +1,122 @@
-Grafana dashboard for GreptimeDB
---------------------------------
+# Grafana dashboards for GreptimeDB
 
-GreptimeDB's official Grafana dashboard.
+## Overview
 
-Status notify: we are still working on this config. It's expected to change frequently in the recent days. Please feel free to submit your feedback and/or contribution to this dashboard 🤗
+This repository contains Grafana dashboards for visualizing metrics and logs of GreptimeDB instances running in either cluster or standalone mode. **The Grafana version should be greater than 9.0**.
 
-If you use Helm [chart](https://github.com/GreptimeTeam/helm-charts) to deploy GreptimeDB cluster, you can enable self-monitoring by setting the following values in your Helm chart:
+We highly recommend using the self-monitoring feature provided by [GreptimeDB Operator](https://github.com/GrepTimeTeam/greptimedb-operator) to automatically collect metrics and logs from your GreptimeDB instances and store them in a dedicated GreptimeDB instance.
+
+- **Metrics Dashboards**
+
+  - `dashboards/metrics/cluster/dashboard.json`: The Grafana dashboard for the GreptimeDB cluster. Read the [dashboard.md](./dashboards/metrics/cluster/dashboard.md) for more details.
+
+  - `dashboards/metrics/standalone/dashboard.json`: The Grafana dashboard for the standalone GreptimeDB instance. **It's generated from the `cluster/dashboard.json` by removing the instance filter through the `make dashboards` command**. Read the [dashboard.md](./dashboards/metrics/standalone/dashboard.md) for more details.
+
+- **Logs Dashboard**
+
+  The `dashboards/logs/dashboard.json` provides a comprehensive Grafana dashboard for visualizing GreptimeDB logs. To utilize this dashboard effectively, you need to collect logs in JSON format from your GreptimeDB instances and store them in a dedicated GreptimeDB instance.
+
+  For proper integration, the logs table must adhere to the following schema design with the table name `_gt_logs`:
+
+  ```sql
+  CREATE TABLE IF NOT EXISTS `_gt_logs` (
+    `pod_ip` STRING NULL,
+    `namespace` STRING NULL,
+    `cluster` STRING NULL,
+    `file` STRING NULL,
+    `module_path` STRING NULL,
+    `level` STRING NULL,
+    `target` STRING NULL,
+    `role` STRING NULL,
+    `pod` STRING NULL SKIPPING INDEX WITH(granularity = '10240', type = 'BLOOM'),
+    `message` STRING NULL FULLTEXT INDEX WITH(analyzer = 'English', backend = 'bloom', case_sensitive = 'false'),
+    `err` STRING NULL FULLTEXT INDEX WITH(analyzer = 'English', backend = 'bloom', case_sensitive = 'false'),
+    `timestamp` TIMESTAMP(9) NOT NULL,
+    TIME INDEX (`timestamp`),
+    PRIMARY KEY (`level`, `target`, `role`)
+  )
+    ENGINE=mito
+  WITH (
+    append_mode = 'true'
+  )
+  ```
+
+## Development
+
+As GreptimeDB evolves rapidly, metrics may change over time. We welcome your feedback and contributions to improve these dashboards 🤗
+
+To modify the metrics dashboards, simply edit the `dashboards/metrics/cluster/dashboard.json` file and run the `make dashboards` command. This will automatically generate the updated `dashboards/metrics/standalone/dashboard.json` and other related files.
+
+For easier dashboard maintenance, we utilize the [`dac`](https://github.com/zyy17/dac) tool to generate human-readable intermediate dashboards and documentation:
+
+- `dashboards/metrics/cluster/dashboard.yaml`: The intermediate dashboard file for the GreptimeDB cluster.
+- `dashboards/metrics/standalone/dashboard.yaml`: The intermediate dashboard file for standalone GreptimeDB instances.
+
+## Data Sources
+
+The following data sources are used to fetch metrics and logs:
+
+- **`${metrics}`**: Prometheus data source for providing the GreptimeDB metrics.
+- **`${logs}`**: MySQL data source for providing the GreptimeDB logs.
+- **`${information_schema}`**: MySQL data source for providing the information schema of the current instance and used for the `overview` panel. It is the MySQL port of the current monitored instance.
+
+## Instance Filters
+
+To deploy the dashboards for multiple scenarios (K8s, bare metal, etc.), we prefer to use the `instance` label when filtering instances.
+
+Additionally, we recommend including the `pod` label in the legend to make it easier to identify each instance, even though this field will be empty in bare metal scenarios.
+
+For example, the following query is recommended:
+
+```promql
+sum(process_resident_memory_bytes{instance=~"$datanode"}) by (instance, pod)
+```
+
+And the legend will be like: `[{{instance}}]-[{{ pod }}]`.
+
+## Deployment
+
+### (Recommended) Helm Chart
+
+If you use the [Helm Chart](https://github.com/GreptimeTeam/helm-charts) to deploy a GreptimeDB cluster, you can enable self-monitoring by setting the following values in your Helm chart:
 
 - `monitoring.enabled=true`: Deploys a standalone GreptimeDB instance dedicated to monitoring the cluster;
 - `grafana.enabled=true`: Deploys Grafana and automatically imports the monitoring dashboard;
 
-The standalone GreptimeDB instance will collect metrics from your cluster and the dashboard will be available in the Grafana UI. For detailed deployment instructions, please refer to our [Kubernetes deployment guide](https://docs.greptime.com/nightly/user-guide/deployments/deploy-on-kubernetes/getting-started).
+The standalone GreptimeDB instance will collect metrics from your cluster, and the dashboard will be available in the Grafana UI. For detailed deployment instructions, please refer to our [Kubernetes deployment guide](https://docs.greptime.com/user-guide/deployments-administration/deploy-on-kubernetes/getting-started).
 
-# How to use
+### Self-host Prometheus and import dashboards manually
 
-## `greptimedb.json`
+1. **Configure Prometheus to scrape the cluster**
 
-Open Grafana Dashboard page, choose `New` -> `Import`. And upload `greptimedb.json` file.
+   The following is an example configuration(**Please modify it according to your actual situation**):
 
-## `greptimedb-cluster.json`
+    ```yml
+    # example config
+    # only to indicate how to assign labels to each target
+    # modify yours accordingly
+    scrape_configs:
+      - job_name: metasrv
+        static_configs:
+        - targets: ['<metasrv-ip>:<port>']
 
-This cluster dashboard provides a comprehensive view of incoming requests, response statuses, and internal activities such as flush and compaction, with a layered structure from frontend to datanode. Designed with a focus on alert functionality, its primary aim is to highlight any anomalies in metrics, allowing users to quickly pinpoint the cause of errors.
+      - job_name: datanode
+        static_configs:
+        - targets: ['<datanode0-ip>:<port>', '<datanode1-ip>:<port>', '<datanode2-ip>:<port>']
 
-We use Prometheus to scrape off metrics from nodes in GreptimeDB cluster, Grafana to visualize the diagram. Any compatible stack should work too.
+      - job_name: frontend
+        static_configs:
+        - targets: ['<frontend-ip>:<port>']
+    ```
 
-__Note__: This dashboard is still in an early stage of development. Any issue or advice on improvement is welcomed.
+2. **Configure the data sources in Grafana**
 
-### Configuration
+   You need to add two data sources in Grafana:
 
-Please ensure the following configuration before importing the dashboard into Grafana.
+   - Prometheus: It is the Prometheus instance that scrapes the GreptimeDB metrics.
+   - Information Schema: It is the MySQL port of the current monitored instance. The dashboard will use this datasource to show the information schema of the current instance.
 
-__1. Prometheus scrape config__
+3. **Import the dashboards based on your deployment scenario**
 
-Configure Prometheus to scrape the cluster.
-
-```yml
-# example config
-# only to indicate how to assign labels to each target
-# modify yours accordingly
-scrape_configs:
-  - job_name: metasrv
-    static_configs:
-    - targets: ['<metasrv-ip>:<port>']
-
-  - job_name: datanode
-    static_configs:
-    - targets: ['<datanode0-ip>:<port>', '<datanode1-ip>:<port>', '<datanode2-ip>:<port>']
-
-  - job_name: frontend
-    static_configs:
-    - targets: ['<frontend-ip>:<port>']
-```
-
-__2. Grafana config__
-
-Create a Prometheus data source in Grafana before using this dashboard. We use `datasource` as a variable in Grafana dashboard so that multiple environments are supported.
-
-### Usage
-
-Use `datasource` or `instance` on the upper-left corner to filter data from certain node.
+   - **Cluster**: Import the `dashboards/metrics/cluster/dashboard.json` dashboard.
+   - **Standalone**: Import the `dashboards/metrics/standalone/dashboard.json` dashboard.

grafana/check.sh

@@ -1,19 +0,0 @@
-#!/usr/bin/env bash
-
-BASEDIR=$(dirname "$0")
-
-# Use jq to check for panels with empty or missing descriptions
-invalid_panels=$(cat $BASEDIR/greptimedb-cluster.json | jq -r '
-  .panels[]
-  | select((.type == "stats" or .type == "timeseries") and (.description == "" or .description == null))
-')
-
-# Check if any invalid panels were found
-if [[ -n "$invalid_panels" ]]; then
-  echo "Error: The following panels have empty or missing descriptions:"
-  echo "$invalid_panels"
-  exit 1
-else
-  echo "All panels with type 'stats' or 'timeseries' have valid descriptions."
-  exit 0
-fi

grafana/dashboards/logs/dashboard.json

@@ -0,0 +1,292 @@
+{
+  "annotations": {
+    "list": [
+      {
+        "builtIn": 1,
+        "datasource": {
+          "type": "grafana",
+          "uid": "-- Grafana --"
+        },
+        "enable": true,
+        "hide": true,
+        "iconColor": "rgba(0, 211, 255, 1)",
+        "name": "Annotations & Alerts",
+        "type": "dashboard"
+      }
+    ]
+  },
+  "editable": true,
+  "fiscalYearStartMonth": 0,
+  "graphTooltip": 0,
+  "id": 12,
+  "links": [],
+  "panels": [
+    {
+      "datasource": {
+        "default": false,
+        "type": "mysql",
+        "uid": "${datasource}"
+      },
+      "fieldConfig": {
+        "defaults": {},
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 20,
+        "w": 24,
+        "x": 0,
+        "y": 0
+      },
+      "id": 1,
+      "options": {
+        "dedupStrategy": "none",
+        "enableInfiniteScrolling": true,
+        "enableLogDetails": true,
+        "prettifyLogMessage": false,
+        "showCommonLabels": false,
+        "showLabels": false,
+        "showTime": true,
+        "sortOrder": "Descending",
+        "wrapLogMessage": false
+      },
+      "pluginVersion": "11.6.0",
+      "targets": [
+        {
+          "dataset": "greptime_private",
+          "datasource": {
+            "type": "mysql",
+            "uid": "${datasource}"
+          },
+          "editorMode": "code",
+          "format": "table",
+          "rawQuery": true,
+          "rawSql": "SELECT `timestamp`, CONCAT('[', `level`, ']', ' ', '<', `target`, '>', ' ', `message`),\n  `role`,\n  `pod`,\n  `pod_ip`,\n  `namespace`,\n  `cluster`,\n  `err`,\n  `file`,\n  `module_path`\nFROM\n  `_gt_logs`\nWHERE\n  (\n    \"$level\" = \"'all'\"\n    OR `level` IN ($level)\n  ) \n  AND (\n    \"$role\" = \"'all'\"\n    OR `role` IN ($role)\n  )\n  AND (\n    \"$pod\" = \"\"\n    OR `pod` = '$pod'\n  )\n  AND (\n    \"$target\" = \"\"\n    OR `target` = '$target'\n  )\n  AND (\n    \"$search\" = \"\"\n    OR matches_term(`message`, '$search')\n  )\n  AND (\n    \"$exclude\" = \"\"\n    OR NOT matches_term(`message`, '$exclude')\n  )\n  AND $__timeFilter(`timestamp`)\nORDER BY `timestamp` DESC\nLIMIT $limit;\n",
+          "refId": "A",
+          "sql": {
+            "columns": [
+              {
+                "parameters": [],
+                "type": "function"
+              }
+            ],
+            "groupBy": [
+              {
+                "property": {
+                  "type": "string"
+                },
+                "type": "groupBy"
+              }
+            ],
+            "limit": 50
+          }
+        }
+      ],
+      "title": "Logs",
+      "type": "logs"
+    }
+  ],
+  "preload": false,
+  "refresh": "",
+  "schemaVersion": 41,
+  "tags": [],
+  "templating": {
+    "list": [
+      {
+        "current": {
+          "text": "logs",
+          "value": "P98F38F12DB221A8C"
+        },
+        "includeAll": false,
+        "name": "datasource",
+        "options": [],
+        "query": "mysql",
+        "refresh": 1,
+        "regex": "",
+        "type": "datasource"
+      },
+      {
+        "allValue": "'all'",
+        "current": {
+          "text": [
+            "$__all"
+          ],
+          "value": [
+            "$__all"
+          ]
+        },
+        "includeAll": true,
+        "label": "level",
+        "multi": true,
+        "name": "level",
+        "options": [
+          {
+            "selected": false,
+            "text": "INFO",
+            "value": "INFO"
+          },
+          {
+            "selected": false,
+            "text": "ERROR",
+            "value": "ERROR"
+          },
+          {
+            "selected": false,
+            "text": "WARN",
+            "value": "WARN"
+          },
+          {
+            "selected": false,
+            "text": "DEBUG",
+            "value": "DEBUG"
+          },
+          {
+            "selected": false,
+            "text": "TRACE",
+            "value": "TRACE"
+          }
+        ],
+        "query": "INFO,ERROR,WARN,DEBUG,TRACE",
+        "type": "custom"
+      },
+      {
+        "allValue": "'all'",
+        "current": {
+          "text": [
+            "$__all"
+          ],
+          "value": [
+            "$__all"
+          ]
+        },
+        "includeAll": true,
+        "label": "role",
+        "multi": true,
+        "name": "role",
+        "options": [
+          {
+            "selected": false,
+            "text": "datanode",
+            "value": "datanode"
+          },
+          {
+            "selected": false,
+            "text": "frontend",
+            "value": "frontend"
+          },
+          {
+            "selected": false,
+            "text": "meta",
+            "value": "meta"
+          }
+        ],
+        "query": "datanode,frontend,meta",
+        "type": "custom"
+      },
+      {
+        "current": {
+          "text": "",
+          "value": ""
+        },
+        "label": "pod",
+        "name": "pod",
+        "options": [
+          {
+            "selected": true,
+            "text": "",
+            "value": ""
+          }
+        ],
+        "query": "",
+        "type": "textbox"
+      },
+      {
+        "current": {
+          "text": "",
+          "value": ""
+        },
+        "label": "target",
+        "name": "target",
+        "options": [
+          {
+            "selected": true,
+            "text": "",
+            "value": ""
+          }
+        ],
+        "query": "",
+        "type": "textbox"
+      },
+      {
+        "current": {
+          "text": "",
+          "value": ""
+        },
+        "label": "search",
+        "name": "search",
+        "options": [
+          {
+            "selected": true,
+            "text": "",
+            "value": ""
+          }
+        ],
+        "query": "",
+        "type": "textbox"
+      },
+      {
+        "current": {
+          "text": "",
+          "value": ""
+        },
+        "label": "exclude",
+        "name": "exclude",
+        "options": [
+          {
+            "selected": true,
+            "text": "",
+            "value": ""
+          }
+        ],
+        "query": "",
+        "type": "textbox"
+      },
+      {
+        "current": {
+          "text": "2000",
+          "value": "2000"
+        },
+        "includeAll": false,
+        "label": "limit",
+        "name": "limit",
+        "options": [
+          {
+            "selected": true,
+            "text": "2000",
+            "value": "2000"
+          },
+          {
+            "selected": false,
+            "text": "5000",
+            "value": "5000"
+          },
+          {
+            "selected": false,
+            "text": "8000",
+            "value": "8000"
+          }
+        ],
+        "query": "2000,5000,8000",
+        "type": "custom"
+      }
+    ]
+  },
+  "time": {
+    "from": "now-6h",
+    "to": "now"
+  },
+  "timepicker": {},
+  "timezone": "browser",
+  "title": "GreptimeDB Logs",
+  "uid": "edx5veo4rd3wge2",
+  "version": 1
+}

grafana/dashboards/metrics/cluster/dashboard.md

@@ -0,0 +1,112 @@
+# Overview
+| Title | Query | Type | Description | Datasource | Unit | Legend Format |
+| --- | --- | --- | --- | --- | --- | --- |
+| Uptime | `time() - process_start_time_seconds` | `stat` | The start time of GreptimeDB. | `prometheus` | `s` | `__auto` |
+| Version | `SELECT pkg_version FROM information_schema.build_info` | `stat` | GreptimeDB version. | `mysql` | -- | -- |
+| Total Ingestion Rate | `sum(rate(greptime_table_operator_ingest_rows[$__rate_interval]))` | `stat` | Total ingestion rate. | `prometheus` | `rowsps` | `__auto` |
+| Total Storage Size | `select SUM(disk_size) from information_schema.region_statistics;` | `stat` | Total number of data file size. | `mysql` | `decbytes` | -- |
+| Total Rows | `select SUM(region_rows) from information_schema.region_statistics;` | `stat` | Total number of data rows in the cluster. Calculated by sum of rows from each region. | `mysql` | `sishort` | -- |
+| Deployment | `SELECT count(*) as datanode FROM information_schema.cluster_info WHERE peer_type = 'DATANODE';`<br/>`SELECT count(*) as frontend FROM information_schema.cluster_info WHERE peer_type = 'FRONTEND';`<br/>`SELECT count(*) as metasrv FROM information_schema.cluster_info WHERE peer_type = 'METASRV';`<br/>`SELECT count(*) as flownode FROM information_schema.cluster_info WHERE peer_type = 'FLOWNODE';` | `stat` | The deployment topology of GreptimeDB. | `mysql` | -- | -- |
+| Database Resources | `SELECT COUNT(*) as databases FROM information_schema.schemata WHERE schema_name NOT IN ('greptime_private', 'information_schema')`<br/>`SELECT COUNT(*) as tables FROM information_schema.tables WHERE table_schema != 'information_schema'`<br/>`SELECT COUNT(region_id) as regions FROM information_schema.region_peers`<br/>`SELECT COUNT(*) as flows FROM information_schema.flows` | `stat` | The number of the key resources in GreptimeDB. | `mysql` | -- | -- |
+| Data Size | `SELECT SUM(memtable_size) * 0.42825 as WAL FROM information_schema.region_statistics;`<br/>`SELECT SUM(index_size) as index FROM information_schema.region_statistics;`<br/>`SELECT SUM(manifest_size) as manifest FROM information_schema.region_statistics;` | `stat` | The data size of wal/index/manifest in the GreptimeDB. | `mysql` | `decbytes` | -- |
+# Ingestion
+| Title | Query | Type | Description | Datasource | Unit | Legend Format |
+| --- | --- | --- | --- | --- | --- | --- |
+| Total Ingestion Rate | `sum(rate(greptime_table_operator_ingest_rows{instance=~"$frontend"}[$__rate_interval]))` | `timeseries` | Total ingestion rate.<br/><br/>Here we listed 3 primary protocols:<br/><br/>- Prometheus remote write<br/>- Greptime's gRPC API (when using our ingest SDK)<br/>- Log ingestion http API<br/> | `prometheus` | `rowsps` | `ingestion` |
+| Ingestion Rate by Type | `sum(rate(greptime_servers_http_logs_ingestion_counter[$__rate_interval]))`<br/>`sum(rate(greptime_servers_prometheus_remote_write_samples[$__rate_interval]))` | `timeseries` | Total ingestion rate.<br/><br/>Here we listed 3 primary protocols:<br/><br/>- Prometheus remote write<br/>- Greptime's gRPC API (when using our ingest SDK)<br/>- Log ingestion http API<br/> | `prometheus` | `rowsps` | `http-logs` |
+# Queries
+| Title | Query | Type | Description | Datasource | Unit | Legend Format |
+| --- | --- | --- | --- | --- | --- | --- |
+| Total Query Rate | `sum (rate(greptime_servers_mysql_query_elapsed_count{instance=~"$frontend"}[$__rate_interval]))`<br/>`sum (rate(greptime_servers_postgres_query_elapsed_count{instance=~"$frontend"}[$__rate_interval]))`<br/>`sum (rate(greptime_servers_http_promql_elapsed_counte{instance=~"$frontend"}[$__rate_interval]))` | `timeseries` | Total rate of query API calls by protocol. This metric is collected from frontends.<br/><br/>Here we listed 3 main protocols:<br/>- MySQL<br/>- Postgres<br/>- Prometheus API<br/><br/>Note that there are some other minor query APIs like /sql are not included | `prometheus` | `reqps` | `mysql` |
+# Resources
+| Title | Query | Type | Description | Datasource | Unit | Legend Format |
+| --- | --- | --- | --- | --- | --- | --- |
+| Datanode Memory per Instance | `sum(process_resident_memory_bytes{instance=~"$datanode"}) by (instance, pod)` | `timeseries` | Current memory usage by instance | `prometheus` | `decbytes` | `[{{instance}}]-[{{ pod }}]` |
+| Datanode CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{instance=~"$datanode"}[$__rate_interval]) * 1000) by (instance, pod)` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]` |
+| Frontend Memory per Instance | `sum(process_resident_memory_bytes{instance=~"$frontend"}) by (instance, pod)` | `timeseries` | Current memory usage by instance | `prometheus` | `decbytes` | `[{{ instance }}]-[{{ pod }}]` |
+| Frontend CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{instance=~"$frontend"}[$__rate_interval]) * 1000) by (instance, pod)` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]-cpu` |
+| Metasrv Memory per Instance | `sum(process_resident_memory_bytes{instance=~"$metasrv"}) by (instance, pod)` | `timeseries` | Current memory usage by instance | `prometheus` | `decbytes` | `[{{ instance }}]-[{{ pod }}]-resident` |
+| Metasrv CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{instance=~"$metasrv"}[$__rate_interval]) * 1000) by (instance, pod)` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]` |
+| Flownode Memory per Instance | `sum(process_resident_memory_bytes{instance=~"$flownode"}) by (instance, pod)` | `timeseries` | Current memory usage by instance | `prometheus` | `decbytes` | `[{{ instance }}]-[{{ pod }}]` |
+| Flownode CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{instance=~"$flownode"}[$__rate_interval]) * 1000) by (instance, pod)` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]` |
+# Frontend Requests
+| Title | Query | Type | Description | Datasource | Unit | Legend Format |
+| --- | --- | --- | --- | --- | --- | --- |
+| HTTP QPS per Instance | `sum by(instance, pod, path, method, code) (rate(greptime_servers_http_requests_elapsed_count{instance=~"$frontend",path!~"/health\|/metrics"}[$__rate_interval]))` | `timeseries` | HTTP QPS per Instance. | `prometheus` | `reqps` | `[{{instance}}]-[{{pod}}]-[{{path}}]-[{{method}}]-[{{code}}]` |
+| HTTP P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, path, method, code) (rate(greptime_servers_http_requests_elapsed_bucket{instance=~"$frontend",path!~"/health\|/metrics"}[$__rate_interval])))` | `timeseries` | HTTP P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{path}}]-[{{method}}]-[{{code}}]-p99` |
+| gRPC QPS per Instance | `sum by(instance, pod, path, code) (rate(greptime_servers_grpc_requests_elapsed_count{instance=~"$frontend"}[$__rate_interval]))` | `timeseries` | gRPC QPS per Instance. | `prometheus` | `reqps` | `[{{instance}}]-[{{pod}}]-[{{path}}]-[{{code}}]` |
+| gRPC P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, path, code) (rate(greptime_servers_grpc_requests_elapsed_bucket{instance=~"$frontend"}[$__rate_interval])))` | `timeseries` | gRPC P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{path}}]-[{{method}}]-[{{code}}]-p99` |
+| MySQL QPS per Instance | `sum by(pod, instance)(rate(greptime_servers_mysql_query_elapsed_count{instance=~"$frontend"}[$__rate_interval]))` | `timeseries` | MySQL QPS per Instance. | `prometheus` | `reqps` | `[{{instance}}]-[{{pod}}]` |
+| MySQL P99 per Instance | `histogram_quantile(0.99, sum by(pod, instance, le) (rate(greptime_servers_mysql_query_elapsed_bucket{instance=~"$frontend"}[$__rate_interval])))` | `timeseries` | MySQL P99 per Instance. | `prometheus` | `s` | `[{{ instance }}]-[{{ pod }}]-p99` |
+| PostgreSQL QPS per Instance | `sum by(pod, instance)(rate(greptime_servers_postgres_query_elapsed_count{instance=~"$frontend"}[$__rate_interval]))` | `timeseries` | PostgreSQL QPS per Instance. | `prometheus` | `reqps` | `[{{instance}}]-[{{pod}}]` |
+| PostgreSQL P99 per Instance | `histogram_quantile(0.99, sum by(pod,instance,le) (rate(greptime_servers_postgres_query_elapsed_bucket{instance=~"$frontend"}[$__rate_interval])))` | `timeseries` | PostgreSQL P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-p99` |
+# Frontend to Datanode
+| Title | Query | Type | Description | Datasource | Unit | Legend Format |
+| --- | --- | --- | --- | --- | --- | --- |
+| Ingest Rows per Instance | `sum by(instance, pod)(rate(greptime_table_operator_ingest_rows{instance=~"$frontend"}[$__rate_interval]))` | `timeseries` | Ingestion rate by row as in each frontend | `prometheus` | `rowsps` | `[{{instance}}]-[{{pod}}]` |
+| Region Call QPS per Instance | `sum by(instance, pod, request_type) (rate(greptime_grpc_region_request_count{instance=~"$frontend"}[$__rate_interval]))` | `timeseries` | Region Call QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{request_type}}]` |
+| Region Call P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, request_type) (rate(greptime_grpc_region_request_bucket{instance=~"$frontend"}[$__rate_interval])))` | `timeseries` | Region Call P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{request_type}}]` |
+| Frontend Handle Bulk Insert Elapsed Time  | `sum by(instance, pod, stage) (rate(greptime_table_operator_handle_bulk_insert_sum[$__rate_interval]))/sum by(instance, pod, stage) (rate(greptime_table_operator_handle_bulk_insert_count[$__rate_interval]))`<br/>`histogram_quantile(0.99, sum by(instance, pod, stage, le) (rate(greptime_table_operator_handle_bulk_insert_bucket[$__rate_interval])))` | `timeseries` | Per-stage time for frontend to handle bulk insert requests | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{stage}}]-AVG` |
+# Mito Engine
+| Title | Query | Type | Description | Datasource | Unit | Legend Format |
+| --- | --- | --- | --- | --- | --- | --- |
+| Request OPS per Instance | `sum by(instance, pod, type) (rate(greptime_mito_handle_request_elapsed_count{instance=~"$datanode"}[$__rate_interval]))` | `timeseries` | Request QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{type}}]` |
+| Request P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, type) (rate(greptime_mito_handle_request_elapsed_bucket{instance=~"$datanode"}[$__rate_interval])))` | `timeseries` | Request P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{type}}]` |
+| Write Buffer per Instance | `greptime_mito_write_buffer_bytes{instance=~"$datanode"}` | `timeseries` | Write Buffer per Instance. | `prometheus` | `decbytes` | `[{{instance}}]-[{{pod}}]` |
+| Write Rows per Instance | `sum by (instance, pod) (rate(greptime_mito_write_rows_total{instance=~"$datanode"}[$__rate_interval]))` | `timeseries` | Ingestion size by row counts. | `prometheus` | `rowsps` | `[{{instance}}]-[{{pod}}]` |
+| Flush OPS per Instance | `sum by(instance, pod, reason) (rate(greptime_mito_flush_requests_total{instance=~"$datanode"}[$__rate_interval]))` | `timeseries` | Flush QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{reason}}]` |
+| Write Stall per Instance | `sum by(instance, pod) (greptime_mito_write_stall_total{instance=~"$datanode"})` | `timeseries` | Write Stall per Instance. | `prometheus` | -- | `[{{instance}}]-[{{pod}}]` |
+| Read Stage OPS per Instance | `sum by(instance, pod) (rate(greptime_mito_read_stage_elapsed_count{instance=~"$datanode", stage="total"}[$__rate_interval]))` | `timeseries` | Read Stage OPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]` |
+| Read Stage P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, stage) (rate(greptime_mito_read_stage_elapsed_bucket{instance=~"$datanode"}[$__rate_interval])))` | `timeseries` | Read Stage P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{stage}}]` |
+| Write Stage P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, stage) (rate(greptime_mito_write_stage_elapsed_bucket{instance=~"$datanode"}[$__rate_interval])))` | `timeseries` | Write Stage P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{stage}}]` |
+| Compaction OPS per Instance | `sum by(instance, pod) (rate(greptime_mito_compaction_total_elapsed_count{instance=~"$datanode"}[$__rate_interval]))` | `timeseries` | Compaction OPS per Instance. | `prometheus` | `ops` | `[{{ instance }}]-[{{pod}}]` |
+| Compaction Elapsed Time per Instance by Stage | `histogram_quantile(0.99, sum by(instance, pod, le, stage) (rate(greptime_mito_compaction_stage_elapsed_bucket{instance=~"$datanode"}[$__rate_interval])))`<br/>`sum by(instance, pod, stage) (rate(greptime_mito_compaction_stage_elapsed_sum{instance=~"$datanode"}[$__rate_interval]))/sum by(instance, pod, stage) (rate(greptime_mito_compaction_stage_elapsed_count{instance=~"$datanode"}[$__rate_interval]))` | `timeseries` | Compaction latency by stage | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{stage}}]-p99` |
+| Compaction P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le,stage) (rate(greptime_mito_compaction_total_elapsed_bucket{instance=~"$datanode"}[$__rate_interval])))` | `timeseries` | Compaction P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{stage}}]-compaction` |
+| WAL write size | `histogram_quantile(0.95, sum by(le,instance, pod) (rate(raft_engine_write_size_bucket[$__rate_interval])))`<br/>`histogram_quantile(0.99, sum by(le,instance,pod) (rate(raft_engine_write_size_bucket[$__rate_interval])))`<br/>`sum by (instance, pod)(rate(raft_engine_write_size_sum[$__rate_interval]))` | `timeseries` | Write-ahead logs write size as bytes. This chart includes stats of p95 and p99 size by instance, total WAL write rate. | `prometheus` | `bytes` | `[{{instance}}]-[{{pod}}]-req-size-p95` |
+| Cached Bytes per Instance | `greptime_mito_cache_bytes{instance=~"$datanode"}` | `timeseries` | Cached Bytes per Instance. | `prometheus` | `decbytes` | `[{{instance}}]-[{{pod}}]-[{{type}}]` |
+| Inflight Compaction | `greptime_mito_inflight_compaction_count` | `timeseries` | Ongoing compaction task count | `prometheus` | `none` | `[{{instance}}]-[{{pod}}]` |
+| WAL sync duration seconds | `histogram_quantile(0.99, sum by(le, type, node, instance, pod) (rate(raft_engine_sync_log_duration_seconds_bucket[$__rate_interval])))` | `timeseries` | Raft engine (local disk) log store sync latency, p99 | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-p99` |
+| Log Store op duration seconds | `histogram_quantile(0.99, sum by(le,logstore,optype,instance, pod) (rate(greptime_logstore_op_elapsed_bucket[$__rate_interval])))` | `timeseries` | Write-ahead log operations latency at p99 | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{logstore}}]-[{{optype}}]-p99` |
+| Inflight Flush | `greptime_mito_inflight_flush_count` | `timeseries` | Ongoing flush task count | `prometheus` | `none` | `[{{instance}}]-[{{pod}}]` |
+| Compaction Input/Output Bytes | `sum by(instance, pod) (greptime_mito_compaction_input_bytes)`<br/>`sum by(instance, pod) (greptime_mito_compaction_output_bytes)` | `timeseries` | Compaction oinput output bytes | `prometheus` | `bytes` | `[{{instance}}]-[{{pod}}]-input` |
+| Region Worker Handle Bulk Insert Requests | `histogram_quantile(0.95, sum by(le,instance, stage, pod) (rate(greptime_region_worker_handle_write_bucket[$__rate_interval])))`<br/>`sum by(instance, stage, pod) (rate(greptime_region_worker_handle_write_sum[$__rate_interval]))/sum by(instance, stage, pod) (rate(greptime_region_worker_handle_write_count[$__rate_interval]))` | `timeseries` | Per-stage elapsed time for region worker to handle bulk insert region requests. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{stage}}]-P95` |
+| Active Series and Field Builders Count | `sum by(instance, pod) (greptime_mito_memtable_active_series_count)`<br/>`sum by(instance, pod) (greptime_mito_memtable_field_builder_count)` | `timeseries` | Compaction oinput output bytes | `prometheus` | `none` | `[{{instance}}]-[{{pod}}]-series` |
+| Region Worker Convert Requests | `histogram_quantile(0.95, sum by(le, instance, stage, pod) (rate(greptime_datanode_convert_region_request_bucket[$__rate_interval])))`<br/>`sum by(le,instance, stage, pod) (rate(greptime_datanode_convert_region_request_sum[$__rate_interval]))/sum by(le,instance, stage, pod) (rate(greptime_datanode_convert_region_request_count[$__rate_interval]))` | `timeseries` | Per-stage elapsed time for region worker to decode requests. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{stage}}]-P95` |
+# OpenDAL
+| Title | Query | Type | Description | Datasource | Unit | Legend Format |
+| --- | --- | --- | --- | --- | --- | --- |
+| QPS per Instance | `sum by(instance, pod, scheme, operation) (rate(opendal_operation_duration_seconds_count{instance=~"$datanode"}[$__rate_interval]))` | `timeseries` | QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]` |
+| Read QPS per Instance | `sum by(instance, pod, scheme) (rate(opendal_operation_duration_seconds_count{instance=~"$datanode", operation="read"}[$__rate_interval]))` | `timeseries` | Read QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]` |
+| Read P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, scheme) (rate(opendal_operation_duration_seconds_bucket{instance=~"$datanode",operation="read"}[$__rate_interval])))` | `timeseries` | Read P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-{{scheme}}` |
+| Write QPS per Instance | `sum by(instance, pod, scheme) (rate(opendal_operation_duration_seconds_count{instance=~"$datanode", operation="write"}[$__rate_interval]))` | `timeseries` | Write QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-{{scheme}}` |
+| Write P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, scheme) (rate(opendal_operation_duration_seconds_bucket{instance=~"$datanode", operation="write"}[$__rate_interval])))` | `timeseries` | Write P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]` |
+| List QPS per Instance | `sum by(instance, pod, scheme) (rate(opendal_operation_duration_seconds_count{instance=~"$datanode", operation="list"}[$__rate_interval]))` | `timeseries` | List QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]` |
+| List P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, scheme) (rate(opendal_operation_duration_seconds_bucket{instance=~"$datanode", operation="list"}[$__rate_interval])))` | `timeseries` | List P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]` |
+| Other Requests per Instance | `sum by(instance, pod, scheme, operation) (rate(opendal_operation_duration_seconds_count{instance=~"$datanode",operation!~"read\|write\|list\|stat"}[$__rate_interval]))` | `timeseries` | Other Requests per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]` |
+| Other Request P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, scheme, operation) (rate(opendal_operation_duration_seconds_bucket{instance=~"$datanode", operation!~"read\|write\|list"}[$__rate_interval])))` | `timeseries` | Other Request P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]` |
+| Opendal traffic | `sum by(instance, pod, scheme, operation) (rate(opendal_operation_bytes_sum{instance=~"$datanode"}[$__rate_interval]))` | `timeseries` | Total traffic as in bytes by instance and operation | `prometheus` | `decbytes` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]` |
+| OpenDAL errors per Instance | `sum by(instance, pod, scheme, operation, error) (rate(opendal_operation_errors_total{instance=~"$datanode", error!="NotFound"}[$__rate_interval]))` | `timeseries` | OpenDAL error counts per Instance. | `prometheus` | -- | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]-[{{error}}]` |
+# Metasrv
+| Title | Query | Type | Description | Datasource | Unit | Legend Format |
+| --- | --- | --- | --- | --- | --- | --- |
+| Region migration datanode | `greptime_meta_region_migration_stat{datanode_type="src"}`<br/>`greptime_meta_region_migration_stat{datanode_type="desc"}` | `status-history` | Counter of region migration by source and destination | `prometheus` | -- | `from-datanode-{{datanode_id}}` |
+| Region migration error | `greptime_meta_region_migration_error` | `timeseries` | Counter of region migration error | `prometheus` | `none` | `{{pod}}-{{state}}-{{error_type}}` |
+| Datanode load | `greptime_datanode_load` | `timeseries` | Gauge of load information of each datanode, collected via heartbeat between datanode and metasrv. This information is for metasrv to schedule workloads. | `prometheus` | `binBps` | `Datanode-{{datanode_id}}-writeload` |
+| Rate of SQL Executions (RDS) | `rate(greptime_meta_rds_pg_sql_execute_elapsed_ms_count[$__rate_interval])` | `timeseries` | Displays the rate of SQL executions processed by the Meta service using the RDS backend. | `prometheus` | `none` | `{{pod}} {{op}} {{type}} {{result}} ` |
+| SQL Execution Latency (RDS) | `histogram_quantile(0.90, sum by(pod, op, type, result, le) (rate(greptime_meta_rds_pg_sql_execute_elapsed_ms_bucket[$__rate_interval])))` | `timeseries` | Measures the response time of SQL executions via the RDS backend.  | `prometheus` | `ms` | `{{pod}} {{op}} {{type}} {{result}} p90` |
+| Handler Execution Latency | `histogram_quantile(0.90, sum by(pod, le, name) (
+  rate(greptime_meta_handler_execute_bucket[$__rate_interval])
+))` | `timeseries` | Shows latency of Meta handlers by pod and handler name, useful for monitoring handler performance and detecting latency spikes.<br/> | `prometheus` | `s` | `{{pod}} {{name}} p90` |
+| Heartbeat Packet Size | `histogram_quantile(0.9, sum by(pod, le) (greptime_meta_heartbeat_stat_memory_size_bucket))` | `timeseries` | Shows p90 heartbeat message sizes, helping track network usage and identify anomalies in heartbeat payload.<br/> | `prometheus` | `bytes` | `{{pod}}` |
+| Meta Heartbeat Receive Rate | `rate(greptime_meta_heartbeat_rate[$__rate_interval])` | `timeseries` | Gauge of load information of each datanode, collected via heartbeat between datanode and metasrv. This information is for metasrv to schedule workloads. | `prometheus` | `s` | `{{pod}}` |
+| Meta KV Ops Latency | `histogram_quantile(0.99, sum by(pod, le, op, target) (greptime_meta_kv_request_elapsed_bucket))` | `timeseries` | Gauge of load information of each datanode, collected via heartbeat between datanode and metasrv. This information is for metasrv to schedule workloads. | `prometheus` | `s` | `{{pod}}-{{op}} p99` |
+| Rate of meta KV Ops | `rate(greptime_meta_kv_request_elapsed_count[$__rate_interval])` | `timeseries` | Gauge of load information of each datanode, collected via heartbeat between datanode and metasrv. This information is for metasrv to schedule workloads. | `prometheus` | `none` | `{{pod}}-{{op}} p99` |
+| DDL Latency | `histogram_quantile(0.9, sum by(le, pod, step) (greptime_meta_procedure_create_tables_bucket))`<br/>`histogram_quantile(0.9, sum by(le, pod, step) (greptime_meta_procedure_create_table))`<br/>`histogram_quantile(0.9, sum by(le, pod, step) (greptime_meta_procedure_create_view))`<br/>`histogram_quantile(0.9, sum by(le, pod, step) (greptime_meta_procedure_create_flow))`<br/>`histogram_quantile(0.9, sum by(le, pod, step) (greptime_meta_procedure_drop_table))`<br/>`histogram_quantile(0.9, sum by(le, pod, step) (greptime_meta_procedure_alter_table))` | `timeseries` | Gauge of load information of each datanode, collected via heartbeat between datanode and metasrv. This information is for metasrv to schedule workloads. | `prometheus` | `s` | `CreateLogicalTables-{{step}} p90` |
+# Flownode
+| Title | Query | Type | Description | Datasource | Unit | Legend Format |
+| --- | --- | --- | --- | --- | --- | --- |
+| Flow Ingest / Output Rate | `sum by(instance, pod, direction) (rate(greptime_flow_processed_rows[$__rate_interval]))` | `timeseries` | Flow Ingest / Output Rate. | `prometheus` | -- | `[{{pod}}]-[{{instance}}]-[{{direction}}]` |
+| Flow Ingest Latency | `histogram_quantile(0.95, sum(rate(greptime_flow_insert_elapsed_bucket[$__rate_interval])) by (le, instance, pod))`<br/>`histogram_quantile(0.99, sum(rate(greptime_flow_insert_elapsed_bucket[$__rate_interval])) by (le, instance, pod))` | `timeseries` | Flow Ingest Latency. | `prometheus` | -- | `[{{instance}}]-[{{pod}}]-p95` |
+| Flow Operation Latency | `histogram_quantile(0.95, sum(rate(greptime_flow_processing_time_bucket[$__rate_interval])) by (le,instance,pod,type))`<br/>`histogram_quantile(0.99, sum(rate(greptime_flow_processing_time_bucket[$__rate_interval])) by (le,instance,pod,type))` | `timeseries` | Flow Operation Latency. | `prometheus` | -- | `[{{instance}}]-[{{pod}}]-[{{type}}]-p95` |
+| Flow Buffer Size per Instance | `greptime_flow_input_buf_size` | `timeseries` | Flow Buffer Size per Instance. | `prometheus` | -- | `[{{instance}}]-[{{pod}]` |
+| Flow Processing Error per Instance | `sum by(instance,pod,code) (rate(greptime_flow_errors[$__rate_interval]))` | `timeseries` | Flow Processing Error per Instance. | `prometheus` | -- | `[{{instance}}]-[{{pod}}]-[{{code}}]` |

grafana/dashboards/metrics/cluster/dashboard.yaml

@@ -0,0 +1,958 @@
+groups:
+    - title: Overview
+      panels:
+        - title: Uptime
+          type: stat
+          description: The start time of GreptimeDB.
+          unit: s
+          queries:
+            - expr: time() - process_start_time_seconds
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: __auto
+        - title: Version
+          type: stat
+          description: GreptimeDB version.
+          queries:
+            - expr: SELECT pkg_version FROM information_schema.build_info
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+        - title: Total Ingestion Rate
+          type: stat
+          description: Total ingestion rate.
+          unit: rowsps
+          queries:
+            - expr: sum(rate(greptime_table_operator_ingest_rows[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: __auto
+        - title: Total Storage Size
+          type: stat
+          description: Total number of data file size.
+          unit: decbytes
+          queries:
+            - expr: select SUM(disk_size) from information_schema.region_statistics;
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+        - title: Total Rows
+          type: stat
+          description: Total number of data rows in the cluster. Calculated by sum of rows from each region.
+          unit: sishort
+          queries:
+            - expr: select SUM(region_rows) from information_schema.region_statistics;
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+        - title: Deployment
+          type: stat
+          description: The deployment topology of GreptimeDB.
+          queries:
+            - expr: SELECT count(*) as datanode FROM information_schema.cluster_info WHERE peer_type = 'DATANODE';
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+            - expr: SELECT count(*) as frontend FROM information_schema.cluster_info WHERE peer_type = 'FRONTEND';
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+            - expr: SELECT count(*) as metasrv FROM information_schema.cluster_info WHERE peer_type = 'METASRV';
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+            - expr: SELECT count(*) as flownode FROM information_schema.cluster_info WHERE peer_type = 'FLOWNODE';
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+        - title: Database Resources
+          type: stat
+          description: The number of the key resources in GreptimeDB.
+          queries:
+            - expr: SELECT COUNT(*) as databases FROM information_schema.schemata WHERE schema_name NOT IN ('greptime_private', 'information_schema')
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+            - expr: SELECT COUNT(*) as tables FROM information_schema.tables WHERE table_schema != 'information_schema'
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+            - expr: SELECT COUNT(region_id) as regions FROM information_schema.region_peers
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+            - expr: SELECT COUNT(*) as flows FROM information_schema.flows
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+        - title: Data Size
+          type: stat
+          description: The data size of wal/index/manifest in the GreptimeDB.
+          unit: decbytes
+          queries:
+            - expr: SELECT SUM(memtable_size) * 0.42825 as WAL FROM information_schema.region_statistics;
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+            - expr: SELECT SUM(index_size) as index FROM information_schema.region_statistics;
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+            - expr: SELECT SUM(manifest_size) as manifest FROM information_schema.region_statistics;
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+    - title: Ingestion
+      panels:
+        - title: Total Ingestion Rate
+          type: timeseries
+          description: |
+            Total ingestion rate.
+
+            Here we listed 3 primary protocols:
+
+            - Prometheus remote write
+            - Greptime's gRPC API (when using our ingest SDK)
+            - Log ingestion http API
+          unit: rowsps
+          queries:
+            - expr: sum(rate(greptime_table_operator_ingest_rows{instance=~"$frontend"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: ingestion
+        - title: Ingestion Rate by Type
+          type: timeseries
+          description: |
+            Total ingestion rate.
+
+            Here we listed 3 primary protocols:
+
+            - Prometheus remote write
+            - Greptime's gRPC API (when using our ingest SDK)
+            - Log ingestion http API
+          unit: rowsps
+          queries:
+            - expr: sum(rate(greptime_servers_http_logs_ingestion_counter[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: http-logs
+            - expr: sum(rate(greptime_servers_prometheus_remote_write_samples[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: prometheus-remote-write
+    - title: Queries
+      panels:
+        - title: Total Query Rate
+          type: timeseries
+          description: |-
+            Total rate of query API calls by protocol. This metric is collected from frontends.
+
+            Here we listed 3 main protocols:
+            - MySQL
+            - Postgres
+            - Prometheus API
+
+            Note that there are some other minor query APIs like /sql are not included
+          unit: reqps
+          queries:
+            - expr: sum (rate(greptime_servers_mysql_query_elapsed_count{instance=~"$frontend"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: mysql
+            - expr: sum (rate(greptime_servers_postgres_query_elapsed_count{instance=~"$frontend"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: pg
+            - expr: sum (rate(greptime_servers_http_promql_elapsed_counte{instance=~"$frontend"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: promql
+    - title: Resources
+      panels:
+        - title: Datanode Memory per Instance
+          type: timeseries
+          description: Current memory usage by instance
+          unit: decbytes
+          queries:
+            - expr: sum(process_resident_memory_bytes{instance=~"$datanode"}) by (instance, pod)
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{ pod }}]'
+        - title: Datanode CPU Usage per Instance
+          type: timeseries
+          description: Current cpu usage by instance
+          unit: none
+          queries:
+            - expr: sum(rate(process_cpu_seconds_total{instance=~"$datanode"}[$__rate_interval]) * 1000) by (instance, pod)
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{ instance }}]-[{{ pod }}]'
+        - title: Frontend Memory per Instance
+          type: timeseries
+          description: Current memory usage by instance
+          unit: decbytes
+          queries:
+            - expr: sum(process_resident_memory_bytes{instance=~"$frontend"}) by (instance, pod)
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{ instance }}]-[{{ pod }}]'
+        - title: Frontend CPU Usage per Instance
+          type: timeseries
+          description: Current cpu usage by instance
+          unit: none
+          queries:
+            - expr: sum(rate(process_cpu_seconds_total{instance=~"$frontend"}[$__rate_interval]) * 1000) by (instance, pod)
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{ instance }}]-[{{ pod }}]-cpu'
+        - title: Metasrv Memory per Instance
+          type: timeseries
+          description: Current memory usage by instance
+          unit: decbytes
+          queries:
+            - expr: sum(process_resident_memory_bytes{instance=~"$metasrv"}) by (instance, pod)
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{ instance }}]-[{{ pod }}]-resident'
+        - title: Metasrv CPU Usage per Instance
+          type: timeseries
+          description: Current cpu usage by instance
+          unit: none
+          queries:
+            - expr: sum(rate(process_cpu_seconds_total{instance=~"$metasrv"}[$__rate_interval]) * 1000) by (instance, pod)
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{ instance }}]-[{{ pod }}]'
+        - title: Flownode Memory per Instance
+          type: timeseries
+          description: Current memory usage by instance
+          unit: decbytes
+          queries:
+            - expr: sum(process_resident_memory_bytes{instance=~"$flownode"}) by (instance, pod)
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{ instance }}]-[{{ pod }}]'
+        - title: Flownode CPU Usage per Instance
+          type: timeseries
+          description: Current cpu usage by instance
+          unit: none
+          queries:
+            - expr: sum(rate(process_cpu_seconds_total{instance=~"$flownode"}[$__rate_interval]) * 1000) by (instance, pod)
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{ instance }}]-[{{ pod }}]'
+    - title: Frontend Requests
+      panels:
+        - title: HTTP QPS per Instance
+          type: timeseries
+          description: HTTP QPS per Instance.
+          unit: reqps
+          queries:
+            - expr: sum by(instance, pod, path, method, code) (rate(greptime_servers_http_requests_elapsed_count{instance=~"$frontend",path!~"/health|/metrics"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{path}}]-[{{method}}]-[{{code}}]'
+        - title: HTTP P99 per Instance
+          type: timeseries
+          description: HTTP P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(instance, pod, le, path, method, code) (rate(greptime_servers_http_requests_elapsed_bucket{instance=~"$frontend",path!~"/health|/metrics"}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{path}}]-[{{method}}]-[{{code}}]-p99'
+        - title: gRPC QPS per Instance
+          type: timeseries
+          description: gRPC QPS per Instance.
+          unit: reqps
+          queries:
+            - expr: sum by(instance, pod, path, code) (rate(greptime_servers_grpc_requests_elapsed_count{instance=~"$frontend"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{path}}]-[{{code}}]'
+        - title: gRPC P99 per Instance
+          type: timeseries
+          description: gRPC P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(instance, pod, le, path, code) (rate(greptime_servers_grpc_requests_elapsed_bucket{instance=~"$frontend"}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{path}}]-[{{method}}]-[{{code}}]-p99'
+        - title: MySQL QPS per Instance
+          type: timeseries
+          description: MySQL QPS per Instance.
+          unit: reqps
+          queries:
+            - expr: sum by(pod, instance)(rate(greptime_servers_mysql_query_elapsed_count{instance=~"$frontend"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]'
+        - title: MySQL P99 per Instance
+          type: timeseries
+          description: MySQL P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(pod, instance, le) (rate(greptime_servers_mysql_query_elapsed_bucket{instance=~"$frontend"}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{ instance }}]-[{{ pod }}]-p99'
+        - title: PostgreSQL QPS per Instance
+          type: timeseries
+          description: PostgreSQL QPS per Instance.
+          unit: reqps
+          queries:
+            - expr: sum by(pod, instance)(rate(greptime_servers_postgres_query_elapsed_count{instance=~"$frontend"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]'
+        - title: PostgreSQL P99 per Instance
+          type: timeseries
+          description: PostgreSQL P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(pod,instance,le) (rate(greptime_servers_postgres_query_elapsed_bucket{instance=~"$frontend"}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-p99'
+    - title: Frontend to Datanode
+      panels:
+        - title: Ingest Rows per Instance
+          type: timeseries
+          description: Ingestion rate by row as in each frontend
+          unit: rowsps
+          queries:
+            - expr: sum by(instance, pod)(rate(greptime_table_operator_ingest_rows{instance=~"$frontend"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]'
+        - title: Region Call QPS per Instance
+          type: timeseries
+          description: Region Call QPS per Instance.
+          unit: ops
+          queries:
+            - expr: sum by(instance, pod, request_type) (rate(greptime_grpc_region_request_count{instance=~"$frontend"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{request_type}}]'
+        - title: Region Call P99 per Instance
+          type: timeseries
+          description: Region Call P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(instance, pod, le, request_type) (rate(greptime_grpc_region_request_bucket{instance=~"$frontend"}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{request_type}}]'
+        - title: 'Frontend Handle Bulk Insert Elapsed Time '
+          type: timeseries
+          description: Per-stage time for frontend to handle bulk insert requests
+          unit: s
+          queries:
+            - expr: sum by(instance, pod, stage) (rate(greptime_table_operator_handle_bulk_insert_sum[$__rate_interval]))/sum by(instance, pod, stage) (rate(greptime_table_operator_handle_bulk_insert_count[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{stage}}]-AVG'
+            - expr: histogram_quantile(0.99, sum by(instance, pod, stage, le) (rate(greptime_table_operator_handle_bulk_insert_bucket[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{stage}}]-P95'
+    - title: Mito Engine
+      panels:
+        - title: Request OPS per Instance
+          type: timeseries
+          description: Request QPS per Instance.
+          unit: ops
+          queries:
+            - expr: sum by(instance, pod, type) (rate(greptime_mito_handle_request_elapsed_count{instance=~"$datanode"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{type}}]'
+        - title: Request P99 per Instance
+          type: timeseries
+          description: Request P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(instance, pod, le, type) (rate(greptime_mito_handle_request_elapsed_bucket{instance=~"$datanode"}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{type}}]'
+        - title: Write Buffer per Instance
+          type: timeseries
+          description: Write Buffer per Instance.
+          unit: decbytes
+          queries:
+            - expr: greptime_mito_write_buffer_bytes{instance=~"$datanode"}
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]'
+        - title: Write Rows per Instance
+          type: timeseries
+          description: Ingestion size by row counts.
+          unit: rowsps
+          queries:
+            - expr: sum by (instance, pod) (rate(greptime_mito_write_rows_total{instance=~"$datanode"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]'
+        - title: Flush OPS per Instance
+          type: timeseries
+          description: Flush QPS per Instance.
+          unit: ops
+          queries:
+            - expr: sum by(instance, pod, reason) (rate(greptime_mito_flush_requests_total{instance=~"$datanode"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{reason}}]'
+        - title: Write Stall per Instance
+          type: timeseries
+          description: Write Stall per Instance.
+          queries:
+            - expr: sum by(instance, pod) (greptime_mito_write_stall_total{instance=~"$datanode"})
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]'
+        - title: Read Stage OPS per Instance
+          type: timeseries
+          description: Read Stage OPS per Instance.
+          unit: ops
+          queries:
+            - expr: sum by(instance, pod) (rate(greptime_mito_read_stage_elapsed_count{instance=~"$datanode", stage="total"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]'
+        - title: Read Stage P99 per Instance
+          type: timeseries
+          description: Read Stage P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(instance, pod, le, stage) (rate(greptime_mito_read_stage_elapsed_bucket{instance=~"$datanode"}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{stage}}]'
+        - title: Write Stage P99 per Instance
+          type: timeseries
+          description: Write Stage P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(instance, pod, le, stage) (rate(greptime_mito_write_stage_elapsed_bucket{instance=~"$datanode"}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{stage}}]'
+        - title: Compaction OPS per Instance
+          type: timeseries
+          description: Compaction OPS per Instance.
+          unit: ops
+          queries:
+            - expr: sum by(instance, pod) (rate(greptime_mito_compaction_total_elapsed_count{instance=~"$datanode"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{ instance }}]-[{{pod}}]'
+        - title: Compaction Elapsed Time per Instance by Stage
+          type: timeseries
+          description: Compaction latency by stage
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(instance, pod, le, stage) (rate(greptime_mito_compaction_stage_elapsed_bucket{instance=~"$datanode"}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{stage}}]-p99'
+            - expr: sum by(instance, pod, stage) (rate(greptime_mito_compaction_stage_elapsed_sum{instance=~"$datanode"}[$__rate_interval]))/sum by(instance, pod, stage) (rate(greptime_mito_compaction_stage_elapsed_count{instance=~"$datanode"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{stage}}]-avg'
+        - title: Compaction P99 per Instance
+          type: timeseries
+          description: Compaction P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(instance, pod, le,stage) (rate(greptime_mito_compaction_total_elapsed_bucket{instance=~"$datanode"}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{stage}}]-compaction'
+        - title: WAL write size
+          type: timeseries
+          description: Write-ahead logs write size as bytes. This chart includes stats of p95 and p99 size by instance, total WAL write rate.
+          unit: bytes
+          queries:
+            - expr: histogram_quantile(0.95, sum by(le,instance, pod) (rate(raft_engine_write_size_bucket[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-req-size-p95'
+            - expr: histogram_quantile(0.99, sum by(le,instance,pod) (rate(raft_engine_write_size_bucket[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-req-size-p99'
+            - expr: sum by (instance, pod)(rate(raft_engine_write_size_sum[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-throughput'
+        - title: Cached Bytes per Instance
+          type: timeseries
+          description: Cached Bytes per Instance.
+          unit: decbytes
+          queries:
+            - expr: greptime_mito_cache_bytes{instance=~"$datanode"}
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{type}}]'
+        - title: Inflight Compaction
+          type: timeseries
+          description: Ongoing compaction task count
+          unit: none
+          queries:
+            - expr: greptime_mito_inflight_compaction_count
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]'
+        - title: WAL sync duration seconds
+          type: timeseries
+          description: Raft engine (local disk) log store sync latency, p99
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(le, type, node, instance, pod) (rate(raft_engine_sync_log_duration_seconds_bucket[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-p99'
+        - title: Log Store op duration seconds
+          type: timeseries
+          description: Write-ahead log operations latency at p99
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(le,logstore,optype,instance, pod) (rate(greptime_logstore_op_elapsed_bucket[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{logstore}}]-[{{optype}}]-p99'
+        - title: Inflight Flush
+          type: timeseries
+          description: Ongoing flush task count
+          unit: none
+          queries:
+            - expr: greptime_mito_inflight_flush_count
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]'
+        - title: Compaction Input/Output Bytes
+          type: timeseries
+          description: Compaction oinput output bytes
+          unit: bytes
+          queries:
+            - expr: sum by(instance, pod) (greptime_mito_compaction_input_bytes)
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-input'
+            - expr: sum by(instance, pod) (greptime_mito_compaction_output_bytes)
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-output'
+        - title: Region Worker Handle Bulk Insert Requests
+          type: timeseries
+          description: Per-stage elapsed time for region worker to handle bulk insert region requests.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.95, sum by(le,instance, stage, pod) (rate(greptime_region_worker_handle_write_bucket[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{stage}}]-P95'
+            - expr: sum by(instance, stage, pod) (rate(greptime_region_worker_handle_write_sum[$__rate_interval]))/sum by(instance, stage, pod) (rate(greptime_region_worker_handle_write_count[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{stage}}]-AVG'
+        - title: Active Series and Field Builders Count
+          type: timeseries
+          description: Compaction oinput output bytes
+          unit: none
+          queries:
+            - expr: sum by(instance, pod) (greptime_mito_memtable_active_series_count)
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-series'
+            - expr: sum by(instance, pod) (greptime_mito_memtable_field_builder_count)
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-field_builders'
+        - title: Region Worker Convert Requests
+          type: timeseries
+          description: Per-stage elapsed time for region worker to decode requests.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.95, sum by(le, instance, stage, pod) (rate(greptime_datanode_convert_region_request_bucket[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{stage}}]-P95'
+            - expr: sum by(le,instance, stage, pod) (rate(greptime_datanode_convert_region_request_sum[$__rate_interval]))/sum by(le,instance, stage, pod) (rate(greptime_datanode_convert_region_request_count[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{stage}}]-AVG'
+    - title: OpenDAL
+      panels:
+        - title: QPS per Instance
+          type: timeseries
+          description: QPS per Instance.
+          unit: ops
+          queries:
+            - expr: sum by(instance, pod, scheme, operation) (rate(opendal_operation_duration_seconds_count{instance=~"$datanode"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]'
+        - title: Read QPS per Instance
+          type: timeseries
+          description: Read QPS per Instance.
+          unit: ops
+          queries:
+            - expr: sum by(instance, pod, scheme) (rate(opendal_operation_duration_seconds_count{instance=~"$datanode", operation="read"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{scheme}}]'
+        - title: Read P99 per Instance
+          type: timeseries
+          description: Read P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(instance, pod, le, scheme) (rate(opendal_operation_duration_seconds_bucket{instance=~"$datanode",operation="read"}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-{{scheme}}'
+        - title: Write QPS per Instance
+          type: timeseries
+          description: Write QPS per Instance.
+          unit: ops
+          queries:
+            - expr: sum by(instance, pod, scheme) (rate(opendal_operation_duration_seconds_count{instance=~"$datanode", operation="write"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-{{scheme}}'
+        - title: Write P99 per Instance
+          type: timeseries
+          description: Write P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(instance, pod, le, scheme) (rate(opendal_operation_duration_seconds_bucket{instance=~"$datanode", operation="write"}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{scheme}}]'
+        - title: List QPS per Instance
+          type: timeseries
+          description: List QPS per Instance.
+          unit: ops
+          queries:
+            - expr: sum by(instance, pod, scheme) (rate(opendal_operation_duration_seconds_count{instance=~"$datanode", operation="list"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{scheme}}]'
+        - title: List P99 per Instance
+          type: timeseries
+          description: List P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(instance, pod, le, scheme) (rate(opendal_operation_duration_seconds_bucket{instance=~"$datanode", operation="list"}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{scheme}}]'
+        - title: Other Requests per Instance
+          type: timeseries
+          description: Other Requests per Instance.
+          unit: ops
+          queries:
+            - expr: sum by(instance, pod, scheme, operation) (rate(opendal_operation_duration_seconds_count{instance=~"$datanode",operation!~"read|write|list|stat"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]'
+        - title: Other Request P99 per Instance
+          type: timeseries
+          description: Other Request P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(instance, pod, le, scheme, operation) (rate(opendal_operation_duration_seconds_bucket{instance=~"$datanode", operation!~"read|write|list"}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]'
+        - title: Opendal traffic
+          type: timeseries
+          description: Total traffic as in bytes by instance and operation
+          unit: decbytes
+          queries:
+            - expr: sum by(instance, pod, scheme, operation) (rate(opendal_operation_bytes_sum{instance=~"$datanode"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]'
+        - title: OpenDAL errors per Instance
+          type: timeseries
+          description: OpenDAL error counts per Instance.
+          queries:
+            - expr: sum by(instance, pod, scheme, operation, error) (rate(opendal_operation_errors_total{instance=~"$datanode", error!="NotFound"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]-[{{error}}]'
+    - title: Metasrv
+      panels:
+        - title: Region migration datanode
+          type: status-history
+          description: Counter of region migration by source and destination
+          queries:
+            - expr: greptime_meta_region_migration_stat{datanode_type="src"}
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: from-datanode-{{datanode_id}}
+            - expr: greptime_meta_region_migration_stat{datanode_type="desc"}
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: to-datanode-{{datanode_id}}
+        - title: Region migration error
+          type: timeseries
+          description: Counter of region migration error
+          unit: none
+          queries:
+            - expr: greptime_meta_region_migration_error
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '{{pod}}-{{state}}-{{error_type}}'
+        - title: Datanode load
+          type: timeseries
+          description: Gauge of load information of each datanode, collected via heartbeat between datanode and metasrv. This information is for metasrv to schedule workloads.
+          unit: binBps
+          queries:
+            - expr: greptime_datanode_load
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: Datanode-{{datanode_id}}-writeload
+        - title: Rate of SQL Executions (RDS)
+          type: timeseries
+          description: Displays the rate of SQL executions processed by the Meta service using the RDS backend.
+          unit: none
+          queries:
+            - expr: rate(greptime_meta_rds_pg_sql_execute_elapsed_ms_count[$__rate_interval])
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '{{pod}} {{op}} {{type}} {{result}} '
+        - title: SQL Execution Latency (RDS)
+          type: timeseries
+          description: 'Measures the response time of SQL executions via the RDS backend. '
+          unit: ms
+          queries:
+            - expr: histogram_quantile(0.90, sum by(pod, op, type, result, le) (rate(greptime_meta_rds_pg_sql_execute_elapsed_ms_bucket[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '{{pod}} {{op}} {{type}} {{result}} p90'
+        - title: Handler Execution Latency
+          type: timeseries
+          description: |
+            Shows latency of Meta handlers by pod and handler name, useful for monitoring handler performance and detecting latency spikes.
+          unit: s
+          queries:
+            - expr: |-
+                histogram_quantile(0.90, sum by(pod, le, name) (
+                  rate(greptime_meta_handler_execute_bucket[$__rate_interval])
+                ))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '{{pod}} {{name}} p90'
+        - title: Heartbeat Packet Size
+          type: timeseries
+          description: |
+            Shows p90 heartbeat message sizes, helping track network usage and identify anomalies in heartbeat payload.
+          unit: bytes
+          queries:
+            - expr: histogram_quantile(0.9, sum by(pod, le) (greptime_meta_heartbeat_stat_memory_size_bucket))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '{{pod}}'
+        - title: Meta Heartbeat Receive Rate
+          type: timeseries
+          description: Gauge of load information of each datanode, collected via heartbeat between datanode and metasrv. This information is for metasrv to schedule workloads.
+          unit: s
+          queries:
+            - expr: rate(greptime_meta_heartbeat_rate[$__rate_interval])
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '{{pod}}'
+        - title: Meta KV Ops Latency
+          type: timeseries
+          description: Gauge of load information of each datanode, collected via heartbeat between datanode and metasrv. This information is for metasrv to schedule workloads.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(pod, le, op, target) (greptime_meta_kv_request_elapsed_bucket))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '{{pod}}-{{op}} p99'
+        - title: Rate of meta KV Ops
+          type: timeseries
+          description: Gauge of load information of each datanode, collected via heartbeat between datanode and metasrv. This information is for metasrv to schedule workloads.
+          unit: none
+          queries:
+            - expr: rate(greptime_meta_kv_request_elapsed_count[$__rate_interval])
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '{{pod}}-{{op}} p99'
+        - title: DDL Latency
+          type: timeseries
+          description: Gauge of load information of each datanode, collected via heartbeat between datanode and metasrv. This information is for metasrv to schedule workloads.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.9, sum by(le, pod, step) (greptime_meta_procedure_create_tables_bucket))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: CreateLogicalTables-{{step}} p90
+            - expr: histogram_quantile(0.9, sum by(le, pod, step) (greptime_meta_procedure_create_table))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: CreateTable-{{step}} p90
+            - expr: histogram_quantile(0.9, sum by(le, pod, step) (greptime_meta_procedure_create_view))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: CreateView-{{step}} p90
+            - expr: histogram_quantile(0.9, sum by(le, pod, step) (greptime_meta_procedure_create_flow))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: CreateFlow-{{step}} p90
+            - expr: histogram_quantile(0.9, sum by(le, pod, step) (greptime_meta_procedure_drop_table))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: DropTable-{{step}} p90
+            - expr: histogram_quantile(0.9, sum by(le, pod, step) (greptime_meta_procedure_alter_table))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: AlterTable-{{step}} p90
+    - title: Flownode
+      panels:
+        - title: Flow Ingest / Output Rate
+          type: timeseries
+          description: Flow Ingest / Output Rate.
+          queries:
+            - expr: sum by(instance, pod, direction) (rate(greptime_flow_processed_rows[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{pod}}]-[{{instance}}]-[{{direction}}]'
+        - title: Flow Ingest Latency
+          type: timeseries
+          description: Flow Ingest Latency.
+          queries:
+            - expr: histogram_quantile(0.95, sum(rate(greptime_flow_insert_elapsed_bucket[$__rate_interval])) by (le, instance, pod))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-p95'
+            - expr: histogram_quantile(0.99, sum(rate(greptime_flow_insert_elapsed_bucket[$__rate_interval])) by (le, instance, pod))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-p99'
+        - title: Flow Operation Latency
+          type: timeseries
+          description: Flow Operation Latency.
+          queries:
+            - expr: histogram_quantile(0.95, sum(rate(greptime_flow_processing_time_bucket[$__rate_interval])) by (le,instance,pod,type))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{type}}]-p95'
+            - expr: histogram_quantile(0.99, sum(rate(greptime_flow_processing_time_bucket[$__rate_interval])) by (le,instance,pod,type))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{type}}]-p99'
+        - title: Flow Buffer Size per Instance
+          type: timeseries
+          description: Flow Buffer Size per Instance.
+          queries:
+            - expr: greptime_flow_input_buf_size
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}]'
+        - title: Flow Processing Error per Instance
+          type: timeseries
+          description: Flow Processing Error per Instance.
+          queries:
+            - expr: sum by(instance,pod,code) (rate(greptime_flow_errors[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{code}}]'

grafana/dashboards/metrics/standalone/dashboard.md

@@ -0,0 +1,112 @@
+# Overview
+| Title | Query | Type | Description | Datasource | Unit | Legend Format |
+| --- | --- | --- | --- | --- | --- | --- |
+| Uptime | `time() - process_start_time_seconds` | `stat` | The start time of GreptimeDB. | `prometheus` | `s` | `__auto` |
+| Version | `SELECT pkg_version FROM information_schema.build_info` | `stat` | GreptimeDB version. | `mysql` | -- | -- |
+| Total Ingestion Rate | `sum(rate(greptime_table_operator_ingest_rows[$__rate_interval]))` | `stat` | Total ingestion rate. | `prometheus` | `rowsps` | `__auto` |
+| Total Storage Size | `select SUM(disk_size) from information_schema.region_statistics;` | `stat` | Total number of data file size. | `mysql` | `decbytes` | -- |
+| Total Rows | `select SUM(region_rows) from information_schema.region_statistics;` | `stat` | Total number of data rows in the cluster. Calculated by sum of rows from each region. | `mysql` | `sishort` | -- |
+| Deployment | `SELECT count(*) as datanode FROM information_schema.cluster_info WHERE peer_type = 'DATANODE';`<br/>`SELECT count(*) as frontend FROM information_schema.cluster_info WHERE peer_type = 'FRONTEND';`<br/>`SELECT count(*) as metasrv FROM information_schema.cluster_info WHERE peer_type = 'METASRV';`<br/>`SELECT count(*) as flownode FROM information_schema.cluster_info WHERE peer_type = 'FLOWNODE';` | `stat` | The deployment topology of GreptimeDB. | `mysql` | -- | -- |
+| Database Resources | `SELECT COUNT(*) as databases FROM information_schema.schemata WHERE schema_name NOT IN ('greptime_private', 'information_schema')`<br/>`SELECT COUNT(*) as tables FROM information_schema.tables WHERE table_schema != 'information_schema'`<br/>`SELECT COUNT(region_id) as regions FROM information_schema.region_peers`<br/>`SELECT COUNT(*) as flows FROM information_schema.flows` | `stat` | The number of the key resources in GreptimeDB. | `mysql` | -- | -- |
+| Data Size | `SELECT SUM(memtable_size) * 0.42825 as WAL FROM information_schema.region_statistics;`<br/>`SELECT SUM(index_size) as index FROM information_schema.region_statistics;`<br/>`SELECT SUM(manifest_size) as manifest FROM information_schema.region_statistics;` | `stat` | The data size of wal/index/manifest in the GreptimeDB. | `mysql` | `decbytes` | -- |
+# Ingestion
+| Title | Query | Type | Description | Datasource | Unit | Legend Format |
+| --- | --- | --- | --- | --- | --- | --- |
+| Total Ingestion Rate | `sum(rate(greptime_table_operator_ingest_rows{}[$__rate_interval]))` | `timeseries` | Total ingestion rate.<br/><br/>Here we listed 3 primary protocols:<br/><br/>- Prometheus remote write<br/>- Greptime's gRPC API (when using our ingest SDK)<br/>- Log ingestion http API<br/> | `prometheus` | `rowsps` | `ingestion` |
+| Ingestion Rate by Type | `sum(rate(greptime_servers_http_logs_ingestion_counter[$__rate_interval]))`<br/>`sum(rate(greptime_servers_prometheus_remote_write_samples[$__rate_interval]))` | `timeseries` | Total ingestion rate.<br/><br/>Here we listed 3 primary protocols:<br/><br/>- Prometheus remote write<br/>- Greptime's gRPC API (when using our ingest SDK)<br/>- Log ingestion http API<br/> | `prometheus` | `rowsps` | `http-logs` |
+# Queries
+| Title | Query | Type | Description | Datasource | Unit | Legend Format |
+| --- | --- | --- | --- | --- | --- | --- |
+| Total Query Rate | `sum (rate(greptime_servers_mysql_query_elapsed_count{}[$__rate_interval]))`<br/>`sum (rate(greptime_servers_postgres_query_elapsed_count{}[$__rate_interval]))`<br/>`sum (rate(greptime_servers_http_promql_elapsed_counte{}[$__rate_interval]))` | `timeseries` | Total rate of query API calls by protocol. This metric is collected from frontends.<br/><br/>Here we listed 3 main protocols:<br/>- MySQL<br/>- Postgres<br/>- Prometheus API<br/><br/>Note that there are some other minor query APIs like /sql are not included | `prometheus` | `reqps` | `mysql` |
+# Resources
+| Title | Query | Type | Description | Datasource | Unit | Legend Format |
+| --- | --- | --- | --- | --- | --- | --- |
+| Datanode Memory per Instance | `sum(process_resident_memory_bytes{}) by (instance, pod)` | `timeseries` | Current memory usage by instance | `prometheus` | `decbytes` | `[{{instance}}]-[{{ pod }}]` |
+| Datanode CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{}[$__rate_interval]) * 1000) by (instance, pod)` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]` |
+| Frontend Memory per Instance | `sum(process_resident_memory_bytes{}) by (instance, pod)` | `timeseries` | Current memory usage by instance | `prometheus` | `decbytes` | `[{{ instance }}]-[{{ pod }}]` |
+| Frontend CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{}[$__rate_interval]) * 1000) by (instance, pod)` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]-cpu` |
+| Metasrv Memory per Instance | `sum(process_resident_memory_bytes{}) by (instance, pod)` | `timeseries` | Current memory usage by instance | `prometheus` | `decbytes` | `[{{ instance }}]-[{{ pod }}]-resident` |
+| Metasrv CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{}[$__rate_interval]) * 1000) by (instance, pod)` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]` |
+| Flownode Memory per Instance | `sum(process_resident_memory_bytes{}) by (instance, pod)` | `timeseries` | Current memory usage by instance | `prometheus` | `decbytes` | `[{{ instance }}]-[{{ pod }}]` |
+| Flownode CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{}[$__rate_interval]) * 1000) by (instance, pod)` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]` |
+# Frontend Requests
+| Title | Query | Type | Description | Datasource | Unit | Legend Format |
+| --- | --- | --- | --- | --- | --- | --- |
+| HTTP QPS per Instance | `sum by(instance, pod, path, method, code) (rate(greptime_servers_http_requests_elapsed_count{path!~"/health\|/metrics"}[$__rate_interval]))` | `timeseries` | HTTP QPS per Instance. | `prometheus` | `reqps` | `[{{instance}}]-[{{pod}}]-[{{path}}]-[{{method}}]-[{{code}}]` |
+| HTTP P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, path, method, code) (rate(greptime_servers_http_requests_elapsed_bucket{path!~"/health\|/metrics"}[$__rate_interval])))` | `timeseries` | HTTP P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{path}}]-[{{method}}]-[{{code}}]-p99` |
+| gRPC QPS per Instance | `sum by(instance, pod, path, code) (rate(greptime_servers_grpc_requests_elapsed_count{}[$__rate_interval]))` | `timeseries` | gRPC QPS per Instance. | `prometheus` | `reqps` | `[{{instance}}]-[{{pod}}]-[{{path}}]-[{{code}}]` |
+| gRPC P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, path, code) (rate(greptime_servers_grpc_requests_elapsed_bucket{}[$__rate_interval])))` | `timeseries` | gRPC P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{path}}]-[{{method}}]-[{{code}}]-p99` |
+| MySQL QPS per Instance | `sum by(pod, instance)(rate(greptime_servers_mysql_query_elapsed_count{}[$__rate_interval]))` | `timeseries` | MySQL QPS per Instance. | `prometheus` | `reqps` | `[{{instance}}]-[{{pod}}]` |
+| MySQL P99 per Instance | `histogram_quantile(0.99, sum by(pod, instance, le) (rate(greptime_servers_mysql_query_elapsed_bucket{}[$__rate_interval])))` | `timeseries` | MySQL P99 per Instance. | `prometheus` | `s` | `[{{ instance }}]-[{{ pod }}]-p99` |
+| PostgreSQL QPS per Instance | `sum by(pod, instance)(rate(greptime_servers_postgres_query_elapsed_count{}[$__rate_interval]))` | `timeseries` | PostgreSQL QPS per Instance. | `prometheus` | `reqps` | `[{{instance}}]-[{{pod}}]` |
+| PostgreSQL P99 per Instance | `histogram_quantile(0.99, sum by(pod,instance,le) (rate(greptime_servers_postgres_query_elapsed_bucket{}[$__rate_interval])))` | `timeseries` | PostgreSQL P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-p99` |
+# Frontend to Datanode
+| Title | Query | Type | Description | Datasource | Unit | Legend Format |
+| --- | --- | --- | --- | --- | --- | --- |
+| Ingest Rows per Instance | `sum by(instance, pod)(rate(greptime_table_operator_ingest_rows{}[$__rate_interval]))` | `timeseries` | Ingestion rate by row as in each frontend | `prometheus` | `rowsps` | `[{{instance}}]-[{{pod}}]` |
+| Region Call QPS per Instance | `sum by(instance, pod, request_type) (rate(greptime_grpc_region_request_count{}[$__rate_interval]))` | `timeseries` | Region Call QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{request_type}}]` |
+| Region Call P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, request_type) (rate(greptime_grpc_region_request_bucket{}[$__rate_interval])))` | `timeseries` | Region Call P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{request_type}}]` |
+| Frontend Handle Bulk Insert Elapsed Time  | `sum by(instance, pod, stage) (rate(greptime_table_operator_handle_bulk_insert_sum[$__rate_interval]))/sum by(instance, pod, stage) (rate(greptime_table_operator_handle_bulk_insert_count[$__rate_interval]))`<br/>`histogram_quantile(0.99, sum by(instance, pod, stage, le) (rate(greptime_table_operator_handle_bulk_insert_bucket[$__rate_interval])))` | `timeseries` | Per-stage time for frontend to handle bulk insert requests | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{stage}}]-AVG` |
+# Mito Engine
+| Title | Query | Type | Description | Datasource | Unit | Legend Format |
+| --- | --- | --- | --- | --- | --- | --- |
+| Request OPS per Instance | `sum by(instance, pod, type) (rate(greptime_mito_handle_request_elapsed_count{}[$__rate_interval]))` | `timeseries` | Request QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{type}}]` |
+| Request P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, type) (rate(greptime_mito_handle_request_elapsed_bucket{}[$__rate_interval])))` | `timeseries` | Request P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{type}}]` |
+| Write Buffer per Instance | `greptime_mito_write_buffer_bytes{}` | `timeseries` | Write Buffer per Instance. | `prometheus` | `decbytes` | `[{{instance}}]-[{{pod}}]` |
+| Write Rows per Instance | `sum by (instance, pod) (rate(greptime_mito_write_rows_total{}[$__rate_interval]))` | `timeseries` | Ingestion size by row counts. | `prometheus` | `rowsps` | `[{{instance}}]-[{{pod}}]` |
+| Flush OPS per Instance | `sum by(instance, pod, reason) (rate(greptime_mito_flush_requests_total{}[$__rate_interval]))` | `timeseries` | Flush QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{reason}}]` |
+| Write Stall per Instance | `sum by(instance, pod) (greptime_mito_write_stall_total{})` | `timeseries` | Write Stall per Instance. | `prometheus` | -- | `[{{instance}}]-[{{pod}}]` |
+| Read Stage OPS per Instance | `sum by(instance, pod) (rate(greptime_mito_read_stage_elapsed_count{ stage="total"}[$__rate_interval]))` | `timeseries` | Read Stage OPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]` |
+| Read Stage P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, stage) (rate(greptime_mito_read_stage_elapsed_bucket{}[$__rate_interval])))` | `timeseries` | Read Stage P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{stage}}]` |
+| Write Stage P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, stage) (rate(greptime_mito_write_stage_elapsed_bucket{}[$__rate_interval])))` | `timeseries` | Write Stage P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{stage}}]` |
+| Compaction OPS per Instance | `sum by(instance, pod) (rate(greptime_mito_compaction_total_elapsed_count{}[$__rate_interval]))` | `timeseries` | Compaction OPS per Instance. | `prometheus` | `ops` | `[{{ instance }}]-[{{pod}}]` |
+| Compaction Elapsed Time per Instance by Stage | `histogram_quantile(0.99, sum by(instance, pod, le, stage) (rate(greptime_mito_compaction_stage_elapsed_bucket{}[$__rate_interval])))`<br/>`sum by(instance, pod, stage) (rate(greptime_mito_compaction_stage_elapsed_sum{}[$__rate_interval]))/sum by(instance, pod, stage) (rate(greptime_mito_compaction_stage_elapsed_count{}[$__rate_interval]))` | `timeseries` | Compaction latency by stage | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{stage}}]-p99` |
+| Compaction P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le,stage) (rate(greptime_mito_compaction_total_elapsed_bucket{}[$__rate_interval])))` | `timeseries` | Compaction P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{stage}}]-compaction` |
+| WAL write size | `histogram_quantile(0.95, sum by(le,instance, pod) (rate(raft_engine_write_size_bucket[$__rate_interval])))`<br/>`histogram_quantile(0.99, sum by(le,instance,pod) (rate(raft_engine_write_size_bucket[$__rate_interval])))`<br/>`sum by (instance, pod)(rate(raft_engine_write_size_sum[$__rate_interval]))` | `timeseries` | Write-ahead logs write size as bytes. This chart includes stats of p95 and p99 size by instance, total WAL write rate. | `prometheus` | `bytes` | `[{{instance}}]-[{{pod}}]-req-size-p95` |
+| Cached Bytes per Instance | `greptime_mito_cache_bytes{}` | `timeseries` | Cached Bytes per Instance. | `prometheus` | `decbytes` | `[{{instance}}]-[{{pod}}]-[{{type}}]` |
+| Inflight Compaction | `greptime_mito_inflight_compaction_count` | `timeseries` | Ongoing compaction task count | `prometheus` | `none` | `[{{instance}}]-[{{pod}}]` |
+| WAL sync duration seconds | `histogram_quantile(0.99, sum by(le, type, node, instance, pod) (rate(raft_engine_sync_log_duration_seconds_bucket[$__rate_interval])))` | `timeseries` | Raft engine (local disk) log store sync latency, p99 | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-p99` |
+| Log Store op duration seconds | `histogram_quantile(0.99, sum by(le,logstore,optype,instance, pod) (rate(greptime_logstore_op_elapsed_bucket[$__rate_interval])))` | `timeseries` | Write-ahead log operations latency at p99 | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{logstore}}]-[{{optype}}]-p99` |
+| Inflight Flush | `greptime_mito_inflight_flush_count` | `timeseries` | Ongoing flush task count | `prometheus` | `none` | `[{{instance}}]-[{{pod}}]` |
+| Compaction Input/Output Bytes | `sum by(instance, pod) (greptime_mito_compaction_input_bytes)`<br/>`sum by(instance, pod) (greptime_mito_compaction_output_bytes)` | `timeseries` | Compaction oinput output bytes | `prometheus` | `bytes` | `[{{instance}}]-[{{pod}}]-input` |
+| Region Worker Handle Bulk Insert Requests | `histogram_quantile(0.95, sum by(le,instance, stage, pod) (rate(greptime_region_worker_handle_write_bucket[$__rate_interval])))`<br/>`sum by(instance, stage, pod) (rate(greptime_region_worker_handle_write_sum[$__rate_interval]))/sum by(instance, stage, pod) (rate(greptime_region_worker_handle_write_count[$__rate_interval]))` | `timeseries` | Per-stage elapsed time for region worker to handle bulk insert region requests. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{stage}}]-P95` |
+| Active Series and Field Builders Count | `sum by(instance, pod) (greptime_mito_memtable_active_series_count)`<br/>`sum by(instance, pod) (greptime_mito_memtable_field_builder_count)` | `timeseries` | Compaction oinput output bytes | `prometheus` | `none` | `[{{instance}}]-[{{pod}}]-series` |
+| Region Worker Convert Requests | `histogram_quantile(0.95, sum by(le, instance, stage, pod) (rate(greptime_datanode_convert_region_request_bucket[$__rate_interval])))`<br/>`sum by(le,instance, stage, pod) (rate(greptime_datanode_convert_region_request_sum[$__rate_interval]))/sum by(le,instance, stage, pod) (rate(greptime_datanode_convert_region_request_count[$__rate_interval]))` | `timeseries` | Per-stage elapsed time for region worker to decode requests. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{stage}}]-P95` |
+# OpenDAL
+| Title | Query | Type | Description | Datasource | Unit | Legend Format |
+| --- | --- | --- | --- | --- | --- | --- |
+| QPS per Instance | `sum by(instance, pod, scheme, operation) (rate(opendal_operation_duration_seconds_count{}[$__rate_interval]))` | `timeseries` | QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]` |
+| Read QPS per Instance | `sum by(instance, pod, scheme) (rate(opendal_operation_duration_seconds_count{ operation="read"}[$__rate_interval]))` | `timeseries` | Read QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]` |
+| Read P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, scheme) (rate(opendal_operation_duration_seconds_bucket{operation="read"}[$__rate_interval])))` | `timeseries` | Read P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-{{scheme}}` |
+| Write QPS per Instance | `sum by(instance, pod, scheme) (rate(opendal_operation_duration_seconds_count{ operation="write"}[$__rate_interval]))` | `timeseries` | Write QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-{{scheme}}` |
+| Write P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, scheme) (rate(opendal_operation_duration_seconds_bucket{ operation="write"}[$__rate_interval])))` | `timeseries` | Write P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]` |
+| List QPS per Instance | `sum by(instance, pod, scheme) (rate(opendal_operation_duration_seconds_count{ operation="list"}[$__rate_interval]))` | `timeseries` | List QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]` |
+| List P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, scheme) (rate(opendal_operation_duration_seconds_bucket{ operation="list"}[$__rate_interval])))` | `timeseries` | List P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]` |
+| Other Requests per Instance | `sum by(instance, pod, scheme, operation) (rate(opendal_operation_duration_seconds_count{operation!~"read\|write\|list\|stat"}[$__rate_interval]))` | `timeseries` | Other Requests per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]` |
+| Other Request P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, scheme, operation) (rate(opendal_operation_duration_seconds_bucket{ operation!~"read\|write\|list"}[$__rate_interval])))` | `timeseries` | Other Request P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]` |
+| Opendal traffic | `sum by(instance, pod, scheme, operation) (rate(opendal_operation_bytes_sum{}[$__rate_interval]))` | `timeseries` | Total traffic as in bytes by instance and operation | `prometheus` | `decbytes` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]` |
+| OpenDAL errors per Instance | `sum by(instance, pod, scheme, operation, error) (rate(opendal_operation_errors_total{ error!="NotFound"}[$__rate_interval]))` | `timeseries` | OpenDAL error counts per Instance. | `prometheus` | -- | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]-[{{error}}]` |
+# Metasrv
+| Title | Query | Type | Description | Datasource | Unit | Legend Format |
+| --- | --- | --- | --- | --- | --- | --- |
+| Region migration datanode | `greptime_meta_region_migration_stat{datanode_type="src"}`<br/>`greptime_meta_region_migration_stat{datanode_type="desc"}` | `status-history` | Counter of region migration by source and destination | `prometheus` | -- | `from-datanode-{{datanode_id}}` |
+| Region migration error | `greptime_meta_region_migration_error` | `timeseries` | Counter of region migration error | `prometheus` | `none` | `{{pod}}-{{state}}-{{error_type}}` |
+| Datanode load | `greptime_datanode_load` | `timeseries` | Gauge of load information of each datanode, collected via heartbeat between datanode and metasrv. This information is for metasrv to schedule workloads. | `prometheus` | `binBps` | `Datanode-{{datanode_id}}-writeload` |
+| Rate of SQL Executions (RDS) | `rate(greptime_meta_rds_pg_sql_execute_elapsed_ms_count[$__rate_interval])` | `timeseries` | Displays the rate of SQL executions processed by the Meta service using the RDS backend. | `prometheus` | `none` | `{{pod}} {{op}} {{type}} {{result}} ` |
+| SQL Execution Latency (RDS) | `histogram_quantile(0.90, sum by(pod, op, type, result, le) (rate(greptime_meta_rds_pg_sql_execute_elapsed_ms_bucket[$__rate_interval])))` | `timeseries` | Measures the response time of SQL executions via the RDS backend.  | `prometheus` | `ms` | `{{pod}} {{op}} {{type}} {{result}} p90` |
+| Handler Execution Latency | `histogram_quantile(0.90, sum by(pod, le, name) (
+  rate(greptime_meta_handler_execute_bucket[$__rate_interval])
+))` | `timeseries` | Shows latency of Meta handlers by pod and handler name, useful for monitoring handler performance and detecting latency spikes.<br/> | `prometheus` | `s` | `{{pod}} {{name}} p90` |
+| Heartbeat Packet Size | `histogram_quantile(0.9, sum by(pod, le) (greptime_meta_heartbeat_stat_memory_size_bucket))` | `timeseries` | Shows p90 heartbeat message sizes, helping track network usage and identify anomalies in heartbeat payload.<br/> | `prometheus` | `bytes` | `{{pod}}` |
+| Meta Heartbeat Receive Rate | `rate(greptime_meta_heartbeat_rate[$__rate_interval])` | `timeseries` | Gauge of load information of each datanode, collected via heartbeat between datanode and metasrv. This information is for metasrv to schedule workloads. | `prometheus` | `s` | `{{pod}}` |
+| Meta KV Ops Latency | `histogram_quantile(0.99, sum by(pod, le, op, target) (greptime_meta_kv_request_elapsed_bucket))` | `timeseries` | Gauge of load information of each datanode, collected via heartbeat between datanode and metasrv. This information is for metasrv to schedule workloads. | `prometheus` | `s` | `{{pod}}-{{op}} p99` |
+| Rate of meta KV Ops | `rate(greptime_meta_kv_request_elapsed_count[$__rate_interval])` | `timeseries` | Gauge of load information of each datanode, collected via heartbeat between datanode and metasrv. This information is for metasrv to schedule workloads. | `prometheus` | `none` | `{{pod}}-{{op}} p99` |
+| DDL Latency | `histogram_quantile(0.9, sum by(le, pod, step) (greptime_meta_procedure_create_tables_bucket))`<br/>`histogram_quantile(0.9, sum by(le, pod, step) (greptime_meta_procedure_create_table))`<br/>`histogram_quantile(0.9, sum by(le, pod, step) (greptime_meta_procedure_create_view))`<br/>`histogram_quantile(0.9, sum by(le, pod, step) (greptime_meta_procedure_create_flow))`<br/>`histogram_quantile(0.9, sum by(le, pod, step) (greptime_meta_procedure_drop_table))`<br/>`histogram_quantile(0.9, sum by(le, pod, step) (greptime_meta_procedure_alter_table))` | `timeseries` | Gauge of load information of each datanode, collected via heartbeat between datanode and metasrv. This information is for metasrv to schedule workloads. | `prometheus` | `s` | `CreateLogicalTables-{{step}} p90` |
+# Flownode
+| Title | Query | Type | Description | Datasource | Unit | Legend Format |
+| --- | --- | --- | --- | --- | --- | --- |
+| Flow Ingest / Output Rate | `sum by(instance, pod, direction) (rate(greptime_flow_processed_rows[$__rate_interval]))` | `timeseries` | Flow Ingest / Output Rate. | `prometheus` | -- | `[{{pod}}]-[{{instance}}]-[{{direction}}]` |
+| Flow Ingest Latency | `histogram_quantile(0.95, sum(rate(greptime_flow_insert_elapsed_bucket[$__rate_interval])) by (le, instance, pod))`<br/>`histogram_quantile(0.99, sum(rate(greptime_flow_insert_elapsed_bucket[$__rate_interval])) by (le, instance, pod))` | `timeseries` | Flow Ingest Latency. | `prometheus` | -- | `[{{instance}}]-[{{pod}}]-p95` |
+| Flow Operation Latency | `histogram_quantile(0.95, sum(rate(greptime_flow_processing_time_bucket[$__rate_interval])) by (le,instance,pod,type))`<br/>`histogram_quantile(0.99, sum(rate(greptime_flow_processing_time_bucket[$__rate_interval])) by (le,instance,pod,type))` | `timeseries` | Flow Operation Latency. | `prometheus` | -- | `[{{instance}}]-[{{pod}}]-[{{type}}]-p95` |
+| Flow Buffer Size per Instance | `greptime_flow_input_buf_size` | `timeseries` | Flow Buffer Size per Instance. | `prometheus` | -- | `[{{instance}}]-[{{pod}]` |
+| Flow Processing Error per Instance | `sum by(instance,pod,code) (rate(greptime_flow_errors[$__rate_interval]))` | `timeseries` | Flow Processing Error per Instance. | `prometheus` | -- | `[{{instance}}]-[{{pod}}]-[{{code}}]` |

grafana/dashboards/metrics/standalone/dashboard.yaml

@@ -0,0 +1,958 @@
+groups:
+    - title: Overview
+      panels:
+        - title: Uptime
+          type: stat
+          description: The start time of GreptimeDB.
+          unit: s
+          queries:
+            - expr: time() - process_start_time_seconds
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: __auto
+        - title: Version
+          type: stat
+          description: GreptimeDB version.
+          queries:
+            - expr: SELECT pkg_version FROM information_schema.build_info
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+        - title: Total Ingestion Rate
+          type: stat
+          description: Total ingestion rate.
+          unit: rowsps
+          queries:
+            - expr: sum(rate(greptime_table_operator_ingest_rows[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: __auto
+        - title: Total Storage Size
+          type: stat
+          description: Total number of data file size.
+          unit: decbytes
+          queries:
+            - expr: select SUM(disk_size) from information_schema.region_statistics;
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+        - title: Total Rows
+          type: stat
+          description: Total number of data rows in the cluster. Calculated by sum of rows from each region.
+          unit: sishort
+          queries:
+            - expr: select SUM(region_rows) from information_schema.region_statistics;
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+        - title: Deployment
+          type: stat
+          description: The deployment topology of GreptimeDB.
+          queries:
+            - expr: SELECT count(*) as datanode FROM information_schema.cluster_info WHERE peer_type = 'DATANODE';
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+            - expr: SELECT count(*) as frontend FROM information_schema.cluster_info WHERE peer_type = 'FRONTEND';
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+            - expr: SELECT count(*) as metasrv FROM information_schema.cluster_info WHERE peer_type = 'METASRV';
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+            - expr: SELECT count(*) as flownode FROM information_schema.cluster_info WHERE peer_type = 'FLOWNODE';
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+        - title: Database Resources
+          type: stat
+          description: The number of the key resources in GreptimeDB.
+          queries:
+            - expr: SELECT COUNT(*) as databases FROM information_schema.schemata WHERE schema_name NOT IN ('greptime_private', 'information_schema')
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+            - expr: SELECT COUNT(*) as tables FROM information_schema.tables WHERE table_schema != 'information_schema'
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+            - expr: SELECT COUNT(region_id) as regions FROM information_schema.region_peers
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+            - expr: SELECT COUNT(*) as flows FROM information_schema.flows
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+        - title: Data Size
+          type: stat
+          description: The data size of wal/index/manifest in the GreptimeDB.
+          unit: decbytes
+          queries:
+            - expr: SELECT SUM(memtable_size) * 0.42825 as WAL FROM information_schema.region_statistics;
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+            - expr: SELECT SUM(index_size) as index FROM information_schema.region_statistics;
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+            - expr: SELECT SUM(manifest_size) as manifest FROM information_schema.region_statistics;
+              datasource:
+                type: mysql
+                uid: ${information_schema}
+    - title: Ingestion
+      panels:
+        - title: Total Ingestion Rate
+          type: timeseries
+          description: |
+            Total ingestion rate.
+
+            Here we listed 3 primary protocols:
+
+            - Prometheus remote write
+            - Greptime's gRPC API (when using our ingest SDK)
+            - Log ingestion http API
+          unit: rowsps
+          queries:
+            - expr: sum(rate(greptime_table_operator_ingest_rows{}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: ingestion
+        - title: Ingestion Rate by Type
+          type: timeseries
+          description: |
+            Total ingestion rate.
+
+            Here we listed 3 primary protocols:
+
+            - Prometheus remote write
+            - Greptime's gRPC API (when using our ingest SDK)
+            - Log ingestion http API
+          unit: rowsps
+          queries:
+            - expr: sum(rate(greptime_servers_http_logs_ingestion_counter[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: http-logs
+            - expr: sum(rate(greptime_servers_prometheus_remote_write_samples[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: prometheus-remote-write
+    - title: Queries
+      panels:
+        - title: Total Query Rate
+          type: timeseries
+          description: |-
+            Total rate of query API calls by protocol. This metric is collected from frontends.
+
+            Here we listed 3 main protocols:
+            - MySQL
+            - Postgres
+            - Prometheus API
+
+            Note that there are some other minor query APIs like /sql are not included
+          unit: reqps
+          queries:
+            - expr: sum (rate(greptime_servers_mysql_query_elapsed_count{}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: mysql
+            - expr: sum (rate(greptime_servers_postgres_query_elapsed_count{}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: pg
+            - expr: sum (rate(greptime_servers_http_promql_elapsed_counte{}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: promql
+    - title: Resources
+      panels:
+        - title: Datanode Memory per Instance
+          type: timeseries
+          description: Current memory usage by instance
+          unit: decbytes
+          queries:
+            - expr: sum(process_resident_memory_bytes{}) by (instance, pod)
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{ pod }}]'
+        - title: Datanode CPU Usage per Instance
+          type: timeseries
+          description: Current cpu usage by instance
+          unit: none
+          queries:
+            - expr: sum(rate(process_cpu_seconds_total{}[$__rate_interval]) * 1000) by (instance, pod)
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{ instance }}]-[{{ pod }}]'
+        - title: Frontend Memory per Instance
+          type: timeseries
+          description: Current memory usage by instance
+          unit: decbytes
+          queries:
+            - expr: sum(process_resident_memory_bytes{}) by (instance, pod)
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{ instance }}]-[{{ pod }}]'
+        - title: Frontend CPU Usage per Instance
+          type: timeseries
+          description: Current cpu usage by instance
+          unit: none
+          queries:
+            - expr: sum(rate(process_cpu_seconds_total{}[$__rate_interval]) * 1000) by (instance, pod)
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{ instance }}]-[{{ pod }}]-cpu'
+        - title: Metasrv Memory per Instance
+          type: timeseries
+          description: Current memory usage by instance
+          unit: decbytes
+          queries:
+            - expr: sum(process_resident_memory_bytes{}) by (instance, pod)
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{ instance }}]-[{{ pod }}]-resident'
+        - title: Metasrv CPU Usage per Instance
+          type: timeseries
+          description: Current cpu usage by instance
+          unit: none
+          queries:
+            - expr: sum(rate(process_cpu_seconds_total{}[$__rate_interval]) * 1000) by (instance, pod)
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{ instance }}]-[{{ pod }}]'
+        - title: Flownode Memory per Instance
+          type: timeseries
+          description: Current memory usage by instance
+          unit: decbytes
+          queries:
+            - expr: sum(process_resident_memory_bytes{}) by (instance, pod)
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{ instance }}]-[{{ pod }}]'
+        - title: Flownode CPU Usage per Instance
+          type: timeseries
+          description: Current cpu usage by instance
+          unit: none
+          queries:
+            - expr: sum(rate(process_cpu_seconds_total{}[$__rate_interval]) * 1000) by (instance, pod)
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{ instance }}]-[{{ pod }}]'
+    - title: Frontend Requests
+      panels:
+        - title: HTTP QPS per Instance
+          type: timeseries
+          description: HTTP QPS per Instance.
+          unit: reqps
+          queries:
+            - expr: sum by(instance, pod, path, method, code) (rate(greptime_servers_http_requests_elapsed_count{path!~"/health|/metrics"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{path}}]-[{{method}}]-[{{code}}]'
+        - title: HTTP P99 per Instance
+          type: timeseries
+          description: HTTP P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(instance, pod, le, path, method, code) (rate(greptime_servers_http_requests_elapsed_bucket{path!~"/health|/metrics"}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{path}}]-[{{method}}]-[{{code}}]-p99'
+        - title: gRPC QPS per Instance
+          type: timeseries
+          description: gRPC QPS per Instance.
+          unit: reqps
+          queries:
+            - expr: sum by(instance, pod, path, code) (rate(greptime_servers_grpc_requests_elapsed_count{}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{path}}]-[{{code}}]'
+        - title: gRPC P99 per Instance
+          type: timeseries
+          description: gRPC P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(instance, pod, le, path, code) (rate(greptime_servers_grpc_requests_elapsed_bucket{}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{path}}]-[{{method}}]-[{{code}}]-p99'
+        - title: MySQL QPS per Instance
+          type: timeseries
+          description: MySQL QPS per Instance.
+          unit: reqps
+          queries:
+            - expr: sum by(pod, instance)(rate(greptime_servers_mysql_query_elapsed_count{}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]'
+        - title: MySQL P99 per Instance
+          type: timeseries
+          description: MySQL P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(pod, instance, le) (rate(greptime_servers_mysql_query_elapsed_bucket{}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{ instance }}]-[{{ pod }}]-p99'
+        - title: PostgreSQL QPS per Instance
+          type: timeseries
+          description: PostgreSQL QPS per Instance.
+          unit: reqps
+          queries:
+            - expr: sum by(pod, instance)(rate(greptime_servers_postgres_query_elapsed_count{}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]'
+        - title: PostgreSQL P99 per Instance
+          type: timeseries
+          description: PostgreSQL P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(pod,instance,le) (rate(greptime_servers_postgres_query_elapsed_bucket{}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-p99'
+    - title: Frontend to Datanode
+      panels:
+        - title: Ingest Rows per Instance
+          type: timeseries
+          description: Ingestion rate by row as in each frontend
+          unit: rowsps
+          queries:
+            - expr: sum by(instance, pod)(rate(greptime_table_operator_ingest_rows{}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]'
+        - title: Region Call QPS per Instance
+          type: timeseries
+          description: Region Call QPS per Instance.
+          unit: ops
+          queries:
+            - expr: sum by(instance, pod, request_type) (rate(greptime_grpc_region_request_count{}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{request_type}}]'
+        - title: Region Call P99 per Instance
+          type: timeseries
+          description: Region Call P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(instance, pod, le, request_type) (rate(greptime_grpc_region_request_bucket{}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{request_type}}]'
+        - title: 'Frontend Handle Bulk Insert Elapsed Time '
+          type: timeseries
+          description: Per-stage time for frontend to handle bulk insert requests
+          unit: s
+          queries:
+            - expr: sum by(instance, pod, stage) (rate(greptime_table_operator_handle_bulk_insert_sum[$__rate_interval]))/sum by(instance, pod, stage) (rate(greptime_table_operator_handle_bulk_insert_count[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{stage}}]-AVG'
+            - expr: histogram_quantile(0.99, sum by(instance, pod, stage, le) (rate(greptime_table_operator_handle_bulk_insert_bucket[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{stage}}]-P95'
+    - title: Mito Engine
+      panels:
+        - title: Request OPS per Instance
+          type: timeseries
+          description: Request QPS per Instance.
+          unit: ops
+          queries:
+            - expr: sum by(instance, pod, type) (rate(greptime_mito_handle_request_elapsed_count{}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{type}}]'
+        - title: Request P99 per Instance
+          type: timeseries
+          description: Request P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(instance, pod, le, type) (rate(greptime_mito_handle_request_elapsed_bucket{}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{type}}]'
+        - title: Write Buffer per Instance
+          type: timeseries
+          description: Write Buffer per Instance.
+          unit: decbytes
+          queries:
+            - expr: greptime_mito_write_buffer_bytes{}
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]'
+        - title: Write Rows per Instance
+          type: timeseries
+          description: Ingestion size by row counts.
+          unit: rowsps
+          queries:
+            - expr: sum by (instance, pod) (rate(greptime_mito_write_rows_total{}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]'
+        - title: Flush OPS per Instance
+          type: timeseries
+          description: Flush QPS per Instance.
+          unit: ops
+          queries:
+            - expr: sum by(instance, pod, reason) (rate(greptime_mito_flush_requests_total{}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{reason}}]'
+        - title: Write Stall per Instance
+          type: timeseries
+          description: Write Stall per Instance.
+          queries:
+            - expr: sum by(instance, pod) (greptime_mito_write_stall_total{})
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]'
+        - title: Read Stage OPS per Instance
+          type: timeseries
+          description: Read Stage OPS per Instance.
+          unit: ops
+          queries:
+            - expr: sum by(instance, pod) (rate(greptime_mito_read_stage_elapsed_count{ stage="total"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]'
+        - title: Read Stage P99 per Instance
+          type: timeseries
+          description: Read Stage P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(instance, pod, le, stage) (rate(greptime_mito_read_stage_elapsed_bucket{}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{stage}}]'
+        - title: Write Stage P99 per Instance
+          type: timeseries
+          description: Write Stage P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(instance, pod, le, stage) (rate(greptime_mito_write_stage_elapsed_bucket{}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{stage}}]'
+        - title: Compaction OPS per Instance
+          type: timeseries
+          description: Compaction OPS per Instance.
+          unit: ops
+          queries:
+            - expr: sum by(instance, pod) (rate(greptime_mito_compaction_total_elapsed_count{}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{ instance }}]-[{{pod}}]'
+        - title: Compaction Elapsed Time per Instance by Stage
+          type: timeseries
+          description: Compaction latency by stage
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(instance, pod, le, stage) (rate(greptime_mito_compaction_stage_elapsed_bucket{}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{stage}}]-p99'
+            - expr: sum by(instance, pod, stage) (rate(greptime_mito_compaction_stage_elapsed_sum{}[$__rate_interval]))/sum by(instance, pod, stage) (rate(greptime_mito_compaction_stage_elapsed_count{}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{stage}}]-avg'
+        - title: Compaction P99 per Instance
+          type: timeseries
+          description: Compaction P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(instance, pod, le,stage) (rate(greptime_mito_compaction_total_elapsed_bucket{}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{stage}}]-compaction'
+        - title: WAL write size
+          type: timeseries
+          description: Write-ahead logs write size as bytes. This chart includes stats of p95 and p99 size by instance, total WAL write rate.
+          unit: bytes
+          queries:
+            - expr: histogram_quantile(0.95, sum by(le,instance, pod) (rate(raft_engine_write_size_bucket[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-req-size-p95'
+            - expr: histogram_quantile(0.99, sum by(le,instance,pod) (rate(raft_engine_write_size_bucket[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-req-size-p99'
+            - expr: sum by (instance, pod)(rate(raft_engine_write_size_sum[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-throughput'
+        - title: Cached Bytes per Instance
+          type: timeseries
+          description: Cached Bytes per Instance.
+          unit: decbytes
+          queries:
+            - expr: greptime_mito_cache_bytes{}
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{type}}]'
+        - title: Inflight Compaction
+          type: timeseries
+          description: Ongoing compaction task count
+          unit: none
+          queries:
+            - expr: greptime_mito_inflight_compaction_count
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]'
+        - title: WAL sync duration seconds
+          type: timeseries
+          description: Raft engine (local disk) log store sync latency, p99
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(le, type, node, instance, pod) (rate(raft_engine_sync_log_duration_seconds_bucket[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-p99'
+        - title: Log Store op duration seconds
+          type: timeseries
+          description: Write-ahead log operations latency at p99
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(le,logstore,optype,instance, pod) (rate(greptime_logstore_op_elapsed_bucket[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{logstore}}]-[{{optype}}]-p99'
+        - title: Inflight Flush
+          type: timeseries
+          description: Ongoing flush task count
+          unit: none
+          queries:
+            - expr: greptime_mito_inflight_flush_count
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]'
+        - title: Compaction Input/Output Bytes
+          type: timeseries
+          description: Compaction oinput output bytes
+          unit: bytes
+          queries:
+            - expr: sum by(instance, pod) (greptime_mito_compaction_input_bytes)
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-input'
+            - expr: sum by(instance, pod) (greptime_mito_compaction_output_bytes)
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-output'
+        - title: Region Worker Handle Bulk Insert Requests
+          type: timeseries
+          description: Per-stage elapsed time for region worker to handle bulk insert region requests.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.95, sum by(le,instance, stage, pod) (rate(greptime_region_worker_handle_write_bucket[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{stage}}]-P95'
+            - expr: sum by(instance, stage, pod) (rate(greptime_region_worker_handle_write_sum[$__rate_interval]))/sum by(instance, stage, pod) (rate(greptime_region_worker_handle_write_count[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{stage}}]-AVG'
+        - title: Active Series and Field Builders Count
+          type: timeseries
+          description: Compaction oinput output bytes
+          unit: none
+          queries:
+            - expr: sum by(instance, pod) (greptime_mito_memtable_active_series_count)
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-series'
+            - expr: sum by(instance, pod) (greptime_mito_memtable_field_builder_count)
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-field_builders'
+        - title: Region Worker Convert Requests
+          type: timeseries
+          description: Per-stage elapsed time for region worker to decode requests.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.95, sum by(le, instance, stage, pod) (rate(greptime_datanode_convert_region_request_bucket[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{stage}}]-P95'
+            - expr: sum by(le,instance, stage, pod) (rate(greptime_datanode_convert_region_request_sum[$__rate_interval]))/sum by(le,instance, stage, pod) (rate(greptime_datanode_convert_region_request_count[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{stage}}]-AVG'
+    - title: OpenDAL
+      panels:
+        - title: QPS per Instance
+          type: timeseries
+          description: QPS per Instance.
+          unit: ops
+          queries:
+            - expr: sum by(instance, pod, scheme, operation) (rate(opendal_operation_duration_seconds_count{}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]'
+        - title: Read QPS per Instance
+          type: timeseries
+          description: Read QPS per Instance.
+          unit: ops
+          queries:
+            - expr: sum by(instance, pod, scheme) (rate(opendal_operation_duration_seconds_count{ operation="read"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{scheme}}]'
+        - title: Read P99 per Instance
+          type: timeseries
+          description: Read P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(instance, pod, le, scheme) (rate(opendal_operation_duration_seconds_bucket{operation="read"}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-{{scheme}}'
+        - title: Write QPS per Instance
+          type: timeseries
+          description: Write QPS per Instance.
+          unit: ops
+          queries:
+            - expr: sum by(instance, pod, scheme) (rate(opendal_operation_duration_seconds_count{ operation="write"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-{{scheme}}'
+        - title: Write P99 per Instance
+          type: timeseries
+          description: Write P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(instance, pod, le, scheme) (rate(opendal_operation_duration_seconds_bucket{ operation="write"}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{scheme}}]'
+        - title: List QPS per Instance
+          type: timeseries
+          description: List QPS per Instance.
+          unit: ops
+          queries:
+            - expr: sum by(instance, pod, scheme) (rate(opendal_operation_duration_seconds_count{ operation="list"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{scheme}}]'
+        - title: List P99 per Instance
+          type: timeseries
+          description: List P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(instance, pod, le, scheme) (rate(opendal_operation_duration_seconds_bucket{ operation="list"}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{scheme}}]'
+        - title: Other Requests per Instance
+          type: timeseries
+          description: Other Requests per Instance.
+          unit: ops
+          queries:
+            - expr: sum by(instance, pod, scheme, operation) (rate(opendal_operation_duration_seconds_count{operation!~"read|write|list|stat"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]'
+        - title: Other Request P99 per Instance
+          type: timeseries
+          description: Other Request P99 per Instance.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(instance, pod, le, scheme, operation) (rate(opendal_operation_duration_seconds_bucket{ operation!~"read|write|list"}[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]'
+        - title: Opendal traffic
+          type: timeseries
+          description: Total traffic as in bytes by instance and operation
+          unit: decbytes
+          queries:
+            - expr: sum by(instance, pod, scheme, operation) (rate(opendal_operation_bytes_sum{}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]'
+        - title: OpenDAL errors per Instance
+          type: timeseries
+          description: OpenDAL error counts per Instance.
+          queries:
+            - expr: sum by(instance, pod, scheme, operation, error) (rate(opendal_operation_errors_total{ error!="NotFound"}[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]-[{{error}}]'
+    - title: Metasrv
+      panels:
+        - title: Region migration datanode
+          type: status-history
+          description: Counter of region migration by source and destination
+          queries:
+            - expr: greptime_meta_region_migration_stat{datanode_type="src"}
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: from-datanode-{{datanode_id}}
+            - expr: greptime_meta_region_migration_stat{datanode_type="desc"}
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: to-datanode-{{datanode_id}}
+        - title: Region migration error
+          type: timeseries
+          description: Counter of region migration error
+          unit: none
+          queries:
+            - expr: greptime_meta_region_migration_error
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '{{pod}}-{{state}}-{{error_type}}'
+        - title: Datanode load
+          type: timeseries
+          description: Gauge of load information of each datanode, collected via heartbeat between datanode and metasrv. This information is for metasrv to schedule workloads.
+          unit: binBps
+          queries:
+            - expr: greptime_datanode_load
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: Datanode-{{datanode_id}}-writeload
+        - title: Rate of SQL Executions (RDS)
+          type: timeseries
+          description: Displays the rate of SQL executions processed by the Meta service using the RDS backend.
+          unit: none
+          queries:
+            - expr: rate(greptime_meta_rds_pg_sql_execute_elapsed_ms_count[$__rate_interval])
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '{{pod}} {{op}} {{type}} {{result}} '
+        - title: SQL Execution Latency (RDS)
+          type: timeseries
+          description: 'Measures the response time of SQL executions via the RDS backend. '
+          unit: ms
+          queries:
+            - expr: histogram_quantile(0.90, sum by(pod, op, type, result, le) (rate(greptime_meta_rds_pg_sql_execute_elapsed_ms_bucket[$__rate_interval])))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '{{pod}} {{op}} {{type}} {{result}} p90'
+        - title: Handler Execution Latency
+          type: timeseries
+          description: |
+            Shows latency of Meta handlers by pod and handler name, useful for monitoring handler performance and detecting latency spikes.
+          unit: s
+          queries:
+            - expr: |-
+                histogram_quantile(0.90, sum by(pod, le, name) (
+                  rate(greptime_meta_handler_execute_bucket[$__rate_interval])
+                ))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '{{pod}} {{name}} p90'
+        - title: Heartbeat Packet Size
+          type: timeseries
+          description: |
+            Shows p90 heartbeat message sizes, helping track network usage and identify anomalies in heartbeat payload.
+          unit: bytes
+          queries:
+            - expr: histogram_quantile(0.9, sum by(pod, le) (greptime_meta_heartbeat_stat_memory_size_bucket))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '{{pod}}'
+        - title: Meta Heartbeat Receive Rate
+          type: timeseries
+          description: Gauge of load information of each datanode, collected via heartbeat between datanode and metasrv. This information is for metasrv to schedule workloads.
+          unit: s
+          queries:
+            - expr: rate(greptime_meta_heartbeat_rate[$__rate_interval])
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '{{pod}}'
+        - title: Meta KV Ops Latency
+          type: timeseries
+          description: Gauge of load information of each datanode, collected via heartbeat between datanode and metasrv. This information is for metasrv to schedule workloads.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.99, sum by(pod, le, op, target) (greptime_meta_kv_request_elapsed_bucket))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '{{pod}}-{{op}} p99'
+        - title: Rate of meta KV Ops
+          type: timeseries
+          description: Gauge of load information of each datanode, collected via heartbeat between datanode and metasrv. This information is for metasrv to schedule workloads.
+          unit: none
+          queries:
+            - expr: rate(greptime_meta_kv_request_elapsed_count[$__rate_interval])
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '{{pod}}-{{op}} p99'
+        - title: DDL Latency
+          type: timeseries
+          description: Gauge of load information of each datanode, collected via heartbeat between datanode and metasrv. This information is for metasrv to schedule workloads.
+          unit: s
+          queries:
+            - expr: histogram_quantile(0.9, sum by(le, pod, step) (greptime_meta_procedure_create_tables_bucket))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: CreateLogicalTables-{{step}} p90
+            - expr: histogram_quantile(0.9, sum by(le, pod, step) (greptime_meta_procedure_create_table))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: CreateTable-{{step}} p90
+            - expr: histogram_quantile(0.9, sum by(le, pod, step) (greptime_meta_procedure_create_view))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: CreateView-{{step}} p90
+            - expr: histogram_quantile(0.9, sum by(le, pod, step) (greptime_meta_procedure_create_flow))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: CreateFlow-{{step}} p90
+            - expr: histogram_quantile(0.9, sum by(le, pod, step) (greptime_meta_procedure_drop_table))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: DropTable-{{step}} p90
+            - expr: histogram_quantile(0.9, sum by(le, pod, step) (greptime_meta_procedure_alter_table))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: AlterTable-{{step}} p90
+    - title: Flownode
+      panels:
+        - title: Flow Ingest / Output Rate
+          type: timeseries
+          description: Flow Ingest / Output Rate.
+          queries:
+            - expr: sum by(instance, pod, direction) (rate(greptime_flow_processed_rows[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{pod}}]-[{{instance}}]-[{{direction}}]'
+        - title: Flow Ingest Latency
+          type: timeseries
+          description: Flow Ingest Latency.
+          queries:
+            - expr: histogram_quantile(0.95, sum(rate(greptime_flow_insert_elapsed_bucket[$__rate_interval])) by (le, instance, pod))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-p95'
+            - expr: histogram_quantile(0.99, sum(rate(greptime_flow_insert_elapsed_bucket[$__rate_interval])) by (le, instance, pod))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-p99'
+        - title: Flow Operation Latency
+          type: timeseries
+          description: Flow Operation Latency.
+          queries:
+            - expr: histogram_quantile(0.95, sum(rate(greptime_flow_processing_time_bucket[$__rate_interval])) by (le,instance,pod,type))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{type}}]-p95'
+            - expr: histogram_quantile(0.99, sum(rate(greptime_flow_processing_time_bucket[$__rate_interval])) by (le,instance,pod,type))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{type}}]-p99'
+        - title: Flow Buffer Size per Instance
+          type: timeseries
+          description: Flow Buffer Size per Instance.
+          queries:
+            - expr: greptime_flow_input_buf_size
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}]'
+        - title: Flow Processing Error per Instance
+          type: timeseries
+          description: Flow Processing Error per Instance.
+          queries:
+            - expr: sum by(instance,pod,code) (rate(greptime_flow_errors[$__rate_interval]))
+              datasource:
+                type: prometheus
+                uid: ${metrics}
+              legendFormat: '[{{instance}}]-[{{pod}}]-[{{code}}]'

grafana/scripts/check.sh

@@ -0,0 +1,54 @@
+#!/usr/bin/env bash
+
+DASHBOARD_DIR=${1:-grafana/dashboards/metrics}
+
+check_dashboard_description() {
+  for dashboard in $(find $DASHBOARD_DIR -name "*.json"); do
+    echo "Checking $dashboard description"
+
+    # Use jq to check for panels with empty or missing descriptions
+    invalid_panels=$(cat $dashboard | jq -r '
+      .panels[]
+    | select((.type == "stats" or .type == "timeseries") and (.description == "" or .description == null))')
+
+    # Check if any invalid panels were found
+    if [[ -n "$invalid_panels" ]]; then
+      echo "Error: The following panels have empty or missing descriptions:"
+      echo "$invalid_panels"
+      exit 1
+    else
+      echo "All panels with type 'stats' or 'timeseries' have valid descriptions."
+    fi
+  done
+}
+
+check_dashboards_generation() {
+  ./grafana/scripts/gen-dashboards.sh
+
+  if [[ -n "$(git diff --name-only grafana/dashboards/metrics)" ]]; then
+    echo "Error: The dashboards are not generated correctly. You should execute the `make dashboards` command."
+    exit 1
+  fi
+}
+
+check_datasource() {
+  for dashboard in $(find $DASHBOARD_DIR -name "*.json"); do
+    echo "Checking $dashboard datasource"
+    jq -r '.panels[] | select(.type != "row") | .targets[] | [.datasource.type, .datasource.uid] | @tsv' $dashboard | while read -r type uid; do
+    # if the datasource is prometheus, check if the uid is ${metrics}
+    if [[ "$type" == "prometheus" && "$uid" != "\${metrics}" ]]; then
+      echo "Error: The datasource uid of $dashboard is not valid. It should be \${metrics}, got $uid"
+      exit 1
+    fi
+    # if the datasource is mysql, check if the uid is ${information_schema}
+    if [[ "$type" == "mysql" && "$uid" != "\${information_schema}" ]]; then
+      echo "Error: The datasource uid of $dashboard is not valid. It should be \${information_schema}, got $uid"
+      exit 1
+    fi
+    done
+  done
+}
+
+check_dashboards_generation
+check_dashboard_description
+check_datasource

grafana/scripts/gen-dashboards.sh

@@ -0,0 +1,25 @@
+#! /usr/bin/env bash
+
+CLUSTER_DASHBOARD_DIR=${1:-grafana/dashboards/metrics/cluster}
+STANDALONE_DASHBOARD_DIR=${2:-grafana/dashboards/metrics/standalone}
+DAC_IMAGE=ghcr.io/zyy17/dac:20250423-522bd35
+
+remove_instance_filters() {
+  # Remove the instance filters for the standalone dashboards.
+  sed -E 's/instance=~\\"(\$datanode|\$frontend|\$metasrv|\$flownode)\\",?//g' "$CLUSTER_DASHBOARD_DIR/dashboard.json" > "$STANDALONE_DASHBOARD_DIR/dashboard.json"
+}
+
+generate_intermediate_dashboards_and_docs() {
+  docker run -v ${PWD}:/greptimedb --rm ${DAC_IMAGE} \
+    -i /greptimedb/$CLUSTER_DASHBOARD_DIR/dashboard.json \
+    -o /greptimedb/$CLUSTER_DASHBOARD_DIR/dashboard.yaml \
+    -m /greptimedb/$CLUSTER_DASHBOARD_DIR/dashboard.md
+
+  docker run -v ${PWD}:/greptimedb --rm ${DAC_IMAGE} \
+    -i /greptimedb/$STANDALONE_DASHBOARD_DIR/dashboard.json \
+    -o /greptimedb/$STANDALONE_DASHBOARD_DIR/dashboard.yaml \
+    -m /greptimedb/$STANDALONE_DASHBOARD_DIR/dashboard.md
+}
+
+remove_instance_filters
+generate_intermediate_dashboards_and_docs

grafana/summary.sh

@@ -1,11 +0,0 @@
-#!/usr/bin/env bash
-
-BASEDIR=$(dirname "$0")
-echo '| Title | Description | Expressions |
-|---|---|---|'
-
-cat $BASEDIR/greptimedb-cluster.json | jq -r '
-  .panels |
-  map(select(.type == "stat" or .type == "timeseries")) |
-  .[] | "| \(.title) | \(.description | gsub("\n"; "<br>")) | \(.targets | map(.expr // .rawSql | "`\(.|gsub("\n"; "<br>"))`")  | join("<br>")) |"
-'

licenserc.toml

@@ -26,6 +26,13 @@ excludes = [
     "src/common/base/src/secrets.rs",
     "src/servers/src/repeated_field.rs",
     "src/servers/src/http/test_helpers.rs",
+    # enterprise
+    "src/common/meta/src/rpc/ddl/trigger.rs",
+    "src/operator/src/expr_helper/trigger.rs",
+    "src/sql/src/statements/create/trigger.rs",
+    "src/sql/src/statements/show/trigger.rs",
+    "src/sql/src/parsers/create_parser/trigger.rs",
+    "src/sql/src/parsers/show_parser/trigger.rs",
 ]
 
 [properties]

rust-toolchain.toml

@@ -1,2 +1,2 @@
 [toolchain]
-channel = "nightly-2024-12-25"
+channel = "nightly-2025-05-19"

src/api/src/helper.rs

@@ -514,6 +514,7 @@ fn query_request_type(request: &QueryRequest) -> &'static str {
         Some(Query::Sql(_)) => "query.sql",
         Some(Query::LogicalPlan(_)) => "query.logical_plan",
         Some(Query::PromRangeQuery(_)) => "query.prom_range",
+        Some(Query::InsertIntoPlan(_)) => "query.insert_into_plan",
         None => "query.empty",
     }
 }
@@ -1049,7 +1050,7 @@ pub fn value_to_grpc_value(value: Value) -> GrpcValue {
             Value::Int64(v) => Some(ValueData::I64Value(v)),
             Value::Float32(v) => Some(ValueData::F32Value(*v)),
             Value::Float64(v) => Some(ValueData::F64Value(*v)),
-            Value::String(v) => Some(ValueData::StringValue(v.as_utf8().to_string())),
+            Value::String(v) => Some(ValueData::StringValue(v.into_string())),
             Value::Binary(v) => Some(ValueData::BinaryValue(v.to_vec())),
             Value::Date(v) => Some(ValueData::DateValue(v.val())),
             Value::Timestamp(v) => Some(match v.unit() {

src/api/src/region.rs

@@ -22,6 +22,7 @@ use greptime_proto::v1::region::RegionResponse as RegionResponseV1;
 pub struct RegionResponse {
     pub affected_rows: AffectedRows,
     pub extensions: HashMap<String, Vec<u8>>,
+    pub metadata: Vec<u8>,
 }
 
 impl RegionResponse {
@@ -29,6 +30,7 @@ impl RegionResponse {
         Self {
             affected_rows: region_response.affected_rows as _,
             extensions: region_response.extensions,
+            metadata: region_response.metadata,
         }
     }
 
@@ -37,6 +39,16 @@ impl RegionResponse {
         Self {
             affected_rows,
             extensions: Default::default(),
+            metadata: Vec::new(),
+        }
+    }
+
+    /// Creates one response with metadata.
+    pub fn from_metadata(metadata: Vec<u8>) -> Self {
+        Self {
+            affected_rows: 0,
+            extensions: Default::default(),
+            metadata,
         }
     }
 }

src/api/src/v1/column_def.rs

@@ -24,7 +24,7 @@ use greptime_proto::v1::{
 };
 use snafu::ResultExt;
 
-use crate::error::{self, Result};
+use crate::error::{self, ConvertColumnDefaultConstraintSnafu, Result};
 use crate::helper::ColumnDataTypeWrapper;
 use crate::v1::{ColumnDef, ColumnOptions, SemanticType};
 
@@ -77,6 +77,48 @@ pub fn try_as_column_schema(column_def: &ColumnDef) -> Result<ColumnSchema> {
         })
 }
 
+/// Tries to construct a `ColumnDef` from the given `ColumnSchema`.
+///
+/// TODO(weny): Add tests for this function.
+pub fn try_as_column_def(column_schema: &ColumnSchema, is_primary_key: bool) -> Result<ColumnDef> {
+    let column_datatype =
+        ColumnDataTypeWrapper::try_from(column_schema.data_type.clone()).map(|w| w.to_parts())?;
+
+    let semantic_type = if column_schema.is_time_index() {
+        SemanticType::Timestamp
+    } else if is_primary_key {
+        SemanticType::Tag
+    } else {
+        SemanticType::Field
+    } as i32;
+    let comment = column_schema
+        .metadata()
+        .get(COMMENT_KEY)
+        .cloned()
+        .unwrap_or_default();
+
+    let default_constraint = match column_schema.default_constraint() {
+        None => vec![],
+        Some(v) => v
+            .clone()
+            .try_into()
+            .context(ConvertColumnDefaultConstraintSnafu {
+                column: &column_schema.name,
+            })?,
+    };
+    let options = options_from_column_schema(column_schema);
+    Ok(ColumnDef {
+        name: column_schema.name.clone(),
+        data_type: column_datatype.0 as i32,
+        is_nullable: column_schema.is_nullable(),
+        default_constraint,
+        semantic_type,
+        comment,
+        datatype_extension: column_datatype.1,
+        options,
+    })
+}
+
 /// Constructs a `ColumnOptions` from the given `ColumnSchema`.
 pub fn options_from_column_schema(column_schema: &ColumnSchema) -> Option<ColumnOptions> {
     let mut options = ColumnOptions::default();
@@ -226,18 +268,20 @@ mod tests {
         assert!(options.is_none());
 
         let mut schema = ColumnSchema::new("test", ConcreteDataType::string_datatype(), true)
-            .with_fulltext_options(FulltextOptions {
-                enable: true,
-                analyzer: FulltextAnalyzer::English,
-                case_sensitive: false,
-                backend: FulltextBackend::Bloom,
-            })
+            .with_fulltext_options(FulltextOptions::new_unchecked(
+                true,
+                FulltextAnalyzer::English,
+                false,
+                FulltextBackend::Bloom,
+                10240,
+                0.01,
+            ))
             .unwrap();
         schema.set_inverted_index(true);
         let options = options_from_column_schema(&schema).unwrap();
         assert_eq!(
             options.options.get(FULLTEXT_GRPC_KEY).unwrap(),
-            "{\"enable\":true,\"analyzer\":\"English\",\"case-sensitive\":false,\"backend\":\"bloom\"}"
+            "{\"enable\":true,\"analyzer\":\"English\",\"case-sensitive\":false,\"backend\":\"bloom\",\"granularity\":10240,\"false-positive-rate-in-10000\":100}"
         );
         assert_eq!(
             options.options.get(INVERTED_INDEX_GRPC_KEY).unwrap(),
@@ -247,16 +291,18 @@ mod tests {
 
     #[test]
     fn test_options_with_fulltext() {
-        let fulltext = FulltextOptions {
-            enable: true,
-            analyzer: FulltextAnalyzer::English,
-            case_sensitive: false,
-            backend: FulltextBackend::Bloom,
-        };
+        let fulltext = FulltextOptions::new_unchecked(
+            true,
+            FulltextAnalyzer::English,
+            false,
+            FulltextBackend::Bloom,
+            10240,
+            0.01,
+        );
         let options = options_from_fulltext(&fulltext).unwrap().unwrap();
         assert_eq!(
             options.options.get(FULLTEXT_GRPC_KEY).unwrap(),
-            "{\"enable\":true,\"analyzer\":\"English\",\"case-sensitive\":false,\"backend\":\"bloom\"}"
+            "{\"enable\":true,\"analyzer\":\"English\",\"case-sensitive\":false,\"backend\":\"bloom\",\"granularity\":10240,\"false-positive-rate-in-10000\":100}"
         );
     }
 

src/auth/src/common.rs

@@ -36,7 +36,7 @@ pub fn userinfo_by_name(username: Option<String>) -> UserInfoRef {
 }
 
 pub fn user_provider_from_option(opt: &String) -> Result<UserProviderRef> {
-    let (name, content) = opt.split_once(':').context(InvalidConfigSnafu {
+    let (name, content) = opt.split_once(':').with_context(|| InvalidConfigSnafu {
         value: opt.to_string(),
         msg: "UserProviderOption must be in format `<option>:<value>`",
     })?;
@@ -57,6 +57,24 @@ pub fn user_provider_from_option(opt: &String) -> Result<UserProviderRef> {
     }
 }
 
+pub fn static_user_provider_from_option(opt: &String) -> Result<StaticUserProvider> {
+    let (name, content) = opt.split_once(':').with_context(|| InvalidConfigSnafu {
+        value: opt.to_string(),
+        msg: "UserProviderOption must be in format `<option>:<value>`",
+    })?;
+    match name {
+        STATIC_USER_PROVIDER => {
+            let provider = StaticUserProvider::new(content)?;
+            Ok(provider)
+        }
+        _ => InvalidConfigSnafu {
+            value: name.to_string(),
+            msg: format!("Invalid UserProviderOption, expect only {STATIC_USER_PROVIDER}"),
+        }
+        .fail(),
+    }
+}
+
 type Username<'a> = &'a str;
 type HostOrIp<'a> = &'a str;
 

src/auth/src/error.rs

@@ -38,6 +38,14 @@ pub enum Error {
         location: Location,
     },
 
+    #[snafu(display("Failed to convert to utf8"))]
+    FromUtf8 {
+        #[snafu(source)]
+        error: std::string::FromUtf8Error,
+        #[snafu(implicit)]
+        location: Location,
+    },
+
     #[snafu(display("Authentication source failure"))]
     AuthBackend {
         #[snafu(implicit)]
@@ -85,7 +93,7 @@ impl ErrorExt for Error {
     fn status_code(&self) -> StatusCode {
         match self {
             Error::InvalidConfig { .. } => StatusCode::InvalidArguments,
-            Error::IllegalParam { .. } => StatusCode::InvalidArguments,
+            Error::IllegalParam { .. } | Error::FromUtf8 { .. } => StatusCode::InvalidArguments,
             Error::FileWatch { .. } => StatusCode::InvalidArguments,
             Error::InternalState { .. } => StatusCode::Unexpected,
             Error::Io { .. } => StatusCode::StorageUnavailable,

src/auth/src/lib.rs

@@ -22,10 +22,12 @@ mod user_provider;
 pub mod tests;
 
 pub use common::{
-    auth_mysql, user_provider_from_option, userinfo_by_name, HashedPassword, Identity, Password,
+    auth_mysql, static_user_provider_from_option, user_provider_from_option, userinfo_by_name,
+    HashedPassword, Identity, Password,
 };
 pub use permission::{PermissionChecker, PermissionReq, PermissionResp};
 pub use user_info::UserInfo;
+pub use user_provider::static_user_provider::StaticUserProvider;
 pub use user_provider::UserProvider;
 
 /// pub type alias

src/auth/src/user_provider/static_user_provider.rs

@@ -15,15 +15,15 @@
 use std::collections::HashMap;
 
 use async_trait::async_trait;
-use snafu::OptionExt;
+use snafu::{OptionExt, ResultExt};
 
-use crate::error::{InvalidConfigSnafu, Result};
+use crate::error::{FromUtf8Snafu, InvalidConfigSnafu, Result};
 use crate::user_provider::{authenticate_with_credential, load_credential_from_file};
 use crate::{Identity, Password, UserInfoRef, UserProvider};
 
 pub(crate) const STATIC_USER_PROVIDER: &str = "static_user_provider";
 
-pub(crate) struct StaticUserProvider {
+pub struct StaticUserProvider {
     users: HashMap<String, Vec<u8>>,
 }
 
@@ -60,6 +60,18 @@ impl StaticUserProvider {
                 .fail(),
         }
     }
+
+    /// Return a random username/password pair
+    /// This is useful for invoking from other components in the cluster
+    pub fn get_one_user_pwd(&self) -> Result<(String, String)> {
+        let kv = self.users.iter().next().context(InvalidConfigSnafu {
+            value: "",
+            msg: "Expect at least one pair of username and password",
+        })?;
+        let username = kv.0;
+        let pwd = String::from_utf8(kv.1.clone()).context(FromUtf8Snafu)?;
+        Ok((username.clone(), pwd))
+    }
 }
 
 #[async_trait]

src/catalog/Cargo.toml

@@ -17,8 +17,10 @@ arrow-schema.workspace = true
 async-stream.workspace = true
 async-trait.workspace = true
 bytes.workspace = true
+common-base.workspace = true
 common-catalog.workspace = true
 common-error.workspace = true
+common-frontend.workspace = true
 common-macro.workspace = true
 common-meta.workspace = true
 common-procedure.workspace = true
@@ -41,6 +43,8 @@ moka = { workspace = true, features = ["future", "sync"] }
 partition.workspace = true
 paste.workspace = true
 prometheus.workspace = true
+promql-parser.workspace = true
+rand.workspace = true
 rustc-hash.workspace = true
 serde_json.workspace = true
 session.workspace = true

src/catalog/src/error.rs

@@ -277,6 +277,26 @@ pub enum Error {
         #[snafu(implicit)]
         location: Location,
     },
+
+    #[snafu(display("Failed to invoke frontend services"))]
+    InvokeFrontend {
+        source: common_frontend::error::Error,
+        #[snafu(implicit)]
+        location: Location,
+    },
+
+    #[snafu(display("Meta client is not provided"))]
+    MetaClientMissing {
+        #[snafu(implicit)]
+        location: Location,
+    },
+
+    #[snafu(display("Failed to find frontend node: {}", addr))]
+    FrontendNotFound {
+        addr: String,
+        #[snafu(implicit)]
+        location: Location,
+    },
 }
 
 impl Error {
@@ -345,6 +365,10 @@ impl ErrorExt for Error {
             Error::GetViewCache { source, .. } | Error::GetTableCache { source, .. } => {
                 source.status_code()
             }
+            Error::InvokeFrontend { source, .. } => source.status_code(),
+            Error::FrontendNotFound { .. } | Error::MetaClientMissing { .. } => {
+                StatusCode::Unexpected
+            }
         }
     }
 

src/catalog/src/information_extension.rs

@@ -16,8 +16,8 @@ use api::v1::meta::ProcedureStatus;
 use common_error::ext::BoxedError;
 use common_meta::cluster::{ClusterInfo, NodeInfo};
 use common_meta::datanode::RegionStat;
-use common_meta::ddl::{ExecutorContext, ProcedureExecutor};
 use common_meta::key::flow::flow_state::FlowStat;
+use common_meta::procedure_executor::{ExecutorContext, ProcedureExecutor};
 use common_meta::rpc::procedure;
 use common_procedure::{ProcedureInfo, ProcedureState};
 use meta_client::MetaClientRef;

src/catalog/src/kvbackend/manager.rs

@@ -22,11 +22,13 @@ use common_catalog::consts::{
     PG_CATALOG_NAME,
 };
 use common_error::ext::BoxedError;
-use common_meta::cache::{LayeredCacheRegistryRef, ViewInfoCacheRef};
+use common_meta::cache::{
+    LayeredCacheRegistryRef, TableRoute, TableRouteCacheRef, ViewInfoCacheRef,
+};
 use common_meta::key::catalog_name::CatalogNameKey;
 use common_meta::key::flow::FlowMetadataManager;
 use common_meta::key::schema_name::SchemaNameKey;
-use common_meta::key::table_info::TableInfoValue;
+use common_meta::key::table_info::{TableInfoManager, TableInfoValue};
 use common_meta::key::table_name::TableNameKey;
 use common_meta::key::{TableMetadataManager, TableMetadataManagerRef};
 use common_meta::kv_backend::KvBackendRef;
@@ -37,6 +39,7 @@ use moka::sync::Cache;
 use partition::manager::{PartitionRuleManager, PartitionRuleManagerRef};
 use session::context::{Channel, QueryContext};
 use snafu::prelude::*;
+use store_api::metric_engine_consts::METRIC_ENGINE_NAME;
 use table::dist_table::DistTable;
 use table::metadata::TableId;
 use table::table::numbers::{NumbersTable, NUMBERS_TABLE_NAME};
@@ -51,6 +54,7 @@ use crate::error::{
 };
 use crate::information_schema::{InformationExtensionRef, InformationSchemaProvider};
 use crate::kvbackend::TableCacheRef;
+use crate::process_manager::ProcessManagerRef;
 use crate::system_schema::pg_catalog::PGCatalogProvider;
 use crate::system_schema::SystemSchemaProvider;
 use crate::CatalogManager;
@@ -84,6 +88,7 @@ impl KvBackendCatalogManager {
         backend: KvBackendRef,
         cache_registry: LayeredCacheRegistryRef,
         procedure_manager: Option<ProcedureManagerRef>,
+        process_manager: Option<ProcessManagerRef>,
     ) -> Arc<Self> {
         Arc::new_cyclic(|me| Self {
             information_extension,
@@ -102,12 +107,14 @@ impl KvBackendCatalogManager {
                     DEFAULT_CATALOG_NAME.to_string(),
                     me.clone(),
                     Arc::new(FlowMetadataManager::new(backend.clone())),
+                    process_manager.clone(),
                 )),
                 pg_catalog_provider: Arc::new(PGCatalogProvider::new(
                     DEFAULT_CATALOG_NAME.to_string(),
                     me.clone(),
                 )),
                 backend,
+                process_manager,
             },
             cache_registry,
             procedure_manager,
@@ -136,6 +143,61 @@ impl KvBackendCatalogManager {
     pub fn procedure_manager(&self) -> Option<ProcedureManagerRef> {
         self.procedure_manager.clone()
     }
+
+    // Override logical table's partition key indices with physical table's.
+    async fn override_logical_table_partition_key_indices(
+        table_route_cache: &TableRouteCacheRef,
+        table_info_manager: &TableInfoManager,
+        table: TableRef,
+    ) -> Result<TableRef> {
+        // If the table is not a metric table, return the table directly.
+        if table.table_info().meta.engine != METRIC_ENGINE_NAME {
+            return Ok(table);
+        }
+
+        if let Some(table_route_value) = table_route_cache
+            .get(table.table_info().table_id())
+            .await
+            .context(TableMetadataManagerSnafu)?
+            && let TableRoute::Logical(logical_route) = &*table_route_value
+            && let Some(physical_table_info_value) = table_info_manager
+                .get(logical_route.physical_table_id())
+                .await
+                .context(TableMetadataManagerSnafu)?
+        {
+            let mut new_table_info = (*table.table_info()).clone();
+
+            // Remap partition key indices from physical table to logical table
+            new_table_info.meta.partition_key_indices = physical_table_info_value
+                .table_info
+                .meta
+                .partition_key_indices
+                .iter()
+                .filter_map(|&physical_index| {
+                    // Get the column name from the physical table using the physical index
+                    physical_table_info_value
+                        .table_info
+                        .meta
+                        .schema
+                        .column_schemas
+                        .get(physical_index)
+                        .and_then(|physical_column| {
+                            // Find the corresponding index in the logical table schema
+                            new_table_info
+                                .meta
+                                .schema
+                                .column_index_by_name(physical_column.name.as_str())
+                        })
+                })
+                .collect();
+
+            let new_table = DistTable::table(Arc::new(new_table_info));
+
+            return Ok(new_table);
+        }
+
+        Ok(table)
+    }
 }
 
 #[async_trait::async_trait]
@@ -262,16 +324,28 @@ impl CatalogManager for KvBackendCatalogManager {
         let table_cache: TableCacheRef = self.cache_registry.get().context(CacheNotFoundSnafu {
             name: "table_cache",
         })?;
-        if let Some(table) = table_cache
+
+        let table = table_cache
             .get_by_ref(&TableName {
                 catalog_name: catalog_name.to_string(),
                 schema_name: schema_name.to_string(),
                 table_name: table_name.to_string(),
             })
             .await
-            .context(GetTableCacheSnafu)?
-        {
-            return Ok(Some(table));
+            .context(GetTableCacheSnafu)?;
+
+        if let Some(table) = table {
+            let table_route_cache: TableRouteCacheRef =
+                self.cache_registry.get().context(CacheNotFoundSnafu {
+                    name: "table_route_cache",
+                })?;
+            return Self::override_logical_table_partition_key_indices(
+                &table_route_cache,
+                self.table_metadata_manager.table_info_manager(),
+                table,
+            )
+            .await
+            .map(Some);
         }
 
         if channel == Channel::Postgres {
@@ -284,7 +358,7 @@ impl CatalogManager for KvBackendCatalogManager {
             }
         }
 
-        return Ok(None);
+        Ok(None)
     }
 
     async fn tables_by_ids(
@@ -336,8 +410,20 @@ impl CatalogManager for KvBackendCatalogManager {
         let catalog = catalog.to_string();
         let schema = schema.to_string();
         let semaphore = Arc::new(Semaphore::new(CONCURRENCY));
+        let table_route_cache: Result<TableRouteCacheRef> =
+            self.cache_registry.get().context(CacheNotFoundSnafu {
+                name: "table_route_cache",
+            });
 
         common_runtime::spawn_global(async move {
+            let table_route_cache = match table_route_cache {
+                Ok(table_route_cache) => table_route_cache,
+                Err(e) => {
+                    let _ = tx.send(Err(e)).await;
+                    return;
+                }
+            };
+
             let table_id_stream = metadata_manager
                 .table_name_manager()
                 .tables(&catalog, &schema)
@@ -364,6 +450,7 @@ impl CatalogManager for KvBackendCatalogManager {
                 let metadata_manager = metadata_manager.clone();
                 let tx = tx.clone();
                 let semaphore = semaphore.clone();
+                let table_route_cache = table_route_cache.clone();
                 common_runtime::spawn_global(async move {
                     // we don't explicitly close the semaphore so just ignore the potential error.
                     let _ = semaphore.acquire().await;
@@ -381,6 +468,16 @@ impl CatalogManager for KvBackendCatalogManager {
                     };
 
                     for table in table_info_values.into_values().map(build_table) {
+                        let table = if let Ok(table) = table {
+                            Self::override_logical_table_partition_key_indices(
+                                &table_route_cache,
+                                metadata_manager.table_info_manager(),
+                                table,
+                            )
+                            .await
+                        } else {
+                            table
+                        };
                         if tx.send(table).await.is_err() {
                             return;
                         }
@@ -419,6 +516,7 @@ struct SystemCatalog {
     information_schema_provider: Arc<InformationSchemaProvider>,
     pg_catalog_provider: Arc<PGCatalogProvider>,
     backend: KvBackendRef,
+    process_manager: Option<ProcessManagerRef>,
 }
 
 impl SystemCatalog {
@@ -486,6 +584,7 @@ impl SystemCatalog {
                         catalog.to_string(),
                         self.catalog_manager.clone(),
                         Arc::new(FlowMetadataManager::new(self.backend.clone())),
+                        self.process_manager.clone(),
                     ))
                 });
             information_schema_provider.table(table_name)

src/catalog/src/lib.rs

@@ -14,6 +14,7 @@
 
 #![feature(assert_matches)]
 #![feature(try_blocks)]
+#![feature(let_chains)]
 
 use std::any::Any;
 use std::fmt::{Debug, Formatter};
@@ -40,6 +41,7 @@ pub mod information_schema {
     pub use crate::system_schema::information_schema::*;
 }
 
+pub mod process_manager;
 pub mod table_source;
 
 #[async_trait::async_trait]

src/catalog/src/memory/manager.rs

@@ -356,6 +356,7 @@ impl MemoryCatalogManager {
             catalog,
             Arc::downgrade(self) as Weak<dyn CatalogManager>,
             Arc::new(FlowMetadataManager::new(Arc::new(MemoryKvBackend::new()))),
+            None, // we don't need ProcessManager on regions server.
         );
         let information_schema = information_schema_provider.tables().clone();
 

src/catalog/src/metrics.rs

@@ -34,4 +34,20 @@ lazy_static! {
         register_histogram!("greptime_catalog_kv_get", "catalog kv get").unwrap();
     pub static ref METRIC_CATALOG_KV_BATCH_GET: Histogram =
         register_histogram!("greptime_catalog_kv_batch_get", "catalog kv batch get").unwrap();
+
+    /// Count of running process in each catalog.
+    pub static ref PROCESS_LIST_COUNT: IntGaugeVec = register_int_gauge_vec!(
+        "greptime_process_list_count",
+        "Running process count per catalog",
+        &["catalog"]
+    )
+    .unwrap();
+
+    /// Count of killed process in each catalog.
+    pub static ref PROCESS_KILL_COUNT: IntCounterVec = register_int_counter_vec!(
+        "greptime_process_kill_count",
+        "Completed kill process requests count",
+        &["catalog"]
+    )
+    .unwrap();
 }

src/catalog/src/process_manager.rs

@@ -0,0 +1,639 @@
+// Copyright 2023 Greptime Team
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+use std::collections::hash_map::Entry;
+use std::collections::HashMap;
+use std::fmt::{Debug, Display, Formatter};
+use std::sync::atomic::{AtomicU32, Ordering};
+use std::sync::{Arc, RwLock};
+use std::time::{Duration, Instant, UNIX_EPOCH};
+
+use api::v1::frontend::{KillProcessRequest, ListProcessRequest, ProcessInfo};
+use common_base::cancellation::CancellationHandle;
+use common_frontend::selector::{FrontendSelector, MetaClientSelector};
+use common_frontend::slow_query_event::SlowQueryEvent;
+use common_telemetry::{debug, error, info, warn};
+use common_time::util::current_time_millis;
+use meta_client::MetaClientRef;
+use promql_parser::parser::EvalStmt;
+use rand::random;
+use session::context::QueryContextRef;
+use snafu::{ensure, OptionExt, ResultExt};
+use sql::statements::statement::Statement;
+use tokio::sync::mpsc::Sender;
+
+use crate::error;
+use crate::metrics::{PROCESS_KILL_COUNT, PROCESS_LIST_COUNT};
+
+pub type ProcessId = u32;
+pub type ProcessManagerRef = Arc<ProcessManager>;
+
+/// Query process manager.
+pub struct ProcessManager {
+    /// Local frontend server address,
+    server_addr: String,
+    /// Next process id for local queries.
+    next_id: AtomicU32,
+    /// Running process per catalog.
+    catalogs: RwLock<HashMap<String, HashMap<ProcessId, CancellableProcess>>>,
+    /// Frontend selector to locate frontend nodes.
+    frontend_selector: Option<MetaClientSelector>,
+}
+
+/// Represents a parsed query statement, functionally equivalent to [query::parser::QueryStatement].
+/// This enum is defined here to avoid cyclic dependencies with the query parser module.
+#[derive(Debug, Clone)]
+pub enum QueryStatement {
+    Sql(Statement),
+    Promql(EvalStmt),
+}
+
+impl Display for QueryStatement {
+    fn fmt(&self, f: &mut Formatter<'_>) -> std::fmt::Result {
+        match self {
+            QueryStatement::Sql(stmt) => write!(f, "{}", stmt),
+            QueryStatement::Promql(eval_stmt) => write!(f, "{}", eval_stmt),
+        }
+    }
+}
+
+impl ProcessManager {
+    /// Create a [ProcessManager] instance with server address and kv client.
+    pub fn new(server_addr: String, meta_client: Option<MetaClientRef>) -> Self {
+        let frontend_selector = meta_client.map(MetaClientSelector::new);
+        Self {
+            server_addr,
+            next_id: Default::default(),
+            catalogs: Default::default(),
+            frontend_selector,
+        }
+    }
+}
+
+impl ProcessManager {
+    /// Registers a submitted query. Use the provided id if present.
+    #[must_use]
+    pub fn register_query(
+        self: &Arc<Self>,
+        catalog: String,
+        schemas: Vec<String>,
+        query: String,
+        client: String,
+        query_id: Option<ProcessId>,
+        _slow_query_timer: Option<SlowQueryTimer>,
+    ) -> Ticket {
+        let id = query_id.unwrap_or_else(|| self.next_id.fetch_add(1, Ordering::Relaxed));
+        let process = ProcessInfo {
+            id,
+            catalog: catalog.clone(),
+            schemas,
+            query,
+            start_timestamp: current_time_millis(),
+            client,
+            frontend: self.server_addr.clone(),
+        };
+        let cancellation_handle = Arc::new(CancellationHandle::default());
+        let cancellable_process = CancellableProcess::new(cancellation_handle.clone(), process);
+
+        self.catalogs
+            .write()
+            .unwrap()
+            .entry(catalog.clone())
+            .or_default()
+            .insert(id, cancellable_process);
+
+        Ticket {
+            catalog,
+            manager: self.clone(),
+            id,
+            cancellation_handle,
+            _slow_query_timer,
+        }
+    }
+
+    /// Generates the next process id.
+    pub fn next_id(&self) -> u32 {
+        self.next_id.fetch_add(1, Ordering::Relaxed)
+    }
+
+    /// De-register a query from process list.
+    pub fn deregister_query(&self, catalog: String, id: ProcessId) {
+        if let Entry::Occupied(mut o) = self.catalogs.write().unwrap().entry(catalog) {
+            let process = o.get_mut().remove(&id);
+            debug!("Deregister process: {:?}", process);
+            if o.get().is_empty() {
+                o.remove();
+            }
+        }
+    }
+
+    /// List local running processes in given catalog.
+    pub fn local_processes(&self, catalog: Option<&str>) -> error::Result<Vec<ProcessInfo>> {
+        let catalogs = self.catalogs.read().unwrap();
+        let result = if let Some(catalog) = catalog {
+            if let Some(catalogs) = catalogs.get(catalog) {
+                catalogs.values().map(|p| p.process.clone()).collect()
+            } else {
+                vec![]
+            }
+        } else {
+            catalogs
+                .values()
+                .flat_map(|v| v.values().map(|p| p.process.clone()))
+                .collect()
+        };
+        Ok(result)
+    }
+
+    pub async fn list_all_processes(
+        &self,
+        catalog: Option<&str>,
+    ) -> error::Result<Vec<ProcessInfo>> {
+        let mut processes = vec![];
+        if let Some(remote_frontend_selector) = self.frontend_selector.as_ref() {
+            let frontends = remote_frontend_selector
+                .select(|node| node.peer.addr != self.server_addr)
+                .await
+                .context(error::InvokeFrontendSnafu)?;
+            for mut f in frontends {
+                let result = f
+                    .list_process(ListProcessRequest {
+                        catalog: catalog.unwrap_or_default().to_string(),
+                    })
+                    .await
+                    .context(error::InvokeFrontendSnafu);
+                match result {
+                    Ok(resp) => {
+                        processes.extend(resp.processes);
+                    }
+                    Err(e) => {
+                        warn!(e; "Skipping failing node: {:?}", f)
+                    }
+                }
+            }
+        }
+        processes.extend(self.local_processes(catalog)?);
+        Ok(processes)
+    }
+
+    /// Kills query with provided catalog and id.
+    pub async fn kill_process(
+        &self,
+        server_addr: String,
+        catalog: String,
+        id: ProcessId,
+    ) -> error::Result<bool> {
+        if server_addr == self.server_addr {
+            self.kill_local_process(catalog, id).await
+        } else {
+            let mut nodes = self
+                .frontend_selector
+                .as_ref()
+                .context(error::MetaClientMissingSnafu)?
+                .select(|node| node.peer.addr == server_addr)
+                .await
+                .context(error::InvokeFrontendSnafu)?;
+            ensure!(
+                !nodes.is_empty(),
+                error::FrontendNotFoundSnafu { addr: server_addr }
+            );
+
+            let request = KillProcessRequest {
+                server_addr,
+                catalog,
+                process_id: id,
+            };
+            nodes[0]
+                .kill_process(request)
+                .await
+                .context(error::InvokeFrontendSnafu)?;
+            Ok(true)
+        }
+    }
+
+    /// Kills local query with provided catalog and id.
+    pub async fn kill_local_process(&self, catalog: String, id: ProcessId) -> error::Result<bool> {
+        if let Some(catalogs) = self.catalogs.write().unwrap().get_mut(&catalog) {
+            if let Some(process) = catalogs.remove(&id) {
+                process.handle.cancel();
+                info!(
+                    "Killed process, catalog: {}, id: {:?}",
+                    process.process.catalog, process.process.id
+                );
+                PROCESS_KILL_COUNT.with_label_values(&[&catalog]).inc();
+                Ok(true)
+            } else {
+                debug!("Failed to kill process, id not found: {}", id);
+                Ok(false)
+            }
+        } else {
+            debug!("Failed to kill process, catalog not found: {}", catalog);
+            Ok(false)
+        }
+    }
+}
+
+pub struct Ticket {
+    pub(crate) catalog: String,
+    pub(crate) manager: ProcessManagerRef,
+    pub(crate) id: ProcessId,
+    pub cancellation_handle: Arc<CancellationHandle>,
+    _slow_query_timer: Option<SlowQueryTimer>,
+}
+
+impl Drop for Ticket {
+    fn drop(&mut self) {
+        self.manager
+            .deregister_query(std::mem::take(&mut self.catalog), self.id);
+    }
+}
+
+struct CancellableProcess {
+    handle: Arc<CancellationHandle>,
+    process: ProcessInfo,
+}
+
+impl Drop for CancellableProcess {
+    fn drop(&mut self) {
+        PROCESS_LIST_COUNT
+            .with_label_values(&[&self.process.catalog])
+            .dec();
+    }
+}
+
+impl CancellableProcess {
+    fn new(handle: Arc<CancellationHandle>, process: ProcessInfo) -> Self {
+        PROCESS_LIST_COUNT
+            .with_label_values(&[&process.catalog])
+            .inc();
+        Self { handle, process }
+    }
+}
+
+impl Debug for CancellableProcess {
+    fn fmt(&self, f: &mut Formatter<'_>) -> std::fmt::Result {
+        f.debug_struct("CancellableProcess")
+            .field("cancelled", &self.handle.is_cancelled())
+            .field("process", &self.process)
+            .finish()
+    }
+}
+
+/// SlowQueryTimer is used to log slow query when it's dropped.
+/// In drop(), it will check if the query is slow and send the slow query event to the handler.
+pub struct SlowQueryTimer {
+    start: Instant,
+    stmt: QueryStatement,
+    query_ctx: QueryContextRef,
+    threshold: Option<Duration>,
+    sample_ratio: Option<f64>,
+    tx: Sender<SlowQueryEvent>,
+}
+
+impl SlowQueryTimer {
+    pub fn new(
+        stmt: QueryStatement,
+        query_ctx: QueryContextRef,
+        threshold: Option<Duration>,
+        sample_ratio: Option<f64>,
+        tx: Sender<SlowQueryEvent>,
+    ) -> Self {
+        Self {
+            start: Instant::now(),
+            stmt,
+            query_ctx,
+            threshold,
+            sample_ratio,
+            tx,
+        }
+    }
+}
+
+impl SlowQueryTimer {
+    fn send_slow_query_event(&self, elapsed: Duration, threshold: Duration) {
+        let mut slow_query_event = SlowQueryEvent {
+            cost: elapsed.as_millis() as u64,
+            threshold: threshold.as_millis() as u64,
+            query: "".to_string(),
+            query_ctx: self.query_ctx.clone(),
+
+            // The following fields are only used for PromQL queries.
+            is_promql: false,
+            promql_range: None,
+            promql_step: None,
+            promql_start: None,
+            promql_end: None,
+        };
+
+        match &self.stmt {
+            QueryStatement::Promql(stmt) => {
+                slow_query_event.is_promql = true;
+                slow_query_event.query = stmt.expr.to_string();
+                slow_query_event.promql_step = Some(stmt.interval.as_millis() as u64);
+
+                let start = stmt
+                    .start
+                    .duration_since(UNIX_EPOCH)
+                    .unwrap_or_default()
+                    .as_millis() as i64;
+
+                let end = stmt
+                    .end
+                    .duration_since(UNIX_EPOCH)
+                    .unwrap_or_default()
+                    .as_millis() as i64;
+
+                slow_query_event.promql_range = Some((end - start) as u64);
+                slow_query_event.promql_start = Some(start);
+                slow_query_event.promql_end = Some(end);
+            }
+            QueryStatement::Sql(stmt) => {
+                slow_query_event.query = stmt.to_string();
+            }
+        }
+
+        // Send SlowQueryEvent to the handler.
+        if let Err(e) = self.tx.try_send(slow_query_event) {
+            error!(e; "Failed to send slow query event");
+        }
+    }
+}
+
+impl Drop for SlowQueryTimer {
+    fn drop(&mut self) {
+        if let Some(threshold) = self.threshold {
+            // Calculate the elaspsed duration since the timer is created.
+            let elapsed = self.start.elapsed();
+            if elapsed > threshold {
+                if let Some(ratio) = self.sample_ratio {
+                    // Only capture a portion of slow queries based on sample_ratio.
+                    // Generate a random number in [0, 1) and compare it with sample_ratio.
+                    if ratio >= 1.0 || random::<f64>() <= ratio {
+                        self.send_slow_query_event(elapsed, threshold);
+                    }
+                } else {
+                    // Captures all slow queries if sample_ratio is not set.
+                    self.send_slow_query_event(elapsed, threshold);
+                }
+            }
+        }
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use std::sync::Arc;
+
+    use crate::process_manager::ProcessManager;
+
+    #[tokio::test]
+    async fn test_register_query() {
+        let process_manager = Arc::new(ProcessManager::new("127.0.0.1:8000".to_string(), None));
+        let ticket = process_manager.clone().register_query(
+            "public".to_string(),
+            vec!["test".to_string()],
+            "SELECT * FROM table".to_string(),
+            "".to_string(),
+            None,
+            None,
+        );
+
+        let running_processes = process_manager.local_processes(None).unwrap();
+        assert_eq!(running_processes.len(), 1);
+        assert_eq!(&running_processes[0].frontend, "127.0.0.1:8000");
+        assert_eq!(running_processes[0].id, ticket.id);
+        assert_eq!(&running_processes[0].query, "SELECT * FROM table");
+
+        drop(ticket);
+        assert_eq!(process_manager.local_processes(None).unwrap().len(), 0);
+    }
+
+    #[tokio::test]
+    async fn test_register_query_with_custom_id() {
+        let process_manager = Arc::new(ProcessManager::new("127.0.0.1:8000".to_string(), None));
+        let custom_id = 12345;
+
+        let ticket = process_manager.clone().register_query(
+            "public".to_string(),
+            vec!["test".to_string()],
+            "SELECT * FROM table".to_string(),
+            "client1".to_string(),
+            Some(custom_id),
+            None,
+        );
+
+        assert_eq!(ticket.id, custom_id);
+
+        let running_processes = process_manager.local_processes(None).unwrap();
+        assert_eq!(running_processes.len(), 1);
+        assert_eq!(running_processes[0].id, custom_id);
+        assert_eq!(&running_processes[0].client, "client1");
+    }
+
+    #[tokio::test]
+    async fn test_multiple_queries_same_catalog() {
+        let process_manager = Arc::new(ProcessManager::new("127.0.0.1:8000".to_string(), None));
+
+        let ticket1 = process_manager.clone().register_query(
+            "public".to_string(),
+            vec!["schema1".to_string()],
+            "SELECT * FROM table1".to_string(),
+            "client1".to_string(),
+            None,
+            None,
+        );
+
+        let ticket2 = process_manager.clone().register_query(
+            "public".to_string(),
+            vec!["schema2".to_string()],
+            "SELECT * FROM table2".to_string(),
+            "client2".to_string(),
+            None,
+            None,
+        );
+
+        let running_processes = process_manager.local_processes(Some("public")).unwrap();
+        assert_eq!(running_processes.len(), 2);
+
+        // Verify both processes are present
+        let ids: Vec<u32> = running_processes.iter().map(|p| p.id).collect();
+        assert!(ids.contains(&ticket1.id));
+        assert!(ids.contains(&ticket2.id));
+    }
+
+    #[tokio::test]
+    async fn test_multiple_catalogs() {
+        let process_manager = Arc::new(ProcessManager::new("127.0.0.1:8000".to_string(), None));
+
+        let _ticket1 = process_manager.clone().register_query(
+            "catalog1".to_string(),
+            vec!["schema1".to_string()],
+            "SELECT * FROM table1".to_string(),
+            "client1".to_string(),
+            None,
+            None,
+        );
+
+        let _ticket2 = process_manager.clone().register_query(
+            "catalog2".to_string(),
+            vec!["schema2".to_string()],
+            "SELECT * FROM table2".to_string(),
+            "client2".to_string(),
+            None,
+            None,
+        );
+
+        // Test listing processes for specific catalog
+        let catalog1_processes = process_manager.local_processes(Some("catalog1")).unwrap();
+        assert_eq!(catalog1_processes.len(), 1);
+        assert_eq!(&catalog1_processes[0].catalog, "catalog1");
+
+        let catalog2_processes = process_manager.local_processes(Some("catalog2")).unwrap();
+        assert_eq!(catalog2_processes.len(), 1);
+        assert_eq!(&catalog2_processes[0].catalog, "catalog2");
+
+        // Test listing all processes
+        let all_processes = process_manager.local_processes(None).unwrap();
+        assert_eq!(all_processes.len(), 2);
+    }
+
+    #[tokio::test]
+    async fn test_deregister_query() {
+        let process_manager = Arc::new(ProcessManager::new("127.0.0.1:8000".to_string(), None));
+
+        let ticket = process_manager.clone().register_query(
+            "public".to_string(),
+            vec!["test".to_string()],
+            "SELECT * FROM table".to_string(),
+            "client1".to_string(),
+            None,
+            None,
+        );
+        assert_eq!(process_manager.local_processes(None).unwrap().len(), 1);
+        process_manager.deregister_query("public".to_string(), ticket.id);
+        assert_eq!(process_manager.local_processes(None).unwrap().len(), 0);
+    }
+
+    #[tokio::test]
+    async fn test_cancellation_handle() {
+        let process_manager = Arc::new(ProcessManager::new("127.0.0.1:8000".to_string(), None));
+
+        let ticket = process_manager.clone().register_query(
+            "public".to_string(),
+            vec!["test".to_string()],
+            "SELECT * FROM table".to_string(),
+            "client1".to_string(),
+            None,
+            None,
+        );
+
+        assert!(!ticket.cancellation_handle.is_cancelled());
+        ticket.cancellation_handle.cancel();
+        assert!(ticket.cancellation_handle.is_cancelled());
+    }
+
+    #[tokio::test]
+    async fn test_kill_local_process() {
+        let process_manager = Arc::new(ProcessManager::new("127.0.0.1:8000".to_string(), None));
+
+        let ticket = process_manager.clone().register_query(
+            "public".to_string(),
+            vec!["test".to_string()],
+            "SELECT * FROM table".to_string(),
+            "client1".to_string(),
+            None,
+            None,
+        );
+        assert!(!ticket.cancellation_handle.is_cancelled());
+        let killed = process_manager
+            .kill_process(
+                "127.0.0.1:8000".to_string(),
+                "public".to_string(),
+                ticket.id,
+            )
+            .await
+            .unwrap();
+
+        assert!(killed);
+        assert_eq!(process_manager.local_processes(None).unwrap().len(), 0);
+    }
+
+    #[tokio::test]
+    async fn test_kill_nonexistent_process() {
+        let process_manager = Arc::new(ProcessManager::new("127.0.0.1:8000".to_string(), None));
+        let killed = process_manager
+            .kill_process("127.0.0.1:8000".to_string(), "public".to_string(), 999)
+            .await
+            .unwrap();
+        assert!(!killed);
+    }
+
+    #[tokio::test]
+    async fn test_kill_process_nonexistent_catalog() {
+        let process_manager = Arc::new(ProcessManager::new("127.0.0.1:8000".to_string(), None));
+        let killed = process_manager
+            .kill_process("127.0.0.1:8000".to_string(), "nonexistent".to_string(), 1)
+            .await
+            .unwrap();
+        assert!(!killed);
+    }
+
+    #[tokio::test]
+    async fn test_process_info_fields() {
+        let process_manager = Arc::new(ProcessManager::new("127.0.0.1:8000".to_string(), None));
+
+        let _ticket = process_manager.clone().register_query(
+            "test_catalog".to_string(),
+            vec!["schema1".to_string(), "schema2".to_string()],
+            "SELECT COUNT(*) FROM users WHERE age > 18".to_string(),
+            "test_client".to_string(),
+            Some(42),
+            None,
+        );
+
+        let processes = process_manager.local_processes(None).unwrap();
+        assert_eq!(processes.len(), 1);
+
+        let process = &processes[0];
+        assert_eq!(process.id, 42);
+        assert_eq!(&process.catalog, "test_catalog");
+        assert_eq!(process.schemas, vec!["schema1", "schema2"]);
+        assert_eq!(&process.query, "SELECT COUNT(*) FROM users WHERE age > 18");
+        assert_eq!(&process.client, "test_client");
+        assert_eq!(&process.frontend, "127.0.0.1:8000");
+        assert!(process.start_timestamp > 0);
+    }
+
+    #[tokio::test]
+    async fn test_ticket_drop_deregisters_process() {
+        let process_manager = Arc::new(ProcessManager::new("127.0.0.1:8000".to_string(), None));
+
+        {
+            let _ticket = process_manager.clone().register_query(
+                "public".to_string(),
+                vec!["test".to_string()],
+                "SELECT * FROM table".to_string(),
+                "client1".to_string(),
+                None,
+                None,
+            );
+
+            // Process should be registered
+            assert_eq!(process_manager.local_processes(None).unwrap().len(), 1);
+        } // ticket goes out of scope here
+
+        // Process should be automatically deregistered
+        assert_eq!(process_manager.local_processes(None).unwrap().len(), 0);
+    }
+}

src/catalog/src/system_schema/information_schema.rs

@@ -19,6 +19,7 @@ mod information_memory_table;
 pub mod key_column_usage;
 mod partitions;
 mod procedure_info;
+mod process_list;
 pub mod region_peers;
 mod region_statistics;
 mod runtime_metrics;
@@ -42,6 +43,7 @@ use common_recordbatch::SendableRecordBatchStream;
 use datatypes::schema::SchemaRef;
 use lazy_static::lazy_static;
 use paste::paste;
+use process_list::InformationSchemaProcessList;
 use store_api::storage::{ScanRequest, TableId};
 use table::metadata::TableType;
 use table::TableRef;
@@ -50,6 +52,7 @@ use views::InformationSchemaViews;
 
 use self::columns::InformationSchemaColumns;
 use crate::error::{Error, Result};
+use crate::process_manager::ProcessManagerRef;
 use crate::system_schema::information_schema::cluster_info::InformationSchemaClusterInfo;
 use crate::system_schema::information_schema::flows::InformationSchemaFlows;
 use crate::system_schema::information_schema::information_memory_table::get_schema_columns;
@@ -113,6 +116,7 @@ macro_rules! setup_memory_table {
 pub struct InformationSchemaProvider {
     catalog_name: String,
     catalog_manager: Weak<dyn CatalogManager>,
+    process_manager: Option<ProcessManagerRef>,
     flow_metadata_manager: Arc<FlowMetadataManager>,
     tables: HashMap<String, TableRef>,
 }
@@ -207,6 +211,10 @@ impl SystemSchemaProviderInner for InformationSchemaProvider {
                     self.catalog_manager.clone(),
                 ),
             ) as _),
+            PROCESS_LIST => self
+                .process_manager
+                .as_ref()
+                .map(|p| Arc::new(InformationSchemaProcessList::new(p.clone())) as _),
             _ => None,
         }
     }
@@ -217,11 +225,13 @@ impl InformationSchemaProvider {
         catalog_name: String,
         catalog_manager: Weak<dyn CatalogManager>,
         flow_metadata_manager: Arc<FlowMetadataManager>,
+        process_manager: Option<ProcessManagerRef>,
     ) -> Self {
         let mut provider = Self {
             catalog_name,
             catalog_manager,
             flow_metadata_manager,
+            process_manager,
             tables: HashMap::new(),
         };
 
@@ -277,6 +287,9 @@ impl InformationSchemaProvider {
             self.build_table(TABLE_CONSTRAINTS).unwrap(),
         );
         tables.insert(FLOWS.to_string(), self.build_table(FLOWS).unwrap());
+        if let Some(process_list) = self.build_table(PROCESS_LIST) {
+            tables.insert(PROCESS_LIST.to_string(), process_list);
+        }
         // Add memory tables
         for name in MEMORY_TABLES.iter() {
             tables.insert((*name).to_string(), self.build_table(name).expect(name));

src/catalog/src/system_schema/information_schema/process_list.rs

@@ -0,0 +1,189 @@
+// Copyright 2023 Greptime Team
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+use std::sync::Arc;
+
+use common_catalog::consts::INFORMATION_SCHEMA_PROCESS_LIST_TABLE_ID;
+use common_error::ext::BoxedError;
+use common_frontend::DisplayProcessId;
+use common_recordbatch::adapter::RecordBatchStreamAdapter;
+use common_recordbatch::{RecordBatch, SendableRecordBatchStream};
+use common_time::util::current_time_millis;
+use common_time::{Duration, Timestamp};
+use datafusion::physical_plan::stream::RecordBatchStreamAdapter as DfRecordBatchStreamAdapter;
+use datatypes::prelude::ConcreteDataType as CDT;
+use datatypes::scalars::ScalarVectorBuilder;
+use datatypes::schema::{ColumnSchema, Schema, SchemaRef};
+use datatypes::value::Value;
+use datatypes::vectors::{
+    DurationMillisecondVectorBuilder, StringVectorBuilder, TimestampMillisecondVectorBuilder,
+    VectorRef,
+};
+use snafu::ResultExt;
+use store_api::storage::{ScanRequest, TableId};
+
+use crate::error::{self, InternalSnafu};
+use crate::information_schema::Predicates;
+use crate::process_manager::ProcessManagerRef;
+use crate::system_schema::information_schema::InformationTable;
+
+/// Column names of `information_schema.process_list`
+const ID: &str = "id";
+const CATALOG: &str = "catalog";
+const SCHEMAS: &str = "schemas";
+const QUERY: &str = "query";
+const CLIENT: &str = "client";
+const FRONTEND: &str = "frontend";
+const START_TIMESTAMP: &str = "start_timestamp";
+const ELAPSED_TIME: &str = "elapsed_time";
+
+/// `information_schema.process_list` table implementation that tracks running
+/// queries in current cluster.
+pub struct InformationSchemaProcessList {
+    schema: SchemaRef,
+    process_manager: ProcessManagerRef,
+}
+
+impl InformationSchemaProcessList {
+    pub fn new(process_manager: ProcessManagerRef) -> Self {
+        Self {
+            schema: Self::schema(),
+            process_manager,
+        }
+    }
+
+    fn schema() -> SchemaRef {
+        Arc::new(Schema::new(vec![
+            ColumnSchema::new(ID, CDT::string_datatype(), false),
+            ColumnSchema::new(CATALOG, CDT::string_datatype(), false),
+            ColumnSchema::new(SCHEMAS, CDT::string_datatype(), false),
+            ColumnSchema::new(QUERY, CDT::string_datatype(), false),
+            ColumnSchema::new(CLIENT, CDT::string_datatype(), false),
+            ColumnSchema::new(FRONTEND, CDT::string_datatype(), false),
+            ColumnSchema::new(
+                START_TIMESTAMP,
+                CDT::timestamp_millisecond_datatype(),
+                false,
+            ),
+            ColumnSchema::new(ELAPSED_TIME, CDT::duration_millisecond_datatype(), false),
+        ]))
+    }
+}
+
+impl InformationTable for InformationSchemaProcessList {
+    fn table_id(&self) -> TableId {
+        INFORMATION_SCHEMA_PROCESS_LIST_TABLE_ID
+    }
+
+    fn table_name(&self) -> &'static str {
+        "process_list"
+    }
+
+    fn schema(&self) -> SchemaRef {
+        self.schema.clone()
+    }
+
+    fn to_stream(&self, request: ScanRequest) -> error::Result<SendableRecordBatchStream> {
+        let process_manager = self.process_manager.clone();
+        let stream = Box::pin(DfRecordBatchStreamAdapter::new(
+            self.schema.arrow_schema().clone(),
+            futures::stream::once(async move {
+                make_process_list(process_manager, request)
+                    .await
+                    .map(RecordBatch::into_df_record_batch)
+                    .map_err(|e| datafusion::error::DataFusionError::External(Box::new(e)))
+            }),
+        ));
+
+        Ok(Box::pin(
+            RecordBatchStreamAdapter::try_new(stream)
+                .map_err(BoxedError::new)
+                .context(InternalSnafu)?,
+        ))
+    }
+}
+
+/// Build running process list.
+async fn make_process_list(
+    process_manager: ProcessManagerRef,
+    request: ScanRequest,
+) -> error::Result<RecordBatch> {
+    let predicates = Predicates::from_scan_request(&Some(request));
+    let current_time = current_time_millis();
+    // todo(hl): find a way to extract user catalog to filter queries from other users.
+    let queries = process_manager.list_all_processes(None).await?;
+
+    let mut id_builder = StringVectorBuilder::with_capacity(queries.len());
+    let mut catalog_builder = StringVectorBuilder::with_capacity(queries.len());
+    let mut schemas_builder = StringVectorBuilder::with_capacity(queries.len());
+    let mut query_builder = StringVectorBuilder::with_capacity(queries.len());
+    let mut client_builder = StringVectorBuilder::with_capacity(queries.len());
+    let mut frontend_builder = StringVectorBuilder::with_capacity(queries.len());
+    let mut start_time_builder = TimestampMillisecondVectorBuilder::with_capacity(queries.len());
+    let mut elapsed_time_builder = DurationMillisecondVectorBuilder::with_capacity(queries.len());
+
+    for process in queries {
+        let display_id = DisplayProcessId {
+            server_addr: process.frontend.to_string(),
+            id: process.id,
+        }
+        .to_string();
+        let schemas = process.schemas.join(",");
+        let id = Value::from(display_id);
+        let catalog = Value::from(process.catalog);
+        let schemas = Value::from(schemas);
+        let query = Value::from(process.query);
+        let client = Value::from(process.client);
+        let frontend = Value::from(process.frontend);
+        let start_timestamp = Value::from(Timestamp::new_millisecond(process.start_timestamp));
+        let elapsed_time = Value::from(Duration::new_millisecond(
+            current_time - process.start_timestamp,
+        ));
+        let row = [
+            (ID, &id),
+            (CATALOG, &catalog),
+            (SCHEMAS, &schemas),
+            (QUERY, &query),
+            (CLIENT, &client),
+            (FRONTEND, &frontend),
+            (START_TIMESTAMP, &start_timestamp),
+            (ELAPSED_TIME, &elapsed_time),
+        ];
+        if predicates.eval(&row) {
+            id_builder.push(id.as_string().as_deref());
+            catalog_builder.push(catalog.as_string().as_deref());
+            schemas_builder.push(schemas.as_string().as_deref());
+            query_builder.push(query.as_string().as_deref());
+            client_builder.push(client.as_string().as_deref());
+            frontend_builder.push(frontend.as_string().as_deref());
+            start_time_builder.push(start_timestamp.as_timestamp().map(|t| t.value().into()));
+            elapsed_time_builder.push(elapsed_time.as_duration().map(|d| d.value().into()));
+        }
+    }
+
+    RecordBatch::new(
+        InformationSchemaProcessList::schema(),
+        vec![
+            Arc::new(id_builder.finish()) as VectorRef,
+            Arc::new(catalog_builder.finish()) as VectorRef,
+            Arc::new(schemas_builder.finish()) as VectorRef,
+            Arc::new(query_builder.finish()) as VectorRef,
+            Arc::new(client_builder.finish()) as VectorRef,
+            Arc::new(frontend_builder.finish()) as VectorRef,
+            Arc::new(start_time_builder.finish()) as VectorRef,
+            Arc::new(elapsed_time_builder.finish()) as VectorRef,
+        ],
+    )
+    .context(error::CreateRecordBatchSnafu)
+}

src/catalog/src/system_schema/information_schema/region_statistics.rs

@@ -40,6 +40,7 @@ const REGION_ID: &str = "region_id";
 const TABLE_ID: &str = "table_id";
 const REGION_NUMBER: &str = "region_number";
 const REGION_ROWS: &str = "region_rows";
+const WRITTEN_BYTES: &str = "written_bytes_since_open";
 const DISK_SIZE: &str = "disk_size";
 const MEMTABLE_SIZE: &str = "memtable_size";
 const MANIFEST_SIZE: &str = "manifest_size";
@@ -56,6 +57,7 @@ const INIT_CAPACITY: usize = 42;
 /// - `table_id`: The table id.
 /// - `region_number`: The region number.
 /// - `region_rows`: The number of rows in region.
+/// - `written_bytes_since_open`: The total bytes written of the region since region opened.
 /// - `memtable_size`: The memtable size in bytes.
 /// - `disk_size`: The approximate disk size in bytes.
 /// - `manifest_size`: The manifest size in bytes.
@@ -83,6 +85,7 @@ impl InformationSchemaRegionStatistics {
             ColumnSchema::new(TABLE_ID, ConcreteDataType::uint32_datatype(), false),
             ColumnSchema::new(REGION_NUMBER, ConcreteDataType::uint32_datatype(), false),
             ColumnSchema::new(REGION_ROWS, ConcreteDataType::uint64_datatype(), true),
+            ColumnSchema::new(WRITTEN_BYTES, ConcreteDataType::uint64_datatype(), true),
             ColumnSchema::new(DISK_SIZE, ConcreteDataType::uint64_datatype(), true),
             ColumnSchema::new(MEMTABLE_SIZE, ConcreteDataType::uint64_datatype(), true),
             ColumnSchema::new(MANIFEST_SIZE, ConcreteDataType::uint64_datatype(), true),
@@ -145,6 +148,7 @@ struct InformationSchemaRegionStatisticsBuilder {
     table_ids: UInt32VectorBuilder,
     region_numbers: UInt32VectorBuilder,
     region_rows: UInt64VectorBuilder,
+    written_bytes: UInt64VectorBuilder,
     disk_sizes: UInt64VectorBuilder,
     memtable_sizes: UInt64VectorBuilder,
     manifest_sizes: UInt64VectorBuilder,
@@ -163,6 +167,7 @@ impl InformationSchemaRegionStatisticsBuilder {
             table_ids: UInt32VectorBuilder::with_capacity(INIT_CAPACITY),
             region_numbers: UInt32VectorBuilder::with_capacity(INIT_CAPACITY),
             region_rows: UInt64VectorBuilder::with_capacity(INIT_CAPACITY),
+            written_bytes: UInt64VectorBuilder::with_capacity(INIT_CAPACITY),
             disk_sizes: UInt64VectorBuilder::with_capacity(INIT_CAPACITY),
             memtable_sizes: UInt64VectorBuilder::with_capacity(INIT_CAPACITY),
             manifest_sizes: UInt64VectorBuilder::with_capacity(INIT_CAPACITY),
@@ -193,6 +198,7 @@ impl InformationSchemaRegionStatisticsBuilder {
             (TABLE_ID, &Value::from(region_stat.id.table_id())),
             (REGION_NUMBER, &Value::from(region_stat.id.region_number())),
             (REGION_ROWS, &Value::from(region_stat.num_rows)),
+            (WRITTEN_BYTES, &Value::from(region_stat.written_bytes)),
             (DISK_SIZE, &Value::from(region_stat.approximate_bytes)),
             (MEMTABLE_SIZE, &Value::from(region_stat.memtable_size)),
             (MANIFEST_SIZE, &Value::from(region_stat.manifest_size)),
@@ -211,6 +217,7 @@ impl InformationSchemaRegionStatisticsBuilder {
         self.region_numbers
             .push(Some(region_stat.id.region_number()));
         self.region_rows.push(Some(region_stat.num_rows));
+        self.written_bytes.push(Some(region_stat.written_bytes));
         self.disk_sizes.push(Some(region_stat.approximate_bytes));
         self.memtable_sizes.push(Some(region_stat.memtable_size));
         self.manifest_sizes.push(Some(region_stat.manifest_size));
@@ -226,6 +233,7 @@ impl InformationSchemaRegionStatisticsBuilder {
             Arc::new(self.table_ids.finish()),
             Arc::new(self.region_numbers.finish()),
             Arc::new(self.region_rows.finish()),
+            Arc::new(self.written_bytes.finish()),
             Arc::new(self.disk_sizes.finish()),
             Arc::new(self.memtable_sizes.finish()),
             Arc::new(self.manifest_sizes.finish()),

src/catalog/src/system_schema/information_schema/table_names.rs

@@ -47,3 +47,4 @@ pub const VIEWS: &str = "views";
 pub const FLOWS: &str = "flows";
 pub const PROCEDURE_INFO: &str = "procedure_info";
 pub const REGION_STATISTICS: &str = "region_statistics";
+pub const PROCESS_LIST: &str = "process_list";

src/catalog/src/system_schema/pg_catalog/pg_namespace/oid_map.rs

@@ -84,12 +84,6 @@ mod tests {
         let key1 = "3178510";
         let key2 = "4215648";
 
-        // have collision
-        assert_eq!(
-            oid_map.hasher.hash_one(key1) as u32,
-            oid_map.hasher.hash_one(key2) as u32
-        );
-
         // insert them into oid_map
         let oid1 = oid_map.get_oid(key1);
         let oid2 = oid_map.get_oid(key2);

src/catalog/src/table_source.rs

@@ -27,7 +27,7 @@ use session::context::QueryContextRef;
 use snafu::{ensure, OptionExt, ResultExt};
 use table::metadata::TableType;
 use table::table::adapter::DfTableProviderAdapter;
-mod dummy_catalog;
+pub mod dummy_catalog;
 use dummy_catalog::DummyCatalogList;
 use table::TableRef;
 
@@ -328,6 +328,7 @@ mod tests {
             backend.clone(),
             layered_cache_registry,
             None,
+            None,
         );
         let table_metadata_manager = TableMetadataManager::new(backend);
         let mut view_info = common_meta::key::test_utils::new_test_table_info(1024, vec![]);

src/cli/Cargo.toml

@@ -5,13 +5,18 @@ edition.workspace = true
 license.workspace = true
 
 [features]
-pg_kvbackend = ["common-meta/pg_kvbackend"]
-mysql_kvbackend = ["common-meta/mysql_kvbackend"]
+default = [
+    "pg_kvbackend",
+    "mysql_kvbackend",
+]
+pg_kvbackend = ["common-meta/pg_kvbackend", "meta-srv/pg_kvbackend"]
+mysql_kvbackend = ["common-meta/mysql_kvbackend", "meta-srv/mysql_kvbackend"]
 
 [lints]
 workspace = true
 
 [dependencies]
+async-stream.workspace = true
 async-trait.workspace = true
 auth.workspace = true
 base64.workspace = true
@@ -38,20 +43,17 @@ common-time.workspace = true
 common-version.workspace = true
 common-wal.workspace = true
 datatypes.workspace = true
-either = "1.8"
 etcd-client.workspace = true
 futures.workspace = true
 humantime.workspace = true
 meta-client.workspace = true
+meta-srv.workspace = true
 nu-ansi-term = "0.46"
-opendal = { version = "0.51.1", features = [
-    "services-fs",
-    "services-s3",
-] }
+object-store.workspace = true
+operator.workspace = true
 query.workspace = true
 rand.workspace = true
 reqwest.workspace = true
-rustyline = "10.1"
 serde.workspace = true
 serde_json.workspace = true
 servers.workspace = true
@@ -64,6 +66,7 @@ tokio.workspace = true
 tracing-appender.workspace = true
 
 [dev-dependencies]
+common-meta = { workspace = true, features = ["testing"] }
 common-version.workspace = true
 serde.workspace = true
 tempfile.workspace = true

src/cli/src/bench.rs

@@ -58,6 +58,7 @@ where
     info!("{desc}, average operation cost: {cost:.2} ms");
 }
 
+/// Command to benchmark table metadata operations.
 #[derive(Debug, Default, Parser)]
 pub struct BenchTableMetadataCommand {
     #[clap(long)]
@@ -159,6 +160,7 @@ fn create_table_info(table_id: TableId, table_name: TableName) -> RawTableInfo {
         options: Default::default(),
         region_numbers: (1..=100).collect(),
         partition_key_indices: vec![],
+        column_ids: vec![],
     };
 
     RawTableInfo {

src/cli/src/cmd.rs

@@ -1,154 +0,0 @@
-// Copyright 2023 Greptime Team
-//
-// Licensed under the Apache License, Version 2.0 (the "License");
-// you may not use this file except in compliance with the License.
-// You may obtain a copy of the License at
-//
-//     http://www.apache.org/licenses/LICENSE-2.0
-//
-// Unless required by applicable law or agreed to in writing, software
-// distributed under the License is distributed on an "AS IS" BASIS,
-// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-// See the License for the specific language governing permissions and
-// limitations under the License.
-
-use crate::error::{Error, InvalidReplCommandSnafu, Result};
-
-/// Represents the parsed command from the user (which may be over many lines)
-#[derive(Debug, PartialEq)]
-pub(crate) enum ReplCommand {
-    Help,
-    UseDatabase { db_name: String },
-    Sql { sql: String },
-    Exit,
-}
-
-impl TryFrom<&str> for ReplCommand {
-    type Error = Error;
-
-    fn try_from(input: &str) -> Result<Self> {
-        let input = input.trim();
-        if input.is_empty() {
-            return InvalidReplCommandSnafu {
-                reason: "No command specified".to_string(),
-            }
-            .fail();
-        }
-
-        // If line ends with ';', it must be treated as a complete input.
-        // However, the opposite is not true.
-        let input_is_completed = input.ends_with(';');
-
-        let input = input.strip_suffix(';').map(|x| x.trim()).unwrap_or(input);
-        let lowercase = input.to_lowercase();
-        match lowercase.as_str() {
-            "help" => Ok(Self::Help),
-            "exit" | "quit" => Ok(Self::Exit),
-            _ => match input.split_once(' ') {
-                Some((maybe_use, database)) if maybe_use.to_lowercase() == "use" => {
-                    Ok(Self::UseDatabase {
-                        db_name: database.trim().to_string(),
-                    })
-                }
-                // Any valid SQL must contains at least one whitespace.
-                Some(_) if input_is_completed => Ok(Self::Sql {
-                    sql: input.to_string(),
-                }),
-                _ => InvalidReplCommandSnafu {
-                    reason: format!("unknown command '{input}', maybe input is not completed"),
-                }
-                .fail(),
-            },
-        }
-    }
-}
-
-impl ReplCommand {
-    pub fn help() -> &'static str {
-        r#"
-Available commands (case insensitive):
-- 'help': print this help
-- 'exit' or 'quit': exit the REPL
-- 'use <your database name>': switch to another database/schema context
-- Other typed in text will be treated as SQL.
-  You can enter new line while typing, just remember to end it with ';'.
-"#
-    }
-}
-
-#[cfg(test)]
-mod tests {
-    use super::*;
-    use crate::error::Error::InvalidReplCommand;
-
-    #[test]
-    fn test_from_str() {
-        fn test_ok(s: &str, expected: ReplCommand) {
-            let actual: ReplCommand = s.try_into().unwrap();
-            assert_eq!(expected, actual, "'{}'", s);
-        }
-
-        fn test_err(s: &str) {
-            let result: Result<ReplCommand> = s.try_into();
-            assert!(matches!(result, Err(InvalidReplCommand { .. })))
-        }
-
-        test_err("");
-        test_err("  ");
-        test_err("\t");
-
-        test_ok("help", ReplCommand::Help);
-        test_ok("help", ReplCommand::Help);
-        test_ok("  help", ReplCommand::Help);
-        test_ok("  help  ", ReplCommand::Help);
-        test_ok("  HELP  ", ReplCommand::Help);
-        test_ok("  Help;  ", ReplCommand::Help);
-        test_ok("  help  ; ", ReplCommand::Help);
-
-        test_ok("exit", ReplCommand::Exit);
-        test_ok("exit;", ReplCommand::Exit);
-        test_ok("exit ;", ReplCommand::Exit);
-        test_ok("EXIT", ReplCommand::Exit);
-
-        test_ok("quit", ReplCommand::Exit);
-        test_ok("quit;", ReplCommand::Exit);
-        test_ok("quit ;", ReplCommand::Exit);
-        test_ok("QUIT", ReplCommand::Exit);
-
-        test_ok(
-            "use Foo",
-            ReplCommand::UseDatabase {
-                db_name: "Foo".to_string(),
-            },
-        );
-        test_ok(
-            "  use Foo ;  ",
-            ReplCommand::UseDatabase {
-                db_name: "Foo".to_string(),
-            },
-        );
-        // ensure that database name is case sensitive
-        test_ok(
-            "  use FOO ;  ",
-            ReplCommand::UseDatabase {
-                db_name: "FOO".to_string(),
-            },
-        );
-
-        // ensure that we aren't messing with capitalization
-        test_ok(
-            "SELECT * from foo;",
-            ReplCommand::Sql {
-                sql: "SELECT * from foo".to_string(),
-            },
-        );
-        // Input line (that don't belong to any other cases above) must ends with ';' to make it a valid SQL.
-        test_err("insert blah");
-        test_ok(
-            "insert blah;",
-            ReplCommand::Sql {
-                sql: "insert blah".to_string(),
-            },
-        );
-    }
-}

src/cli/src/data.rs

@@ -0,0 +1,39 @@
+// Copyright 2023 Greptime Team
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+mod export;
+mod import;
+
+use clap::Subcommand;
+use common_error::ext::BoxedError;
+
+use crate::data::export::ExportCommand;
+use crate::data::import::ImportCommand;
+use crate::Tool;
+
+/// Command for data operations including exporting data from and importing data into GreptimeDB.
+#[derive(Subcommand)]
+pub enum DataCommand {
+    Export(ExportCommand),
+    Import(ImportCommand),
+}
+
+impl DataCommand {
+    pub async fn build(&self) -> std::result::Result<Box<dyn Tool>, BoxedError> {
+        match self {
+            DataCommand::Export(cmd) => cmd.build().await,
+            DataCommand::Import(cmd) => cmd.build().await,
+        }
+    }
+}

src/cli/src/data/export.rs

@@ -19,10 +19,12 @@ use std::time::Duration;
 
 use async_trait::async_trait;
 use clap::{Parser, ValueEnum};
+use common_base::secrets::{ExposeSecret, SecretString};
 use common_error::ext::BoxedError;
 use common_telemetry::{debug, error, info};
-use opendal::layers::LoggingLayer;
-use opendal::{services, Operator};
+use object_store::layers::LoggingLayer;
+use object_store::services::Oss;
+use object_store::{services, ObjectStore};
 use serde_json::Value;
 use snafu::{OptionExt, ResultExt};
 use tokio::sync::Semaphore;
@@ -48,6 +50,7 @@ enum ExportTarget {
     All,
 }
 
+/// Command for exporting data from the GreptimeDB.
 #[derive(Debug, Default, Parser)]
 pub struct ExportCommand {
     /// Server address to connect
@@ -110,11 +113,26 @@ pub struct ExportCommand {
     #[clap(long)]
     s3: bool,
 
+    /// if both `ddl_local_dir` and remote storage (s3/oss) are set, `ddl_local_dir` will be only used for
+    /// exported SQL files, and the data will be exported to remote storage.
+    ///
+    /// Note that `ddl_local_dir` export sql files to **LOCAL** file system, this is useful if export client don't have
+    /// direct access to remote storage.
+    ///
+    /// if remote storage is set but `ddl_local_dir` is not set, both SQL&data will be exported to remote storage.
+    #[clap(long)]
+    ddl_local_dir: Option<String>,
+
     /// The s3 bucket name
     /// if s3 is set, this is required
     #[clap(long)]
     s3_bucket: Option<String>,
 
+    // The s3 root path
+    /// if s3 is set, this is required
+    #[clap(long)]
+    s3_root: Option<String>,
+
     /// The s3 endpoint
     /// if s3 is set, this is required
     #[clap(long)]
@@ -134,6 +152,30 @@ pub struct ExportCommand {
     /// if s3 is set, this is required
     #[clap(long)]
     s3_region: Option<String>,
+
+    /// if export data to oss
+    #[clap(long)]
+    oss: bool,
+
+    /// The oss bucket name
+    /// if oss is set, this is required
+    #[clap(long)]
+    oss_bucket: Option<String>,
+
+    /// The oss endpoint
+    /// if oss is set, this is required
+    #[clap(long)]
+    oss_endpoint: Option<String>,
+
+    /// The oss access key id
+    /// if oss is set, this is required
+    #[clap(long)]
+    oss_access_key_id: Option<String>,
+
+    /// The oss access key secret
+    /// if oss is set, this is required
+    #[clap(long)]
+    oss_access_key_secret: Option<String>,
 }
 
 impl ExportCommand {
@@ -147,7 +189,7 @@ impl ExportCommand {
         {
             return Err(BoxedError::new(S3ConfigNotSetSnafu {}.build()));
         }
-        if !self.s3 && self.output_dir.is_none() {
+        if !self.s3 && !self.oss && self.output_dir.is_none() {
             return Err(BoxedError::new(OutputDirNotSetSnafu {}.build()));
         }
         let (catalog, schema) =
@@ -172,11 +214,32 @@ impl ExportCommand {
             start_time: self.start_time.clone(),
             end_time: self.end_time.clone(),
             s3: self.s3,
+            ddl_local_dir: self.ddl_local_dir.clone(),
             s3_bucket: self.s3_bucket.clone(),
+            s3_root: self.s3_root.clone(),
             s3_endpoint: self.s3_endpoint.clone(),
-            s3_access_key: self.s3_access_key.clone(),
-            s3_secret_key: self.s3_secret_key.clone(),
+            // Wrap sensitive values in SecretString
+            s3_access_key: self
+                .s3_access_key
+                .as_ref()
+                .map(|k| SecretString::from(k.clone())),
+            s3_secret_key: self
+                .s3_secret_key
+                .as_ref()
+                .map(|k| SecretString::from(k.clone())),
             s3_region: self.s3_region.clone(),
+            oss: self.oss,
+            oss_bucket: self.oss_bucket.clone(),
+            oss_endpoint: self.oss_endpoint.clone(),
+            // Wrap sensitive values in SecretString
+            oss_access_key_id: self
+                .oss_access_key_id
+                .as_ref()
+                .map(|k| SecretString::from(k.clone())),
+            oss_access_key_secret: self
+                .oss_access_key_secret
+                .as_ref()
+                .map(|k| SecretString::from(k.clone())),
         }))
     }
 }
@@ -192,21 +255,30 @@ pub struct Export {
     start_time: Option<String>,
     end_time: Option<String>,
     s3: bool,
+    ddl_local_dir: Option<String>,
     s3_bucket: Option<String>,
+    s3_root: Option<String>,
     s3_endpoint: Option<String>,
-    s3_access_key: Option<String>,
-    s3_secret_key: Option<String>,
+    // Changed to SecretString for sensitive data
+    s3_access_key: Option<SecretString>,
+    s3_secret_key: Option<SecretString>,
     s3_region: Option<String>,
+    oss: bool,
+    oss_bucket: Option<String>,
+    oss_endpoint: Option<String>,
+    // Changed to SecretString for sensitive data
+    oss_access_key_id: Option<SecretString>,
+    oss_access_key_secret: Option<SecretString>,
 }
 
 impl Export {
     fn catalog_path(&self) -> PathBuf {
-        if self.s3 {
+        if self.s3 || self.oss {
             PathBuf::from(&self.catalog)
         } else if let Some(dir) = &self.output_dir {
             PathBuf::from(dir).join(&self.catalog)
         } else {
-            unreachable!("catalog_path: output_dir must be set when not using s3")
+            unreachable!("catalog_path: output_dir must be set when not using remote storage")
         }
     }
 
@@ -364,7 +436,7 @@ impl Export {
         let timer = Instant::now();
         let db_names = self.get_db_names().await?;
         let db_count = db_names.len();
-        let operator = self.build_operator().await?;
+        let operator = self.build_prefer_fs_operator().await?;
 
         for schema in db_names {
             let create_database = self
@@ -394,7 +466,7 @@ impl Export {
         let semaphore = Arc::new(Semaphore::new(self.parallelism));
         let db_names = self.get_db_names().await?;
         let db_count = db_names.len();
-        let operator = Arc::new(self.build_operator().await?);
+        let operator = Arc::new(self.build_prefer_fs_operator().await?);
         let mut tasks = Vec::with_capacity(db_names.len());
 
         for schema in db_names {
@@ -408,7 +480,7 @@ impl Export {
                     .await?;
 
                 // Create directory if needed for file system storage
-                if !export_self.s3 {
+                if !export_self.s3 && !export_self.oss {
                     let db_dir = format!("{}/{}/", export_self.catalog, schema);
                     operator.create_dir(&db_dir).await.context(OpenDalSnafu)?;
                 }
@@ -451,21 +523,45 @@ impl Export {
         Ok(())
     }
 
-    async fn build_operator(&self) -> Result<Operator> {
+    async fn build_operator(&self) -> Result<ObjectStore> {
         if self.s3 {
             self.build_s3_operator().await
+        } else if self.oss {
+            self.build_oss_operator().await
         } else {
             self.build_fs_operator().await
         }
     }
 
-    async fn build_s3_operator(&self) -> Result<Operator> {
-        let mut builder = services::S3::default().root("").bucket(
+    /// build operator with preference for file system
+    async fn build_prefer_fs_operator(&self) -> Result<ObjectStore> {
+        if (self.s3 || self.oss) && self.ddl_local_dir.is_some() {
+            let root = self.ddl_local_dir.as_ref().unwrap().clone();
+            let op = ObjectStore::new(services::Fs::default().root(&root))
+                .context(OpenDalSnafu)?
+                .layer(LoggingLayer::default())
+                .finish();
+            Ok(op)
+        } else if self.s3 {
+            self.build_s3_operator().await
+        } else if self.oss {
+            self.build_oss_operator().await
+        } else {
+            self.build_fs_operator().await
+        }
+    }
+
+    async fn build_s3_operator(&self) -> Result<ObjectStore> {
+        let mut builder = services::S3::default().bucket(
             self.s3_bucket
                 .as_ref()
                 .expect("s3_bucket must be provided when s3 is enabled"),
         );
 
+        if let Some(root) = self.s3_root.as_ref() {
+            builder = builder.root(root);
+        }
+
         if let Some(endpoint) = self.s3_endpoint.as_ref() {
             builder = builder.endpoint(endpoint);
         }
@@ -475,27 +571,51 @@ impl Export {
         }
 
         if let Some(key_id) = self.s3_access_key.as_ref() {
-            builder = builder.access_key_id(key_id);
+            builder = builder.access_key_id(key_id.expose_secret());
         }
 
         if let Some(secret_key) = self.s3_secret_key.as_ref() {
-            builder = builder.secret_access_key(secret_key);
+            builder = builder.secret_access_key(secret_key.expose_secret());
         }
 
-        let op = Operator::new(builder)
+        let op = ObjectStore::new(builder)
             .context(OpenDalSnafu)?
             .layer(LoggingLayer::default())
             .finish();
         Ok(op)
     }
 
-    async fn build_fs_operator(&self) -> Result<Operator> {
+    async fn build_oss_operator(&self) -> Result<ObjectStore> {
+        let mut builder = Oss::default()
+            .bucket(self.oss_bucket.as_ref().expect("oss_bucket must be set"))
+            .endpoint(
+                self.oss_endpoint
+                    .as_ref()
+                    .expect("oss_endpoint must be set"),
+            );
+
+        // Use expose_secret() to access the actual secret value
+        if let Some(key_id) = self.oss_access_key_id.as_ref() {
+            builder = builder.access_key_id(key_id.expose_secret());
+        }
+        if let Some(secret_key) = self.oss_access_key_secret.as_ref() {
+            builder = builder.access_key_secret(secret_key.expose_secret());
+        }
+
+        let op = ObjectStore::new(builder)
+            .context(OpenDalSnafu)?
+            .layer(LoggingLayer::default())
+            .finish();
+        Ok(op)
+    }
+
+    async fn build_fs_operator(&self) -> Result<ObjectStore> {
         let root = self
             .output_dir
             .as_ref()
             .context(OutputDirNotSetSnafu)?
             .clone();
-        let op = Operator::new(services::Fs::default().root(&root))
+        let op = ObjectStore::new(services::Fs::default().root(&root))
             .context(OpenDalSnafu)?
             .layer(LoggingLayer::default())
             .finish();
@@ -509,6 +629,7 @@ impl Export {
         let db_count = db_names.len();
         let mut tasks = Vec::with_capacity(db_count);
         let operator = Arc::new(self.build_operator().await?);
+        let fs_first_operator = Arc::new(self.build_prefer_fs_operator().await?);
         let with_options = build_with_options(&self.start_time, &self.end_time);
 
         for schema in db_names {
@@ -516,12 +637,13 @@ impl Export {
             let export_self = self.clone();
             let with_options_clone = with_options.clone();
             let operator = operator.clone();
+            let fs_first_operator = fs_first_operator.clone();
 
             tasks.push(async move {
                 let _permit = semaphore_moved.acquire().await.unwrap();
 
-                // Create directory if not using S3
-                if !export_self.s3 {
+                // Create directory if not using remote storage
+                if !export_self.s3 && !export_self.oss {
                     let db_dir = format!("{}/{}/", export_self.catalog, schema);
                     operator.create_dir(&db_dir).await.context(OpenDalSnafu)?;
                 }
@@ -533,7 +655,11 @@ impl Export {
                     r#"COPY DATABASE "{}"."{}" TO '{}' WITH ({}){};"#,
                     export_self.catalog, schema, path, with_options_clone, connection_part
                 );
-                info!("Executing sql: {sql}");
+
+                // Log SQL command but mask sensitive information
+                let safe_sql = export_self.mask_sensitive_sql(&sql);
+                info!("Executing sql: {}", safe_sql);
+
                 export_self.database_client.sql_in_public(&sql).await?;
                 info!(
                     "Finished exporting {}.{} data to {}",
@@ -549,7 +675,7 @@ impl Export {
                 let copy_from_path = export_self.get_file_path(&schema, "copy_from.sql");
                 export_self
                     .write_to_storage(
-                        &operator,
+                        &fs_first_operator,
                         &copy_from_path,
                         copy_database_from_sql.into_bytes(),
                     )
@@ -573,6 +699,29 @@ impl Export {
         Ok(())
     }
 
+    /// Mask sensitive information in SQL commands for safe logging
+    fn mask_sensitive_sql(&self, sql: &str) -> String {
+        let mut masked_sql = sql.to_string();
+
+        // Mask S3 credentials
+        if let Some(access_key) = &self.s3_access_key {
+            masked_sql = masked_sql.replace(access_key.expose_secret(), "[REDACTED]");
+        }
+        if let Some(secret_key) = &self.s3_secret_key {
+            masked_sql = masked_sql.replace(secret_key.expose_secret(), "[REDACTED]");
+        }
+
+        // Mask OSS credentials
+        if let Some(access_key_id) = &self.oss_access_key_id {
+            masked_sql = masked_sql.replace(access_key_id.expose_secret(), "[REDACTED]");
+        }
+        if let Some(access_key_secret) = &self.oss_access_key_secret {
+            masked_sql = masked_sql.replace(access_key_secret.expose_secret(), "[REDACTED]");
+        }
+
+        masked_sql
+    }
+
     fn get_file_path(&self, schema: &str, file_name: &str) -> String {
         format!("{}/{}/{}", self.catalog, schema, file_name)
     }
@@ -580,8 +729,20 @@ impl Export {
     fn format_output_path(&self, file_path: &str) -> String {
         if self.s3 {
             format!(
-                "s3://{}/{}",
+                "s3://{}{}/{}",
                 self.s3_bucket.as_ref().unwrap_or(&String::new()),
+                if let Some(root) = &self.s3_root {
+                    format!("/{}", root)
+                } else {
+                    String::new()
+                },
+                file_path
+            )
+        } else if self.oss {
+            format!(
+                "oss://{}/{}/{}",
+                self.oss_bucket.as_ref().unwrap_or(&String::new()),
+                self.catalog,
                 file_path
             )
         } else {
@@ -595,19 +756,27 @@ impl Export {
 
     async fn write_to_storage(
         &self,
-        op: &Operator,
+        op: &ObjectStore,
         file_path: &str,
         content: Vec<u8>,
     ) -> Result<()> {
-        op.write(file_path, content).await.context(OpenDalSnafu)
+        op.write(file_path, content)
+            .await
+            .context(OpenDalSnafu)
+            .map(|_| ())
     }
 
     fn get_storage_params(&self, schema: &str) -> (String, String) {
         if self.s3 {
             let s3_path = format!(
-                "s3://{}/{}/{}/",
+                "s3://{}{}/{}/{}/",
                 // Safety: s3_bucket is required when s3 is enabled
                 self.s3_bucket.as_ref().unwrap(),
+                if let Some(root) = &self.s3_root {
+                    format!("/{}", root)
+                } else {
+                    String::new()
+                },
                 self.catalog,
                 schema
             );
@@ -620,15 +789,36 @@ impl Export {
             };
 
             // Safety: All s3 options are required
+            // Use expose_secret() to access the actual secret values
             let connection_options = format!(
                 "ACCESS_KEY_ID='{}', SECRET_ACCESS_KEY='{}', REGION='{}'{}",
-                self.s3_access_key.as_ref().unwrap(),
-                self.s3_secret_key.as_ref().unwrap(),
+                self.s3_access_key.as_ref().unwrap().expose_secret(),
+                self.s3_secret_key.as_ref().unwrap().expose_secret(),
                 self.s3_region.as_ref().unwrap(),
                 endpoint_option
             );
 
             (s3_path, format!(" CONNECTION ({})", connection_options))
+        } else if self.oss {
+            let oss_path = format!(
+                "oss://{}/{}/{}/",
+                self.oss_bucket.as_ref().unwrap(),
+                self.catalog,
+                schema
+            );
+            let endpoint_option = if let Some(endpoint) = self.oss_endpoint.as_ref() {
+                format!(", ENDPOINT='{}'", endpoint)
+            } else {
+                String::new()
+            };
+
+            let connection_options = format!(
+                "ACCESS_KEY_ID='{}', ACCESS_KEY_SECRET='{}'{}",
+                self.oss_access_key_id.as_ref().unwrap().expose_secret(),
+                self.oss_access_key_secret.as_ref().unwrap().expose_secret(),
+                endpoint_option
+            );
+            (oss_path, format!(" CONNECTION ({})", connection_options))
         } else {
             (
                 self.catalog_path()

src/cli/src/data/import.rs

@@ -40,6 +40,7 @@ enum ImportTarget {
     All,
 }
 
+/// Command to import data from a directory into a GreptimeDB instance.
 #[derive(Debug, Default, Parser)]
 pub struct ImportCommand {
     /// Server address to connect

src/cli/src/error.rs

@@ -17,7 +17,10 @@ use std::any::Any;
 use common_error::ext::{BoxedError, ErrorExt};
 use common_error::status_code::StatusCode;
 use common_macro::stack_trace_debug;
+use common_meta::peer::Peer;
+use object_store::Error as ObjectStoreError;
 use snafu::{Location, Snafu};
+use store_api::storage::TableId;
 
 #[derive(Snafu)]
 #[snafu(visibility(pub))]
@@ -29,6 +32,7 @@ pub enum Error {
         location: Location,
         msg: String,
     },
+
     #[snafu(display("Failed to create default catalog and schema"))]
     InitMetadata {
         #[snafu(implicit)]
@@ -71,6 +75,20 @@ pub enum Error {
         source: common_meta::error::Error,
     },
 
+    #[snafu(display("Failed to get table metadata"))]
+    TableMetadata {
+        #[snafu(implicit)]
+        location: Location,
+        source: common_meta::error::Error,
+    },
+
+    #[snafu(display("Unexpected error: {}", msg))]
+    Unexpected {
+        msg: String,
+        #[snafu(implicit)]
+        location: Location,
+    },
+
     #[snafu(display("Missing config, msg: {}", msg))]
     MissingConfig {
         msg: String,
@@ -101,9 +119,6 @@ pub enum Error {
         error: reqwest::Error,
     },
 
-    #[snafu(display("Invalid REPL command: {reason}"))]
-    InvalidReplCommand { reason: String },
-
     #[snafu(display("Failed to parse SQL: {}", sql))]
     ParseSql {
         sql: String,
@@ -223,23 +238,81 @@ pub enum Error {
         location: Location,
     },
 
+    #[snafu(display("Table not found: {table_id}"))]
+    TableNotFound {
+        table_id: TableId,
+        #[snafu(implicit)]
+        location: Location,
+    },
+
     #[snafu(display("OpenDAL operator failed"))]
     OpenDal {
         #[snafu(implicit)]
         location: Location,
         #[snafu(source)]
-        error: opendal::Error,
+        error: ObjectStoreError,
     },
+
     #[snafu(display("S3 config need be set"))]
     S3ConfigNotSet {
         #[snafu(implicit)]
         location: Location,
     },
+
     #[snafu(display("Output directory not set"))]
     OutputDirNotSet {
         #[snafu(implicit)]
         location: Location,
     },
+
+    #[snafu(display("Empty store addresses"))]
+    EmptyStoreAddrs {
+        #[snafu(implicit)]
+        location: Location,
+    },
+
+    #[snafu(display("Unsupported memory backend"))]
+    UnsupportedMemoryBackend {
+        #[snafu(implicit)]
+        location: Location,
+    },
+
+    #[snafu(display("File path invalid: {}", msg))]
+    InvalidFilePath {
+        msg: String,
+        #[snafu(implicit)]
+        location: Location,
+    },
+
+    #[snafu(display("Invalid arguments: {}", msg))]
+    InvalidArguments {
+        msg: String,
+        #[snafu(implicit)]
+        location: Location,
+    },
+
+    #[snafu(display("Failed to init backend"))]
+    InitBackend {
+        #[snafu(implicit)]
+        location: Location,
+        #[snafu(source)]
+        error: ObjectStoreError,
+    },
+
+    #[snafu(display("Covert column schemas to defs failed"))]
+    CovertColumnSchemasToDefs {
+        #[snafu(implicit)]
+        location: Location,
+        source: operator::error::Error,
+    },
+
+    #[snafu(display("Failed to send request to datanode: {}", peer))]
+    SendRequestToDatanode {
+        peer: Peer,
+        #[snafu(implicit)]
+        location: Location,
+        source: common_meta::error::Error,
+    },
 }
 
 pub type Result<T> = std::result::Result<T, Error>;
@@ -247,20 +320,25 @@ pub type Result<T> = std::result::Result<T, Error>;
 impl ErrorExt for Error {
     fn status_code(&self) -> StatusCode {
         match self {
-            Error::InitMetadata { source, .. } | Error::InitDdlManager { source, .. } => {
-                source.status_code()
-            }
+            Error::InitMetadata { source, .. }
+            | Error::InitDdlManager { source, .. }
+            | Error::TableMetadata { source, .. } => source.status_code(),
 
             Error::MissingConfig { .. }
             | Error::LoadLayeredConfig { .. }
             | Error::IllegalConfig { .. }
-            | Error::InvalidReplCommand { .. }
             | Error::InitTimezone { .. }
             | Error::ConnectEtcd { .. }
             | Error::CreateDir { .. }
             | Error::EmptyResult { .. }
+            | Error::InvalidFilePath { .. }
+            | Error::UnsupportedMemoryBackend { .. }
+            | Error::InvalidArguments { .. }
             | Error::ParseProxyOpts { .. } => StatusCode::InvalidArguments,
 
+            Error::CovertColumnSchemasToDefs { source, .. } => source.status_code(),
+            Error::SendRequestToDatanode { source, .. } => source.status_code(),
+
             Error::StartProcedureManager { source, .. }
             | Error::StopProcedureManager { source, .. } => source.status_code(),
             Error::StartWalOptionsAllocator { source, .. } => source.status_code(),
@@ -268,6 +346,7 @@ impl ErrorExt for Error {
             Error::ParseSql { source, .. } | Error::PlanStatement { source, .. } => {
                 source.status_code()
             }
+            Error::Unexpected { .. } => StatusCode::Unexpected,
 
             Error::SerdeJson { .. }
             | Error::FileIo { .. }
@@ -276,14 +355,16 @@ impl ErrorExt for Error {
             | Error::BuildClient { .. } => StatusCode::Unexpected,
 
             Error::Other { source, .. } => source.status_code(),
-            Error::OpenDal { .. } => StatusCode::Internal,
-            Error::S3ConfigNotSet { .. } => StatusCode::InvalidArguments,
-            Error::OutputDirNotSet { .. } => StatusCode::InvalidArguments,
+            Error::OpenDal { .. } | Error::InitBackend { .. } => StatusCode::Internal,
+            Error::S3ConfigNotSet { .. }
+            | Error::OutputDirNotSet { .. }
+            | Error::EmptyStoreAddrs { .. } => StatusCode::InvalidArguments,
 
             Error::BuildRuntime { source, .. } => source.status_code(),
 
             Error::CacheRequired { .. } | Error::BuildCacheRegistry { .. } => StatusCode::Internal,
             Error::MetaClientInit { source, .. } => source.status_code(),
+            Error::TableNotFound { .. } => StatusCode::TableNotFound,
             Error::SchemaNotFound { .. } => StatusCode::DatabaseNotFound,
         }
     }

src/cli/src/helper.rs

@@ -1,112 +0,0 @@
-// Copyright 2023 Greptime Team
-//
-// Licensed under the Apache License, Version 2.0 (the "License");
-// you may not use this file except in compliance with the License.
-// You may obtain a copy of the License at
-//
-//     http://www.apache.org/licenses/LICENSE-2.0
-//
-// Unless required by applicable law or agreed to in writing, software
-// distributed under the License is distributed on an "AS IS" BASIS,
-// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-// See the License for the specific language governing permissions and
-// limitations under the License.
-
-use std::borrow::Cow;
-
-use rustyline::completion::Completer;
-use rustyline::highlight::{Highlighter, MatchingBracketHighlighter};
-use rustyline::hint::{Hinter, HistoryHinter};
-use rustyline::validate::{ValidationContext, ValidationResult, Validator};
-
-use crate::cmd::ReplCommand;
-
-pub(crate) struct RustylineHelper {
-    hinter: HistoryHinter,
-    highlighter: MatchingBracketHighlighter,
-}
-
-impl Default for RustylineHelper {
-    fn default() -> Self {
-        Self {
-            hinter: HistoryHinter {},
-            highlighter: MatchingBracketHighlighter::default(),
-        }
-    }
-}
-
-impl rustyline::Helper for RustylineHelper {}
-
-impl Validator for RustylineHelper {
-    fn validate(&self, ctx: &mut ValidationContext<'_>) -> rustyline::Result<ValidationResult> {
-        let input = ctx.input();
-        match ReplCommand::try_from(input) {
-            Ok(_) => Ok(ValidationResult::Valid(None)),
-            Err(e) => {
-                if input.trim_end().ends_with(';') {
-                    // If line ends with ';', it HAS to be a valid command.
-                    Ok(ValidationResult::Invalid(Some(e.to_string())))
-                } else {
-                    Ok(ValidationResult::Incomplete)
-                }
-            }
-        }
-    }
-}
-
-impl Hinter for RustylineHelper {
-    type Hint = String;
-
-    fn hint(&self, line: &str, pos: usize, ctx: &rustyline::Context<'_>) -> Option<Self::Hint> {
-        self.hinter.hint(line, pos, ctx)
-    }
-}
-
-impl Highlighter for RustylineHelper {
-    fn highlight<'l>(&self, line: &'l str, pos: usize) -> Cow<'l, str> {
-        self.highlighter.highlight(line, pos)
-    }
-
-    fn highlight_prompt<'b, 's: 'b, 'p: 'b>(
-        &'s self,
-        prompt: &'p str,
-        default: bool,
-    ) -> Cow<'b, str> {
-        self.highlighter.highlight_prompt(prompt, default)
-    }
-
-    fn highlight_hint<'h>(&self, hint: &'h str) -> Cow<'h, str> {
-        use nu_ansi_term::Style;
-        Cow::Owned(Style::new().dimmed().paint(hint).to_string())
-    }
-
-    fn highlight_candidate<'c>(
-        &self,
-        candidate: &'c str,
-        completion: rustyline::CompletionType,
-    ) -> Cow<'c, str> {
-        self.highlighter.highlight_candidate(candidate, completion)
-    }
-
-    fn highlight_char(&self, line: &str, pos: usize) -> bool {
-        self.highlighter.highlight_char(line, pos)
-    }
-}
-
-impl Completer for RustylineHelper {
-    type Candidate = String;
-
-    fn complete(
-        &self,
-        line: &str,
-        pos: usize,
-        ctx: &rustyline::Context<'_>,
-    ) -> rustyline::Result<(usize, Vec<Self::Candidate>)> {
-        // If there is a hint, use that as the auto-complete when user hits `tab`
-        if let Some(hint) = self.hinter.hint(line, pos, ctx) {
-            Ok((pos, vec![hint]))
-        } else {
-            Ok((0, vec![]))
-        }
-    }
-}

src/cli/src/lib.rs

@@ -13,16 +13,10 @@
 // limitations under the License.
 
 mod bench;
-pub mod error;
-// Wait for https://github.com/GreptimeTeam/greptimedb/issues/2373
-#[allow(unused)]
-mod cmd;
-mod export;
-mod helper;
-
-// Wait for https://github.com/GreptimeTeam/greptimedb/issues/2373
+mod data;
 mod database;
-mod import;
+pub mod error;
+mod metadata;
 
 use async_trait::async_trait;
 use clap::Parser;
@@ -31,8 +25,8 @@ pub use database::DatabaseClient;
 use error::Result;
 
 pub use crate::bench::BenchTableMetadataCommand;
-pub use crate::export::ExportCommand;
-pub use crate::import::ImportCommand;
+pub use crate::data::DataCommand;
+pub use crate::metadata::MetadataCommand;
 
 #[async_trait]
 pub trait Tool: Send + Sync {

src/cli/src/metadata.rs

@@ -0,0 +1,52 @@
+// Copyright 2023 Greptime Team
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+mod common;
+mod control;
+mod repair;
+mod snapshot;
+mod utils;
+
+use clap::Subcommand;
+use common_error::ext::BoxedError;
+
+use crate::metadata::control::{DelCommand, GetCommand};
+use crate::metadata::repair::RepairLogicalTablesCommand;
+use crate::metadata::snapshot::SnapshotCommand;
+use crate::Tool;
+
+/// Command for managing metadata operations,
+/// including saving and restoring metadata snapshots,
+/// controlling metadata operations, and diagnosing and repairing metadata.
+#[derive(Subcommand)]
+pub enum MetadataCommand {
+    #[clap(subcommand)]
+    Snapshot(SnapshotCommand),
+    #[clap(subcommand)]
+    Get(GetCommand),
+    #[clap(subcommand)]
+    Del(DelCommand),
+    RepairLogicalTables(RepairLogicalTablesCommand),
+}
+
+impl MetadataCommand {
+    pub async fn build(&self) -> Result<Box<dyn Tool>, BoxedError> {
+        match self {
+            MetadataCommand::Snapshot(cmd) => cmd.build().await,
+            MetadataCommand::RepairLogicalTables(cmd) => cmd.build().await,
+            MetadataCommand::Get(cmd) => cmd.build().await,
+            MetadataCommand::Del(cmd) => cmd.build().await,
+        }
+    }
+}

src/cli/src/metadata/common.rs

@@ -0,0 +1,116 @@
+// Copyright 2023 Greptime Team
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+use std::sync::Arc;
+
+use clap::Parser;
+use common_error::ext::BoxedError;
+use common_meta::kv_backend::chroot::ChrootKvBackend;
+use common_meta::kv_backend::etcd::EtcdStore;
+use common_meta::kv_backend::KvBackendRef;
+use meta_srv::bootstrap::create_etcd_client;
+use meta_srv::metasrv::BackendImpl;
+
+use crate::error::{EmptyStoreAddrsSnafu, UnsupportedMemoryBackendSnafu};
+
+#[derive(Debug, Default, Parser)]
+pub(crate) struct StoreConfig {
+    /// The endpoint of store. one of etcd, postgres or mysql.
+    ///
+    /// For postgres store, the format is:
+    /// "password=password dbname=postgres user=postgres host=localhost port=5432"
+    ///
+    /// For etcd store, the format is:
+    /// "127.0.0.1:2379"
+    ///
+    /// For mysql store, the format is:
+    /// "mysql://user:password@ip:port/dbname"
+    #[clap(long, alias = "store-addr", value_delimiter = ',', num_args = 1..)]
+    store_addrs: Vec<String>,
+
+    /// The maximum number of operations in a transaction. Only used when using [etcd-store].
+    #[clap(long, default_value = "128")]
+    max_txn_ops: usize,
+
+    /// The metadata store backend.
+    #[clap(long, value_enum, default_value = "etcd-store")]
+    backend: BackendImpl,
+
+    /// The key prefix of the metadata store.
+    #[clap(long, default_value = "")]
+    store_key_prefix: String,
+
+    /// The table name in RDS to store metadata. Only used when using [postgres-store] or [mysql-store].
+    #[cfg(any(feature = "pg_kvbackend", feature = "mysql_kvbackend"))]
+    #[clap(long, default_value = common_meta::kv_backend::DEFAULT_META_TABLE_NAME)]
+    meta_table_name: String,
+}
+
+impl StoreConfig {
+    /// Builds a [`KvBackendRef`] from the store configuration.
+    pub async fn build(&self) -> Result<KvBackendRef, BoxedError> {
+        let max_txn_ops = self.max_txn_ops;
+        let store_addrs = &self.store_addrs;
+        if store_addrs.is_empty() {
+            EmptyStoreAddrsSnafu.fail().map_err(BoxedError::new)
+        } else {
+            let kvbackend = match self.backend {
+                BackendImpl::EtcdStore => {
+                    let etcd_client = create_etcd_client(store_addrs)
+                        .await
+                        .map_err(BoxedError::new)?;
+                    Ok(EtcdStore::with_etcd_client(etcd_client, max_txn_ops))
+                }
+                #[cfg(feature = "pg_kvbackend")]
+                BackendImpl::PostgresStore => {
+                    let table_name = &self.meta_table_name;
+                    let pool = meta_srv::bootstrap::create_postgres_pool(store_addrs)
+                        .await
+                        .map_err(BoxedError::new)?;
+                    Ok(common_meta::kv_backend::rds::PgStore::with_pg_pool(
+                        pool,
+                        table_name,
+                        max_txn_ops,
+                    )
+                    .await
+                    .map_err(BoxedError::new)?)
+                }
+                #[cfg(feature = "mysql_kvbackend")]
+                BackendImpl::MysqlStore => {
+                    let table_name = &self.meta_table_name;
+                    let pool = meta_srv::bootstrap::create_mysql_pool(store_addrs)
+                        .await
+                        .map_err(BoxedError::new)?;
+                    Ok(common_meta::kv_backend::rds::MySqlStore::with_mysql_pool(
+                        pool,
+                        table_name,
+                        max_txn_ops,
+                    )
+                    .await
+                    .map_err(BoxedError::new)?)
+                }
+                BackendImpl::MemoryStore => UnsupportedMemoryBackendSnafu
+                    .fail()
+                    .map_err(BoxedError::new),
+            };
+            if self.store_key_prefix.is_empty() {
+                kvbackend
+            } else {
+                let chroot_kvbackend =
+                    ChrootKvBackend::new(self.store_key_prefix.as_bytes().to_vec(), kvbackend?);
+                Ok(Arc::new(chroot_kvbackend))
+            }
+        }
+    }
+}

src/cli/src/metadata/control.rs

@@ -0,0 +1,22 @@
+// Copyright 2023 Greptime Team
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+mod del;
+mod get;
+#[cfg(test)]
+mod test_utils;
+mod utils;
+
+pub(crate) use del::DelCommand;
+pub(crate) use get::GetCommand;

src/cli/src/metadata/control/del.rs

@@ -0,0 +1,42 @@
+// Copyright 2023 Greptime Team
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+mod key;
+mod table;
+
+use clap::Subcommand;
+use common_error::ext::BoxedError;
+
+use crate::metadata::control::del::key::DelKeyCommand;
+use crate::metadata::control::del::table::DelTableCommand;
+use crate::Tool;
+
+/// The prefix of the tombstone keys.
+pub(crate) const CLI_TOMBSTONE_PREFIX: &str = "__cli_tombstone/";
+
+/// Subcommand for deleting metadata from the metadata store.
+#[derive(Subcommand)]
+pub enum DelCommand {
+    Key(DelKeyCommand),
+    Table(DelTableCommand),
+}
+
+impl DelCommand {
+    pub async fn build(&self) -> Result<Box<dyn Tool>, BoxedError> {
+        match self {
+            DelCommand::Key(cmd) => cmd.build().await,
+            DelCommand::Table(cmd) => cmd.build().await,
+        }
+    }
+}

src/cli/src/metadata/control/del/key.rs

@@ -0,0 +1,132 @@
+// Copyright 2023 Greptime Team
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+use async_trait::async_trait;
+use clap::Parser;
+use common_error::ext::BoxedError;
+use common_meta::key::tombstone::TombstoneManager;
+use common_meta::kv_backend::KvBackendRef;
+use common_meta::rpc::store::RangeRequest;
+
+use crate::metadata::common::StoreConfig;
+use crate::metadata::control::del::CLI_TOMBSTONE_PREFIX;
+use crate::Tool;
+
+/// Delete key-value pairs logically from the metadata store.
+#[derive(Debug, Default, Parser)]
+pub struct DelKeyCommand {
+    /// The key to delete from the metadata store.
+    key: String,
+
+    /// Delete key-value pairs with the given prefix.
+    #[clap(long)]
+    prefix: bool,
+
+    #[clap(flatten)]
+    store: StoreConfig,
+}
+
+impl DelKeyCommand {
+    pub async fn build(&self) -> Result<Box<dyn Tool>, BoxedError> {
+        let kv_backend = self.store.build().await?;
+        Ok(Box::new(DelKeyTool {
+            key: self.key.to_string(),
+            prefix: self.prefix,
+            key_deleter: KeyDeleter::new(kv_backend),
+        }))
+    }
+}
+
+struct KeyDeleter {
+    kv_backend: KvBackendRef,
+    tombstone_manager: TombstoneManager,
+}
+
+impl KeyDeleter {
+    fn new(kv_backend: KvBackendRef) -> Self {
+        Self {
+            kv_backend: kv_backend.clone(),
+            tombstone_manager: TombstoneManager::new_with_prefix(kv_backend, CLI_TOMBSTONE_PREFIX),
+        }
+    }
+
+    async fn delete(&self, key: &str, prefix: bool) -> Result<usize, BoxedError> {
+        let mut req = RangeRequest::default().with_keys_only();
+        if prefix {
+            req = req.with_prefix(key.as_bytes());
+        } else {
+            req = req.with_key(key.as_bytes());
+        }
+        let resp = self.kv_backend.range(req).await.map_err(BoxedError::new)?;
+        let keys = resp.kvs.iter().map(|kv| kv.key.clone()).collect::<Vec<_>>();
+        self.tombstone_manager
+            .create(keys)
+            .await
+            .map_err(BoxedError::new)
+    }
+}
+
+struct DelKeyTool {
+    key: String,
+    prefix: bool,
+    key_deleter: KeyDeleter,
+}
+
+#[async_trait]
+impl Tool for DelKeyTool {
+    async fn do_work(&self) -> Result<(), BoxedError> {
+        let deleted = self.key_deleter.delete(&self.key, self.prefix).await?;
+        // Print the number of deleted keys.
+        println!("{}", deleted);
+        Ok(())
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use std::sync::Arc;
+
+    use common_meta::kv_backend::chroot::ChrootKvBackend;
+    use common_meta::kv_backend::memory::MemoryKvBackend;
+    use common_meta::kv_backend::{KvBackend, KvBackendRef};
+    use common_meta::rpc::store::RangeRequest;
+
+    use crate::metadata::control::del::key::KeyDeleter;
+    use crate::metadata::control::del::CLI_TOMBSTONE_PREFIX;
+    use crate::metadata::control::test_utils::put_key;
+
+    #[tokio::test]
+    async fn test_delete_keys() {
+        let kv_backend = Arc::new(MemoryKvBackend::new()) as KvBackendRef;
+        let key_deleter = KeyDeleter::new(kv_backend.clone());
+        put_key(&kv_backend, "foo", "bar").await;
+        put_key(&kv_backend, "foo/bar", "baz").await;
+        put_key(&kv_backend, "foo/baz", "qux").await;
+        let deleted = key_deleter.delete("foo", true).await.unwrap();
+        assert_eq!(deleted, 3);
+        let deleted = key_deleter.delete("foo/bar", false).await.unwrap();
+        assert_eq!(deleted, 0);
+
+        let chroot = ChrootKvBackend::new(CLI_TOMBSTONE_PREFIX.as_bytes().to_vec(), kv_backend);
+        let req = RangeRequest::default().with_prefix(b"foo");
+        let resp = chroot.range(req).await.unwrap();
+        assert_eq!(resp.kvs.len(), 3);
+        assert_eq!(resp.kvs[0].key, b"foo");
+        assert_eq!(resp.kvs[0].value, b"bar");
+        assert_eq!(resp.kvs[1].key, b"foo/bar");
+        assert_eq!(resp.kvs[1].value, b"baz");
+        assert_eq!(resp.kvs[2].key, b"foo/baz");
+        assert_eq!(resp.kvs[2].value, b"qux");
+    }
+}