fix: config test failed and use similar_asserts::assert_eq to replace assert_eq for long string compare (#4731)

* fix: config test failed and use 'similar_asserts::assert_eq' to replace 'assert_eq' for long string compare

* Update Cargo.toml

Co-authored-by: Yingwen <realevenyag@gmail.com>

* Update src/cmd/tests/load_config_test.rs

Co-authored-by: Yingwen <realevenyag@gmail.com>

---------

Co-authored-by: Ruihang Xia <waynestxia@gmail.com>
Co-authored-by: Yingwen <realevenyag@gmail.com>

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

@@ -17,6 +17,12 @@ inputs:
     description: Enable dev mode, only build standard greptime
     required: false
     default: "false"
+  image-namespace:
+    description: Image Namespace
+    required: true
+  image-registry:
+    description: Image Registry
+    required: true
   working-dir:
     description: Working directory to build the artifacts
     required: false
@@ -31,8 +37,8 @@ runs:
       run: |
         cd ${{ inputs.working-dir }} && \
         make run-it-in-container BUILD_JOBS=4 \
-        IMAGE_NAMESPACE=i8k6a5e1/greptime \
-        IMAGE_REGISTRY=public.ecr.aws
+        IMAGE_NAMESPACE=${{ inputs.image-namespace }} \
+        IMAGE_REGISTRY=${{ inputs.image-registry }}
 
     - name: Upload sqlness logs
       if: ${{ failure() && inputs.disable-run-tests == 'false' }} # Only upload logs when the integration tests failed.
@@ -51,8 +57,8 @@ runs:
         artifacts-dir: greptime-linux-${{ inputs.arch }}-pyo3-${{ inputs.version }}
         version: ${{ inputs.version }}
         working-dir: ${{ inputs.working-dir }}
-        image-registry: public.ecr.aws
-        image-namespace: i8k6a5e1/greptime        
+        image-registry: ${{ inputs.image-registry }}
+        image-namespace: ${{ inputs.image-namespace }}
 
     - name: Build greptime without pyo3
       if: ${{ inputs.dev-mode == 'false' }}
@@ -64,8 +70,8 @@ runs:
         artifacts-dir: greptime-linux-${{ inputs.arch }}-${{ inputs.version }}
         version: ${{ inputs.version }}
         working-dir: ${{ inputs.working-dir }}
-        image-registry: public.ecr.aws
-        image-namespace: i8k6a5e1/greptime
+        image-registry: ${{ inputs.image-registry }}
+        image-namespace: ${{ inputs.image-namespace }}
 
     - name: Clean up the target directory # Clean up the target directory for the centos7 base image, or it will still use the objects of last build.
       shell: bash
@@ -82,8 +88,8 @@ runs:
         artifacts-dir: greptime-linux-${{ inputs.arch }}-centos-${{ inputs.version }}
         version: ${{ inputs.version }}
         working-dir: ${{ inputs.working-dir }}
-        image-registry: public.ecr.aws
-        image-namespace: i8k6a5e1/greptime
+        image-registry: ${{ inputs.image-registry }}
+        image-namespace: ${{ inputs.image-namespace }}
 
     - name: Build greptime on android base image
       uses: ./.github/actions/build-greptime-binary
@@ -94,5 +100,5 @@ runs:
         version: ${{ inputs.version }}
         working-dir: ${{ inputs.working-dir }}
         build-android-artifacts: true
-        image-registry: public.ecr.aws
-        image-namespace: i8k6a5e1/greptime
+        image-registry: ${{ inputs.image-registry }}
+        image-namespace: ${{ inputs.image-namespace }}

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

@@ -4,9 +4,6 @@ inputs:
   arch:
     description: Architecture to build
     required: true
-  rust-toolchain:
-    description: Rust toolchain to use
-    required: true
   cargo-profile:
     description: Cargo profile to build
     required: true
@@ -43,10 +40,9 @@ runs:
         brew install protobuf
 
     - name: Install rust toolchain
-      uses: dtolnay/rust-toolchain@master
+      uses: actions-rust-lang/setup-rust-toolchain@v1
       with:
-        toolchain: ${{ inputs.rust-toolchain }}
-        targets: ${{ inputs.arch }}
+        target: ${{ inputs.arch }}
 
     - name: Start etcd # For integration tests.
       if: ${{ inputs.disable-run-tests == 'false' }}

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

@@ -4,9 +4,6 @@ inputs:
   arch:
     description: Architecture to build
     required: true
-  rust-toolchain:
-    description: Rust toolchain to use
-    required: true
   cargo-profile:
     description: Cargo profile to build
     required: true
@@ -28,10 +25,9 @@ runs:
     - uses: arduino/setup-protoc@v3
 
     - name: Install rust toolchain
-      uses: dtolnay/rust-toolchain@master
+      uses: actions-rust-lang/setup-rust-toolchain@v1
       with:
-        toolchain: ${{ inputs.rust-toolchain }}
-        targets: ${{ inputs.arch }}
+        target: ${{ inputs.arch }}
         components: llvm-tools-preview
 
     - name: Rust Cache

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

@@ -18,6 +18,8 @@ runs:
         --set replicaCount=${{ inputs.etcd-replicas }} \
         --set resources.requests.cpu=50m \
         --set resources.requests.memory=128Mi \
+        --set resources.limits.cpu=1000m \
+        --set resources.limits.memory=2Gi \
         --set auth.rbac.create=false \
         --set auth.rbac.token.enabled=false \
         --set persistence.size=2Gi \

.github/actions/start-runner/action.yml

@@ -38,7 +38,7 @@ runs:
   steps:
     - name: Configure AWS credentials
       if: startsWith(inputs.runner, 'ec2')
-      uses: aws-actions/configure-aws-credentials@v2
+      uses: aws-actions/configure-aws-credentials@v4
       with:
         aws-access-key-id: ${{ inputs.aws-access-key-id }}
         aws-secret-access-key: ${{ inputs.aws-secret-access-key }}

.github/actions/stop-runner/action.yml

@@ -25,7 +25,7 @@ runs:
   steps:
     - name: Configure AWS credentials
       if: ${{ inputs.label && inputs.ec2-instance-id }}
-      uses: aws-actions/configure-aws-credentials@v2
+      uses: aws-actions/configure-aws-credentials@v4
       with:
         aws-access-key-id: ${{ inputs.aws-access-key-id }}
         aws-secret-access-key: ${{ inputs.aws-secret-access-key }}

.github/workflows/apidoc.yml

@@ -12,9 +12,6 @@ on:
 
 name: Build API docs
 
-env:
-  RUST_TOOLCHAIN: nightly-2024-04-20
-
 jobs:
   apidoc:
     runs-on: ubuntu-20.04
@@ -23,9 +20,7 @@ jobs:
     - uses: arduino/setup-protoc@v3
       with:
         repo-token: ${{ secrets.GITHUB_TOKEN }}
-    - uses: dtolnay/rust-toolchain@master
-      with:
-        toolchain: ${{ env.RUST_TOOLCHAIN }}
+    - uses: actions-rust-lang/setup-rust-toolchain@v1
     - run: cargo doc --workspace --no-deps --document-private-items
     - run: |
         cat <<EOF > target/doc/index.html

.github/workflows/dev-build.yml

@@ -177,6 +177,8 @@ jobs:
           disable-run-tests: ${{ env.DISABLE_RUN_TESTS }}
           dev-mode: true # Only build the standard greptime binary.
           working-dir: ${{ env.CHECKOUT_GREPTIMEDB_PATH }}
+          image-registry: ${{ vars.ECR_IMAGE_REGISTRY }}
+          image-namespace: ${{ vars.ECR_IMAGE_NAMESPACE }}
 
   build-linux-arm64-artifacts:
     name: Build linux-arm64 artifacts
@@ -206,6 +208,8 @@ jobs:
           disable-run-tests: ${{ env.DISABLE_RUN_TESTS }}
           dev-mode: true # Only build the standard greptime binary.
           working-dir: ${{ env.CHECKOUT_GREPTIMEDB_PATH }}
+          image-registry: ${{ vars.ECR_IMAGE_REGISTRY }}
+          image-namespace: ${{ vars.ECR_IMAGE_NAMESPACE }}
 
   release-images-to-dockerhub:
     name: Build and push images to DockerHub

.github/workflows/develop.yml

@@ -29,9 +29,6 @@ concurrency:
   group: ${{ github.workflow }}-${{ github.head_ref || github.run_id }}
   cancel-in-progress: true
 
-env:
-  RUST_TOOLCHAIN: nightly-2024-04-20
-
 jobs:
   check-typos-and-docs:
     name: Check typos and docs
@@ -64,9 +61,7 @@ jobs:
       - uses: arduino/setup-protoc@v3
         with:
           repo-token: ${{ secrets.GITHUB_TOKEN }}
-      - uses: dtolnay/rust-toolchain@master
-        with:
-          toolchain: ${{ env.RUST_TOOLCHAIN }}
+      - uses: actions-rust-lang/setup-rust-toolchain@v1
       - name: Rust Cache
         uses: Swatinem/rust-cache@v2
         with:
@@ -82,9 +77,7 @@ jobs:
     timeout-minutes: 60
     steps:
       - uses: actions/checkout@v4
-      - uses: dtolnay/rust-toolchain@master
-        with:
-          toolchain: stable
+      - uses: actions-rust-lang/setup-rust-toolchain@v1
       - name: Rust Cache
         uses: Swatinem/rust-cache@v2
         with:
@@ -107,9 +100,7 @@ jobs:
       - uses: arduino/setup-protoc@v3
         with:
           repo-token: ${{ secrets.GITHUB_TOKEN }}
-      - uses: dtolnay/rust-toolchain@master
-        with:
-          toolchain: ${{ env.RUST_TOOLCHAIN }}
+      - uses: actions-rust-lang/setup-rust-toolchain@v1
       - uses: Swatinem/rust-cache@v2
         with:
           # Shares across multiple jobs
@@ -161,9 +152,7 @@ jobs:
       - uses: arduino/setup-protoc@v3
         with:
           repo-token: ${{ secrets.GITHUB_TOKEN }}
-      - uses: dtolnay/rust-toolchain@master
-        with:
-          toolchain: ${{ env.RUST_TOOLCHAIN }}
+      - uses: actions-rust-lang/setup-rust-toolchain@v1
       - name: Rust Cache
         uses: Swatinem/rust-cache@v2
         with:
@@ -181,7 +170,7 @@ jobs:
           name: bins
           path: .
       - name: Unzip binaries
-        run: | 
+        run: |
           tar -xvf ./bins.tar.gz
           rm ./bins.tar.gz
       - name: Run GreptimeDB
@@ -221,9 +210,7 @@ jobs:
       - uses: arduino/setup-protoc@v3
         with:
           repo-token: ${{ secrets.GITHUB_TOKEN }}
-      - uses: dtolnay/rust-toolchain@master
-        with:
-          toolchain: ${{ env.RUST_TOOLCHAIN }}
+      - uses: actions-rust-lang/setup-rust-toolchain@v1
       - name: Rust Cache
         uses: Swatinem/rust-cache@v2
         with:
@@ -274,9 +261,7 @@ jobs:
       - uses: arduino/setup-protoc@v3
         with:
           repo-token: ${{ secrets.GITHUB_TOKEN }}
-      - uses: dtolnay/rust-toolchain@master
-        with:
-          toolchain: ${{ env.RUST_TOOLCHAIN }}
+      - uses: actions-rust-lang/setup-rust-toolchain@v1
       - uses: Swatinem/rust-cache@v2
         with:
           # Shares across multiple jobs
@@ -287,7 +272,7 @@ jobs:
       - name: Build greptime bianry
         shell: bash
         # `cargo gc` will invoke `cargo build` with specified args
-        run: cargo gc --profile ci -- --bin greptime 
+        run: cargo gc --profile ci -- --bin greptime
       - name: Pack greptime binary
         shell: bash
         run: |
@@ -301,7 +286,7 @@ jobs:
           artifacts-dir: bin
           version: current
 
-  distributed-fuzztest: 
+  distributed-fuzztest:
     name: Fuzz Test (Distributed, ${{ matrix.mode.name }}, ${{ matrix.target }})
     runs-on: ubuntu-latest
     needs:  build-greptime-ci
@@ -344,9 +329,7 @@ jobs:
       - uses: arduino/setup-protoc@v3
         with:
           repo-token: ${{ secrets.GITHUB_TOKEN }}
-      - uses: dtolnay/rust-toolchain@master
-        with:
-          toolchain: ${{ env.RUST_TOOLCHAIN }}
+      - uses: actions-rust-lang/setup-rust-toolchain@v1
       - name: Rust Cache
         uses: Swatinem/rust-cache@v2
         with:
@@ -416,12 +399,12 @@ jobs:
       - name: Describe Nodes
         if: failure()
         shell: bash
-        run: | 
-          kubectl describe nodes      
+        run: |
+          kubectl describe nodes
       - name: Export kind logs
         if: failure()
         shell: bash
-        run: | 
+        run: |
           kind export logs /tmp/kind
       - name: Upload logs
         if: failure()
@@ -433,13 +416,13 @@ jobs:
       - name: Delete cluster
         if: success()
         shell: bash
-        run: | 
+        run: |
           kind delete cluster
           docker stop $(docker ps -a -q)
           docker rm $(docker ps -a -q)
           docker system prune -f
 
-  distributed-fuzztest-with-chaos: 
+  distributed-fuzztest-with-chaos:
     name: Fuzz Test with Chaos (Distributed, ${{ matrix.mode.name }}, ${{ matrix.target }})
     runs-on: ubuntu-latest
     needs:  build-greptime-ci
@@ -447,7 +430,7 @@ jobs:
     strategy:
       matrix:
         target: ["fuzz_migrate_mito_regions", "fuzz_failover_mito_regions", "fuzz_failover_metric_regions"]
-        mode: 
+        mode:
           - name: "Remote WAL"
             minio: true
             kafka: true
@@ -484,9 +467,7 @@ jobs:
       - uses: arduino/setup-protoc@v3
         with:
           repo-token: ${{ secrets.GITHUB_TOKEN }}
-      - uses: dtolnay/rust-toolchain@master
-        with:
-          toolchain: ${{ env.RUST_TOOLCHAIN }}
+      - uses: actions-rust-lang/setup-rust-toolchain@v1
       - name: Rust Cache
         uses: Swatinem/rust-cache@v2
         with:
@@ -557,12 +538,12 @@ jobs:
       - name: Describe Nodes
         if: failure()
         shell: bash
-        run: | 
-          kubectl describe nodes      
+        run: |
+          kubectl describe nodes
       - name: Export kind logs
         if: failure()
         shell: bash
-        run: | 
+        run: |
           kind export logs /tmp/kind
       - name: Upload logs
         if: failure()
@@ -574,7 +555,7 @@ jobs:
       - name: Delete cluster
         if: success()
         shell: bash
-        run: | 
+        run: |
           kind delete cluster
           docker stop $(docker ps -a -q)
           docker rm $(docker ps -a -q)
@@ -627,17 +608,16 @@ jobs:
       - uses: arduino/setup-protoc@v3
         with:
           repo-token: ${{ secrets.GITHUB_TOKEN }}
-      - uses: dtolnay/rust-toolchain@master
+      - uses: actions-rust-lang/setup-rust-toolchain@v1
         with:
-          toolchain: ${{ env.RUST_TOOLCHAIN }}
           components: rustfmt
       - name: Rust Cache
         uses: Swatinem/rust-cache@v2
         with:
           # Shares across multiple jobs
           shared-key: "check-rust-fmt"
-      - name: Run cargo fmt
-        run: cargo fmt --all -- --check
+      - name: Check format
+        run: make fmt-check
 
   clippy:
     name: Clippy
@@ -648,9 +628,8 @@ jobs:
       - uses: arduino/setup-protoc@v3
         with:
           repo-token: ${{ secrets.GITHUB_TOKEN }}
-      - uses: dtolnay/rust-toolchain@master
+      - uses: actions-rust-lang/setup-rust-toolchain@v1
         with:
-          toolchain: ${{ env.RUST_TOOLCHAIN }}
           components: clippy
       - name: Rust Cache
         uses: Swatinem/rust-cache@v2
@@ -674,9 +653,8 @@ jobs:
         with:
           version: "14.0"
       - name: Install toolchain
-        uses: dtolnay/rust-toolchain@master
+        uses: actions-rust-lang/setup-rust-toolchain@v1
         with:
-          toolchain: ${{ env.RUST_TOOLCHAIN }}
           components: llvm-tools-preview
       - name: Rust Cache
         uses: Swatinem/rust-cache@v2

.github/workflows/nightly-build.yml

@@ -154,6 +154,8 @@ jobs:
           cargo-profile: ${{ env.CARGO_PROFILE }}
           version: ${{ needs.allocate-runners.outputs.version }}
           disable-run-tests: ${{ env.DISABLE_RUN_TESTS }}
+          image-registry: ${{ vars.ECR_IMAGE_REGISTRY }}
+          image-namespace: ${{ vars.ECR_IMAGE_NAMESPACE }}
 
   build-linux-arm64-artifacts:
     name: Build linux-arm64 artifacts
@@ -173,6 +175,8 @@ jobs:
           cargo-profile: ${{ env.CARGO_PROFILE }}
           version: ${{ needs.allocate-runners.outputs.version }}
           disable-run-tests: ${{ env.DISABLE_RUN_TESTS }}
+          image-registry: ${{ vars.ECR_IMAGE_REGISTRY }}
+          image-namespace: ${{ vars.ECR_IMAGE_NAMESPACE }}
 
   release-images-to-dockerhub:
     name: Build and push images to DockerHub

.github/workflows/nightly-ci.yml

@@ -9,9 +9,6 @@ concurrency:
   group: ${{ github.workflow }}-${{ github.head_ref || github.run_id }}
   cancel-in-progress: true
 
-env:
-  RUST_TOOLCHAIN: nightly-2024-04-20
-
 permissions:
   issues: write
 
@@ -52,9 +49,7 @@ jobs:
       - uses: arduino/setup-protoc@v3
         with:
           repo-token: ${{ secrets.GITHUB_TOKEN }}
-      - uses: dtolnay/rust-toolchain@master
-        with:
-          toolchain: ${{ env.RUST_TOOLCHAIN }}
+      - uses: actions-rust-lang/setup-rust-toolchain@v1
       - name: Rust Cache
         uses: Swatinem/rust-cache@v2
       - name: Run sqlness
@@ -85,9 +80,8 @@ jobs:
         with:
           version: "14.0"
       - name: Install Rust toolchain
-        uses: dtolnay/rust-toolchain@master
+        uses: actions-rust-lang/setup-rust-toolchain@v1
         with:
-          toolchain: ${{ env.RUST_TOOLCHAIN }}
           components: llvm-tools-preview
       - name: Rust Cache
         uses: Swatinem/rust-cache@v2

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

@@ -1,12 +1,14 @@
 name: Release dev-builder images
 
 on:
+  push:
+    branches:
+      - main
+    paths:
+      - rust-toolchain.toml
+      - 'docker/dev-builder/**'
   workflow_dispatch: # Allows you to run this workflow manually.
     inputs:
-      version:
-        description: Version of the dev-builder
-        required: false
-        default: latest
       release_dev_builder_ubuntu_image:
         type: boolean
         description: Release dev-builder-ubuntu image
@@ -28,22 +30,103 @@ jobs:
     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.
     runs-on: ubuntu-20.04-16-cores
+    outputs:
+      version: ${{ steps.set-version.outputs.version }}
     steps:
       - name: Checkout
         uses: actions/checkout@v4
         with:
           fetch-depth: 0
 
+      - name: Configure build image version
+        id: set-version
+        shell: bash
+        run: |
+          commitShortSHA=`echo ${{ github.sha }} | cut -c1-8`
+          buildTime=`date +%Y%m%d%H%M%S`
+          BUILD_VERSION="$commitShortSHA-$buildTime"
+          RUST_TOOLCHAIN_VERSION=$(cat rust-toolchain.toml | grep -Eo '[0-9]{4}-[0-9]{2}-[0-9]{2}')
+          IMAGE_VERSION="${RUST_TOOLCHAIN_VERSION}-${BUILD_VERSION}"
+          echo "VERSION=${IMAGE_VERSION}" >> $GITHUB_ENV
+          echo "version=$IMAGE_VERSION" >> $GITHUB_OUTPUT
+
       - name: Build and push dev builder images
         uses: ./.github/actions/build-dev-builder-images
         with:
-          version: ${{ inputs.version }}
+          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 }}
 
+  release-dev-builder-images-ecr:
+    name: Release dev builder images to AWS ECR
+    runs-on: ubuntu-20.04
+    needs: [
+      release-dev-builder-images
+    ]
+    steps:
+      - name: Configure AWS credentials
+        uses: aws-actions/configure-aws-credentials@v4
+        with:
+          aws-access-key-id: ${{ secrets.AWS_ECR_ACCESS_KEY_ID }}
+          aws-secret-access-key: ${{ secrets.AWS_ECR_SECRET_ACCESS_KEY }}
+          aws-region: ${{ vars.ECR_REGION }}
+
+      - name: Login to Amazon ECR
+        id: login-ecr-public
+        uses: aws-actions/amazon-ecr-login@v2
+        env:
+          AWS_REGION: ${{ vars.ECR_REGION }}
+        with:
+          registry-type: public
+
+      - name: Push dev-builder-ubuntu image
+        shell: bash
+        if: ${{ inputs.release_dev_builder_ubuntu_image }}
+        run: |
+          docker run -v "${DOCKER_CONFIG:-$HOME/.docker}:/root/.docker:ro" \
+            -e "REGISTRY_AUTH_FILE=/root/.docker/config.json" \
+            quay.io/skopeo/stable:latest \
+            copy -a docker://docker.io/${{ vars.IMAGE_NAMESPACE }}/dev-builder-ubuntu:${{ needs.release-dev-builder-images.outputs.version }} \
+            docker://${{ vars.ECR_IMAGE_REGISTRY }}/${{ vars.ECR_IMAGE_NAMESPACE }}/dev-builder-ubuntu:${{ needs.release-dev-builder-images.outputs.version }}
+
+          docker run -v "${DOCKER_CONFIG:-$HOME/.docker}:/root/.docker:ro" \
+            -e "REGISTRY_AUTH_FILE=/root/.docker/config.json" \
+            quay.io/skopeo/stable:latest \
+            copy -a docker://docker.io/${{ vars.IMAGE_NAMESPACE }}/dev-builder-ubuntu:latest \
+            docker://${{ vars.ECR_IMAGE_REGISTRY }}/${{ vars.ECR_IMAGE_NAMESPACE }}/dev-builder-ubuntu:latest
+      - name: Push dev-builder-centos image
+        shell: bash
+        if: ${{ inputs.release_dev_builder_centos_image }}
+        run: |
+          docker run -v "${DOCKER_CONFIG:-$HOME/.docker}:/root/.docker:ro" \
+            -e "REGISTRY_AUTH_FILE=/root/.docker/config.json" \
+            quay.io/skopeo/stable:latest \
+            copy -a docker://docker.io/${{ vars.IMAGE_NAMESPACE }}/dev-builder-centos:${{ needs.release-dev-builder-images.outputs.version }} \
+            docker://${{ vars.ECR_IMAGE_REGISTRY }}/${{ vars.ECR_IMAGE_NAMESPACE }}/dev-builder-centos:${{ needs.release-dev-builder-images.outputs.version }}
+
+          docker run -v "${DOCKER_CONFIG:-$HOME/.docker}:/root/.docker:ro" \
+            -e "REGISTRY_AUTH_FILE=/root/.docker/config.json" \
+            quay.io/skopeo/stable:latest \
+            copy -a docker://docker.io/${{ vars.IMAGE_NAMESPACE }}/dev-builder-centos:latest \
+            docker://${{ vars.ECR_IMAGE_REGISTRY }}/${{ vars.ECR_IMAGE_NAMESPACE }}/dev-builder-centos:latest
+      - name: Push dev-builder-android image
+        shell: bash
+        if: ${{ inputs.release_dev_builder_android_image }}
+        run: |
+          docker run -v "${DOCKER_CONFIG:-$HOME/.docker}:/root/.docker:ro" \
+            -e "REGISTRY_AUTH_FILE=/root/.docker/config.json" \
+            quay.io/skopeo/stable:latest \
+            copy -a docker://docker.io/${{ vars.IMAGE_NAMESPACE }}/dev-builder-android:${{ needs.release-dev-builder-images.outputs.version }} \
+            docker://${{ vars.ECR_IMAGE_REGISTRY }}/${{ vars.ECR_IMAGE_NAMESPACE }}/dev-builder-android:${{ needs.release-dev-builder-images.outputs.version }}
+
+          docker run -v "${DOCKER_CONFIG:-$HOME/.docker}:/root/.docker:ro" \
+            -e "REGISTRY_AUTH_FILE=/root/.docker/config.json" \
+            quay.io/skopeo/stable:latest \
+            copy -a docker://docker.io/${{ vars.IMAGE_NAMESPACE }}/dev-builder-android:latest \
+            docker://${{ vars.ECR_IMAGE_REGISTRY }}/${{ vars.ECR_IMAGE_NAMESPACE }}/dev-builder-android:latest
   release-dev-builder-images-cn: # Note: Be careful issue: https://github.com/containers/skopeo/issues/1874 and we decide to use the latest stable skopeo container.
     name: Release dev builder images to CN region
     runs-on: ubuntu-20.04
@@ -51,35 +134,39 @@ jobs:
       release-dev-builder-images
     ]
     steps:
+      - name: Login to AliCloud Container Registry
+        uses: docker/login-action@v3
+        with:
+          registry: ${{ vars.ACR_IMAGE_REGISTRY }}
+          username: ${{ secrets.ALICLOUD_USERNAME }}
+          password: ${{ secrets.ALICLOUD_PASSWORD }}
+
       - name: Push dev-builder-ubuntu image
         shell: bash
         if: ${{ inputs.release_dev_builder_ubuntu_image }}
-        env:
-          DST_REGISTRY_USERNAME: ${{ secrets.ALICLOUD_USERNAME }}
-          DST_REGISTRY_PASSWORD: ${{ secrets.ALICLOUD_PASSWORD }}
         run: |
-          docker run quay.io/skopeo/stable:latest copy -a docker://docker.io/${{ vars.IMAGE_NAMESPACE }}/dev-builder-ubuntu:${{ inputs.version }} \
-            --dest-creds "$DST_REGISTRY_USERNAME":"$DST_REGISTRY_PASSWORD" \
-            docker://${{ vars.ACR_IMAGE_REGISTRY }}/${{ vars.IMAGE_NAMESPACE }}/dev-builder-ubuntu:${{ inputs.version }}
+          docker run -v "${DOCKER_CONFIG:-$HOME/.docker}:/root/.docker:ro" \
+            -e "REGISTRY_AUTH_FILE=/root/.docker/config.json" \
+            quay.io/skopeo/stable:latest \
+            copy -a docker://docker.io/${{ vars.IMAGE_NAMESPACE }}/dev-builder-ubuntu:${{ needs.release-dev-builder-images.outputs.version }} \
+            docker://${{ vars.ACR_IMAGE_REGISTRY }}/${{ vars.IMAGE_NAMESPACE }}/dev-builder-ubuntu:${{ needs.release-dev-builder-images.outputs.version }}
 
       - name: Push dev-builder-centos image
         shell: bash
         if: ${{ inputs.release_dev_builder_centos_image }}
-        env:
-          DST_REGISTRY_USERNAME: ${{ secrets.ALICLOUD_USERNAME }}
-          DST_REGISTRY_PASSWORD: ${{ secrets.ALICLOUD_PASSWORD }}
         run: |
-          docker run quay.io/skopeo/stable:latest copy -a docker://docker.io/${{ vars.IMAGE_NAMESPACE }}/dev-builder-centos:${{ inputs.version }} \
-            --dest-creds "$DST_REGISTRY_USERNAME":"$DST_REGISTRY_PASSWORD" \
-            docker://${{ vars.ACR_IMAGE_REGISTRY }}/${{ vars.IMAGE_NAMESPACE }}/dev-builder-centos:${{ inputs.version }}
+          docker run -v "${DOCKER_CONFIG:-$HOME/.docker}:/root/.docker:ro" \
+            -e "REGISTRY_AUTH_FILE=/root/.docker/config.json" \
+            quay.io/skopeo/stable:latest \
+            copy -a docker://docker.io/${{ vars.IMAGE_NAMESPACE }}/dev-builder-centos:${{ needs.release-dev-builder-images.outputs.version }} \
+            docker://${{ vars.ACR_IMAGE_REGISTRY }}/${{ vars.IMAGE_NAMESPACE }}/dev-builder-centos:${{ needs.release-dev-builder-images.outputs.version }}
 
       - name: Push dev-builder-android image
         shell: bash
         if: ${{ inputs.release_dev_builder_android_image }}
-        env:
-          DST_REGISTRY_USERNAME: ${{ secrets.ALICLOUD_USERNAME }}
-          DST_REGISTRY_PASSWORD: ${{ secrets.ALICLOUD_PASSWORD }}
         run: |
-          docker run quay.io/skopeo/stable:latest copy -a docker://docker.io/${{ vars.IMAGE_NAMESPACE }}/dev-builder-android:${{ inputs.version }} \
-            --dest-creds "$DST_REGISTRY_USERNAME":"$DST_REGISTRY_PASSWORD" \
-            docker://${{ vars.ACR_IMAGE_REGISTRY }}/${{ vars.IMAGE_NAMESPACE }}/dev-builder-android:${{ inputs.version }}
+          docker run -v "${DOCKER_CONFIG:-$HOME/.docker}:/root/.docker:ro" \
+            -e "REGISTRY_AUTH_FILE=/root/.docker/config.json" \
+            quay.io/skopeo/stable:latest \
+            copy -a docker://docker.io/${{ vars.IMAGE_NAMESPACE }}/dev-builder-android:${{ needs.release-dev-builder-images.outputs.version }} \
+            docker://${{ vars.ACR_IMAGE_REGISTRY }}/${{ vars.IMAGE_NAMESPACE }}/dev-builder-android:${{ needs.release-dev-builder-images.outputs.version }}

.github/workflows/release.yml

@@ -33,6 +33,7 @@ on:
         description: The runner uses to build linux-arm64 artifacts
         default: ec2-c6g.4xlarge-arm64
         options:
+          - ubuntu-2204-32-cores-arm
           - ec2-c6g.xlarge-arm64 # 4C8G
           - ec2-c6g.2xlarge-arm64 # 8C16G
           - ec2-c6g.4xlarge-arm64 # 16C32G
@@ -82,7 +83,6 @@ on:
 # Use env variables to control all the release process.
 env:
   # The arguments of building greptime.
-  RUST_TOOLCHAIN: nightly-2024-04-20
   CARGO_PROFILE: nightly
 
   # Controls whether to run tests, include unit-test, integration-test and sqlness.
@@ -123,6 +123,11 @@ jobs:
         with:
           fetch-depth: 0
 
+      - name: Check Rust toolchain version
+        shell: bash
+        run: |
+          ./scripts/check-builder-rust-version.sh
+
       # 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;
@@ -183,6 +188,8 @@ jobs:
           cargo-profile: ${{ env.CARGO_PROFILE }}
           version: ${{ needs.allocate-runners.outputs.version }}
           disable-run-tests: ${{ env.DISABLE_RUN_TESTS }}
+          image-registry: ${{ vars.ECR_IMAGE_REGISTRY }}
+          image-namespace: ${{ vars.ECR_IMAGE_NAMESPACE }}
 
   build-linux-arm64-artifacts:
     name: Build linux-arm64 artifacts
@@ -202,6 +209,8 @@ jobs:
           cargo-profile: ${{ env.CARGO_PROFILE }}
           version: ${{ needs.allocate-runners.outputs.version }}
           disable-run-tests: ${{ env.DISABLE_RUN_TESTS }}
+          image-registry: ${{ vars.ECR_IMAGE_REGISTRY }}
+          image-namespace: ${{ vars.ECR_IMAGE_NAMESPACE }}
 
   build-macos-artifacts:
     name: Build macOS artifacts
@@ -240,11 +249,11 @@ jobs:
       - uses: ./.github/actions/build-macos-artifacts
         with:
           arch: ${{ matrix.arch }}
-          rust-toolchain: ${{ env.RUST_TOOLCHAIN }}
           cargo-profile: ${{ env.CARGO_PROFILE }}
           features: ${{ matrix.features }}
           version: ${{ needs.allocate-runners.outputs.version }}
-          disable-run-tests: ${{ env.DISABLE_RUN_TESTS }}
+          # We decide to disable the integration tests on macOS because it's unnecessary and time-consuming.
+          disable-run-tests: true
           artifacts-dir: ${{ matrix.artifacts-dir-prefix }}-${{ needs.allocate-runners.outputs.version }}
 
       - name: Set build macos result
@@ -283,7 +292,6 @@ jobs:
       - uses: ./.github/actions/build-windows-artifacts
         with:
           arch: ${{ matrix.arch }}
-          rust-toolchain: ${{ env.RUST_TOOLCHAIN }}
           cargo-profile: ${{ env.CARGO_PROFILE }}
           features: ${{ matrix.features }}
           version: ${{ needs.allocate-runners.outputs.version }}

Cargo.lock

@@ -214,7 +214,7 @@ checksum = "d301b3b94cb4b2f23d7917810addbbaff90738e0ca2be692bd027e70d7e0330c"
 
 [[package]]
 name = "api"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "common-base",
  "common-decimal",
@@ -762,7 +762,7 @@ dependencies = [
 
 [[package]]
 name = "auth"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "api",
  "async-trait",
@@ -897,6 +897,15 @@ dependencies = [
  "tokio",
 ]
 
+[[package]]
+name = "backon"
+version = "1.0.2"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "2caf634d05fe0642d0fb1ab43497fa627088ecd93f84b2d0f2a5d7b91f7730db"
+dependencies = [
+ "fastrand",
+]
+
 [[package]]
 name = "backtrace"
 version = "0.3.73"
@@ -1166,6 +1175,17 @@ dependencies = [
  "regex-automata 0.1.10",
 ]
 
+[[package]]
+name = "bstr"
+version = "1.10.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "40723b8fb387abc38f4f4a37c09073622e41dd12327033091ef8950659e6dc0c"
+dependencies = [
+ "memchr",
+ "regex-automata 0.4.7",
+ "serde",
+]
+
 [[package]]
 name = "btoi"
 version = "0.4.3"
@@ -1277,7 +1297,7 @@ dependencies = [
 
 [[package]]
 name = "cache"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "catalog",
  "common-error",
@@ -1285,7 +1305,7 @@ dependencies = [
  "common-meta",
  "moka",
  "snafu 0.8.4",
- "substrait 0.9.2",
+ "substrait 0.9.3",
 ]
 
 [[package]]
@@ -1312,7 +1332,7 @@ checksum = "37b2a672a2cb129a2e41c10b1224bb368f9f37a2b16b612598138befd7b37eb5"
 
 [[package]]
 name = "catalog"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "api",
  "arrow",
@@ -1638,7 +1658,7 @@ checksum = "4b82cf0babdbd58558212896d1a4272303a57bdb245c2bf1147185fb45640e70"
 
 [[package]]
 name = "client"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "api",
  "arc-swap",
@@ -1668,7 +1688,7 @@ dependencies = [
  "serde_json",
  "snafu 0.8.4",
  "substrait 0.37.3",
- "substrait 0.9.2",
+ "substrait 0.9.3",
  "tokio",
  "tokio-stream",
  "tonic 0.11.0",
@@ -1698,7 +1718,7 @@ dependencies = [
 
 [[package]]
 name = "cmd"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "async-trait",
  "auth",
@@ -1752,9 +1772,10 @@ dependencies = [
  "serde_json",
  "servers",
  "session",
+ "similar-asserts",
  "snafu 0.8.4",
  "store-api",
- "substrait 0.9.2",
+ "substrait 0.9.3",
  "table",
  "temp-env",
  "tempfile",
@@ -1800,13 +1821,15 @@ checksum = "55b672471b4e9f9e95499ea597ff64941a309b2cdbffcc46f2cc5e2d971fd335"
 
 [[package]]
 name = "common-base"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "anymap",
+ "async-trait",
  "bitvec",
  "bytes",
  "common-error",
  "common-macro",
+ "futures",
  "paste",
  "serde",
  "snafu 0.8.4",
@@ -1816,7 +1839,7 @@ dependencies = [
 
 [[package]]
 name = "common-catalog"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "chrono",
  "common-error",
@@ -1827,7 +1850,7 @@ dependencies = [
 
 [[package]]
 name = "common-config"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "common-base",
  "common-error",
@@ -1850,7 +1873,7 @@ dependencies = [
 
 [[package]]
 name = "common-datasource"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "arrow",
  "arrow-schema",
@@ -1887,7 +1910,7 @@ dependencies = [
 
 [[package]]
 name = "common-decimal"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "bigdecimal",
  "common-error",
@@ -1900,7 +1923,7 @@ dependencies = [
 
 [[package]]
 name = "common-error"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "snafu 0.8.4",
  "strum 0.25.0",
@@ -1909,7 +1932,7 @@ dependencies = [
 
 [[package]]
 name = "common-frontend"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "api",
  "async-trait",
@@ -1924,7 +1947,7 @@ dependencies = [
 
 [[package]]
 name = "common-function"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "api",
  "arc-swap",
@@ -1941,6 +1964,9 @@ dependencies = [
  "common-version",
  "datafusion",
  "datatypes",
+ "geohash",
+ "h3o",
+ "jsonb",
  "num",
  "num-traits",
  "once_cell",
@@ -1954,11 +1980,12 @@ dependencies = [
  "statrs",
  "store-api",
  "table",
+ "tokio",
 ]
 
 [[package]]
 name = "common-greptimedb-telemetry"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "async-trait",
  "common-runtime",
@@ -1975,7 +2002,7 @@ dependencies = [
 
 [[package]]
 name = "common-grpc"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "api",
  "arrow-flight",
@@ -2001,7 +2028,7 @@ dependencies = [
 
 [[package]]
 name = "common-grpc-expr"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "api",
  "common-base",
@@ -2019,7 +2046,7 @@ dependencies = [
 
 [[package]]
 name = "common-macro"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "arc-swap",
  "common-query",
@@ -2033,7 +2060,7 @@ dependencies = [
 
 [[package]]
 name = "common-mem-prof"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "common-error",
  "common-macro",
@@ -2046,7 +2073,7 @@ dependencies = [
 
 [[package]]
 name = "common-meta"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "anymap2",
  "api",
@@ -2102,15 +2129,15 @@ dependencies = [
 
 [[package]]
 name = "common-plugins"
-version = "0.9.2"
+version = "0.9.3"
 
 [[package]]
 name = "common-procedure"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "async-stream",
  "async-trait",
- "backon",
+ "backon 1.0.2",
  "common-base",
  "common-error",
  "common-macro",
@@ -2132,7 +2159,7 @@ dependencies = [
 
 [[package]]
 name = "common-procedure-test"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "async-trait",
  "common-procedure",
@@ -2140,7 +2167,7 @@ dependencies = [
 
 [[package]]
 name = "common-query"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "api",
  "async-trait",
@@ -2166,7 +2193,7 @@ dependencies = [
 
 [[package]]
 name = "common-recordbatch"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "arc-swap",
  "common-error",
@@ -2185,7 +2212,7 @@ dependencies = [
 
 [[package]]
 name = "common-runtime"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "async-trait",
  "common-error",
@@ -2207,7 +2234,7 @@ dependencies = [
 
 [[package]]
 name = "common-telemetry"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "atty",
  "backtrace",
@@ -2234,7 +2261,7 @@ dependencies = [
 
 [[package]]
 name = "common-test-util"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "client",
  "common-query",
@@ -2246,7 +2273,7 @@ dependencies = [
 
 [[package]]
 name = "common-time"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "arrow",
  "chrono",
@@ -2262,7 +2289,7 @@ dependencies = [
 
 [[package]]
 name = "common-version"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "build-data",
  "const_format",
@@ -2273,7 +2300,7 @@ dependencies = [
 
 [[package]]
 name = "common-wal"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "common-base",
  "common-error",
@@ -2281,6 +2308,7 @@ dependencies = [
  "common-telemetry",
  "futures-util",
  "humantime-serde",
+ "num_cpus",
  "rskafka",
  "rustls 0.23.10",
  "rustls-native-certs",
@@ -3081,7 +3109,7 @@ dependencies = [
 
 [[package]]
 name = "datanode"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "api",
  "arrow-flight",
@@ -3130,7 +3158,7 @@ dependencies = [
  "session",
  "snafu 0.8.4",
  "store-api",
- "substrait 0.9.2",
+ "substrait 0.9.3",
  "table",
  "tokio",
  "toml 0.8.14",
@@ -3139,11 +3167,12 @@ dependencies = [
 
 [[package]]
 name = "datatypes"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "arrow",
  "arrow-array",
  "arrow-schema",
+ "base64 0.21.7",
  "common-base",
  "common-decimal",
  "common-error",
@@ -3152,6 +3181,8 @@ dependencies = [
  "common-time",
  "datafusion-common",
  "enum_dispatch",
+ "greptime-proto",
+ "jsonb",
  "num",
  "num-traits",
  "ordered-float 3.9.2",
@@ -3684,6 +3715,12 @@ version = "0.1.9"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "7360491ce676a36bf9bb3c56c1aa791658183a54d2744120f27285738d90465a"
 
+[[package]]
+name = "fast-float"
+version = "0.2.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "95765f67b4b18863968b4a1bd5bb576f732b29a4a28c7cd84c09fa3e2875f33c"
+
 [[package]]
 name = "fastdivide"
 version = "0.4.1"
@@ -3709,7 +3746,7 @@ dependencies = [
 
 [[package]]
 name = "file-engine"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "api",
  "async-trait",
@@ -3803,9 +3840,15 @@ dependencies = [
  "num-traits",
 ]
 
+[[package]]
+name = "float_eq"
+version = "1.0.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "28a80e3145d8ad11ba0995949bbcf48b9df2be62772b3d351ef017dff6ecb853"
+
 [[package]]
 name = "flow"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "api",
  "arrow",
@@ -3862,7 +3905,7 @@ dependencies = [
  "snafu 0.8.4",
  "store-api",
  "strum 0.25.0",
- "substrait 0.9.2",
+ "substrait 0.9.3",
  "table",
  "tokio",
  "tonic 0.11.0",
@@ -3909,7 +3952,7 @@ checksum = "6c2141d6d6c8512188a7891b4b01590a45f6dac67afb4f255c4124dbb86d4eaa"
 
 [[package]]
 name = "frontend"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "api",
  "arc-swap",
@@ -4201,6 +4244,27 @@ dependencies = [
  "version_check",
 ]
 
+[[package]]
+name = "geo-types"
+version = "0.7.13"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "9ff16065e5720f376fbced200a5ae0f47ace85fd70b7e54269790281353b6d61"
+dependencies = [
+ "approx",
+ "num-traits",
+ "serde",
+]
+
+[[package]]
+name = "geohash"
+version = "0.13.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "0fb94b1a65401d6cbf22958a9040aa364812c26674f841bee538b12c135db1e6"
+dependencies = [
+ "geo-types",
+ "libm",
+]
+
 [[package]]
 name = "gethostname"
 version = "0.2.3"
@@ -4261,7 +4325,7 @@ checksum = "d2fabcfbdc87f4758337ca535fb41a6d701b65693ce38287d856d1674551ec9b"
 [[package]]
 name = "greptime-proto"
 version = "0.1.0"
-source = "git+https://github.com/GreptimeTeam/greptime-proto.git?rev=c437b55725b7f5224fe9d46db21072b4a682ee4b#c437b55725b7f5224fe9d46db21072b4a682ee4b"
+source = "git+https://github.com/GreptimeTeam/greptime-proto.git?rev=973f49cde88a582fb65755cc572ebcf6fb93ccf7#973f49cde88a582fb65755cc572ebcf6fb93ccf7"
 dependencies = [
  "prost 0.12.6",
  "serde",
@@ -4291,6 +4355,25 @@ dependencies = [
  "tracing",
 ]
 
+[[package]]
+name = "h3o"
+version = "0.6.4"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "0de3592e1f699692aa0525c42ff7879ec3ee7e36329af20967bc910a1cdc39c7"
+dependencies = [
+ "ahash 0.8.11",
+ "either",
+ "float_eq",
+ "h3o-bit",
+ "libm",
+]
+
+[[package]]
+name = "h3o-bit"
+version = "0.1.1"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "6fb45e8060378c0353781abf67e1917b545a6b710d0342d85b70c125af7ef320"
+
 [[package]]
 name = "half"
 version = "1.8.3"
@@ -4568,9 +4651,9 @@ checksum = "df3b46402a9d5adb4c86a0cf463f42e19994e3ee891101b1841f30a545cb49a9"
 
 [[package]]
 name = "human-panic"
-version = "1.2.3"
+version = "2.0.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "c4f016c89920bbb30951a8405ecacbb4540db5524313b9445736e7e1855cf370"
+checksum = "1c5a08ed290eac04006e21e63d32e90086b6182c7cd0452d10f4264def1fec9a"
 dependencies = [
  "anstream",
  "anstyle",
@@ -5020,7 +5103,7 @@ dependencies = [
 
 [[package]]
 name = "index"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "async-trait",
  "asynchronous-codec",
@@ -5349,6 +5432,21 @@ dependencies = [
  "serde",
 ]
 
+[[package]]
+name = "jsonb"
+version = "0.4.1"
+source = "git+https://github.com/CookiePieWw/jsonb.git?rev=d0166c130fce903bf6c58643417a3173a6172d31#d0166c130fce903bf6c58643417a3173a6172d31"
+dependencies = [
+ "byteorder",
+ "fast-float",
+ "itoa",
+ "nom",
+ "ordered-float 4.2.0",
+ "rand",
+ "ryu",
+ "serde_json",
+]
+
 [[package]]
 name = "jsonpath-rust"
 version = "0.5.1"
@@ -5800,7 +5898,7 @@ checksum = "90ed8c1e510134f979dbc4f070f87d4313098b704861a105fe34231c70a3901c"
 
 [[package]]
 name = "log-store"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "async-stream",
  "async-trait",
@@ -5816,6 +5914,7 @@ dependencies = [
  "common-time",
  "common-wal",
  "delta-encoding",
+ "derive_builder 0.12.0",
  "futures",
  "futures-util",
  "itertools 0.10.5",
@@ -6111,7 +6210,7 @@ dependencies = [
 
 [[package]]
 name = "meta-client"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "api",
  "async-trait",
@@ -6137,7 +6236,7 @@ dependencies = [
 
 [[package]]
 name = "meta-srv"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "api",
  "async-trait",
@@ -6215,7 +6314,7 @@ dependencies = [
 
 [[package]]
 name = "metric-engine"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "api",
  "aquamarine",
@@ -6306,7 +6405,7 @@ dependencies = [
 
 [[package]]
 name = "mito2"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "api",
  "aquamarine",
@@ -6953,7 +7052,7 @@ dependencies = [
 
 [[package]]
 name = "object-store"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "anyhow",
  "bytes",
@@ -7010,13 +7109,13 @@ checksum = "0ab1bc2a289d34bd04a330323ac98a1b4bc82c9d9fcb1e66b63caa84da26b575"
 
 [[package]]
 name = "opendal"
-version = "0.48.0"
+version = "0.49.1"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "615d41187deea0ea7fab5b48e9afef6ae8fc742fdcfa248846ee3d92ff71e986"
+checksum = "ba615070686c8781ce97376fdafca29d7c42f47b31d2230d7c8c1642ec823950"
 dependencies = [
  "anyhow",
  "async-trait",
- "backon",
+ "backon 0.4.4",
  "base64 0.22.1",
  "bytes",
  "chrono",
@@ -7141,9 +7240,11 @@ version = "0.5.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "3a8fddc9b68f5b80dae9d6f510b88e02396f006ad48cac349411fbecc80caae4"
 dependencies = [
+ "hex",
  "opentelemetry 0.22.0",
  "opentelemetry_sdk 0.22.1",
  "prost 0.12.6",
+ "serde",
  "tonic 0.11.0",
 ]
 
@@ -7200,7 +7301,7 @@ dependencies = [
 
 [[package]]
 name = "operator"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "api",
  "async-trait",
@@ -7245,7 +7346,7 @@ dependencies = [
  "sql",
  "sqlparser 0.45.0 (git+https://github.com/GreptimeTeam/sqlparser-rs.git?rev=54a267ac89c09b11c0c88934690530807185d3e7)",
  "store-api",
- "substrait 0.9.2",
+ "substrait 0.9.3",
  "table",
  "tokio",
  "tokio-util",
@@ -7495,7 +7596,7 @@ dependencies = [
 
 [[package]]
 name = "partition"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "api",
  "async-trait",
@@ -7784,7 +7885,7 @@ checksum = "8b870d8c151b6f2fb93e84a13146138f05d02ed11c7e7c54f8826aaaf7c9f184"
 
 [[package]]
 name = "pipeline"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "ahash 0.8.11",
  "api",
@@ -7945,7 +8046,7 @@ dependencies = [
 
 [[package]]
 name = "plugins"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "auth",
  "common-base",
@@ -8001,6 +8102,8 @@ dependencies = [
  "chrono",
  "fallible-iterator",
  "postgres-protocol",
+ "serde",
+ "serde_json",
 ]
 
 [[package]]
@@ -8214,7 +8317,7 @@ dependencies = [
 
 [[package]]
 name = "promql"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "ahash 0.8.11",
  "async-trait",
@@ -8449,7 +8552,7 @@ dependencies = [
 
 [[package]]
 name = "puffin"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "async-compression 0.4.11",
  "async-trait",
@@ -8571,7 +8674,7 @@ dependencies = [
 
 [[package]]
 name = "query"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "ahash 0.8.11",
  "api",
@@ -8634,7 +8737,7 @@ dependencies = [
  "stats-cli",
  "store-api",
  "streaming-stats",
- "substrait 0.9.2",
+ "substrait 0.9.3",
  "table",
  "tokio",
  "tokio-stream",
@@ -9170,8 +9273,9 @@ dependencies = [
 
 [[package]]
 name = "rsasl"
-version = "2.0.2"
-source = "git+https://github.com/wenyxu/rsasl.git?rev=06ebb683d5539c3410de4ce9fa37ff9b97e790a4#06ebb683d5539c3410de4ce9fa37ff9b97e790a4"
+version = "2.1.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "45035615cdd68c71daac89aef75b130d4b2cad29599966e1b4671f8fbb463559"
 dependencies = [
  "base64 0.22.1",
  "core2",
@@ -9188,9 +9292,8 @@ dependencies = [
 [[package]]
 name = "rskafka"
 version = "0.5.0"
-source = "git+https://github.com/WenyXu/rskafka.git?rev=940c6030012c5b746fad819fb72e3325b26e39de#940c6030012c5b746fad819fb72e3325b26e39de"
+source = "git+https://github.com/influxdata/rskafka.git?rev=75535b5ad9bae4a5dbb582c82e44dfd81ec10105#75535b5ad9bae4a5dbb582c82e44dfd81ec10105"
 dependencies = [
- "async-trait",
  "bytes",
  "chrono",
  "crc32c",
@@ -9199,7 +9302,6 @@ dependencies = [
  "integer-encoding 4.0.0",
  "lz4",
  "parking_lot 0.12.3",
- "pin-project-lite",
  "rand",
  "rsasl",
  "rustls 0.23.10",
@@ -9567,7 +9669,7 @@ source = "git+https://github.com/discord9/RustPython?rev=9ed5137412#9ed51374125b
 dependencies = [
  "ascii",
  "bitflags 1.3.2",
- "bstr",
+ "bstr 0.2.17",
  "cfg-if",
  "hexf-parse",
  "itertools 0.10.5",
@@ -9602,7 +9704,7 @@ version = "0.2.0"
 source = "git+https://github.com/discord9/RustPython?rev=9ed5137412#9ed51374125b5f1a9e5cee5dd7e27023b8591f1e"
 dependencies = [
  "bitflags 1.3.2",
- "bstr",
+ "bstr 0.2.17",
  "itertools 0.10.5",
  "lz4_flex 0.9.5",
  "num-bigint",
@@ -9755,7 +9857,7 @@ dependencies = [
  "ascii",
  "atty",
  "bitflags 1.3.2",
- "bstr",
+ "bstr 0.2.17",
  "caseless",
  "cfg-if",
  "chrono",
@@ -9997,7 +10099,7 @@ checksum = "94143f37725109f92c262ed2cf5e59bce7498c01bcc1502d7b9afe439a4e9f49"
 
 [[package]]
 name = "script"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "api",
  "arc-swap",
@@ -10291,7 +10393,7 @@ dependencies = [
 
 [[package]]
 name = "servers"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "aide",
  "api",
@@ -10340,6 +10442,7 @@ dependencies = [
  "hyper 0.14.29",
  "influxdb_line_protocol",
  "itertools 0.10.5",
+ "jsonb",
  "lazy_static",
  "mime_guess",
  "mysql_async",
@@ -10397,7 +10500,7 @@ dependencies = [
 
 [[package]]
 name = "session"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "api",
  "arc-swap",
@@ -10522,6 +10625,26 @@ version = "0.1.4"
 source = "registry+https://github.com/rust-lang/crates.io-index"
 checksum = "f27f6278552951f1f2b8cf9da965d10969b2efdea95a6ec47987ab46edfe263a"
 
+[[package]]
+name = "similar"
+version = "2.6.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "1de1d4f81173b03af4c0cbed3c898f6bff5b870e4a7f5d6f4057d62a7a4b686e"
+dependencies = [
+ "bstr 1.10.0",
+ "unicode-segmentation",
+]
+
+[[package]]
+name = "similar-asserts"
+version = "1.6.0"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "cfe85670573cd6f0fa97940f26e7e6601213c3b0555246c24234131f88c5709e"
+dependencies = [
+ "console",
+ "similar",
+]
+
 [[package]]
 name = "simple_asn1"
 version = "0.6.2"
@@ -10698,7 +10821,7 @@ dependencies = [
 
 [[package]]
 name = "sql"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "api",
  "chrono",
@@ -10719,6 +10842,7 @@ dependencies = [
  "hex",
  "iso8601",
  "itertools 0.10.5",
+ "jsonb",
  "lazy_static",
  "regex",
  "serde_json",
@@ -10758,7 +10882,7 @@ dependencies = [
 
 [[package]]
 name = "sqlness-runner"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "async-trait",
  "clap 4.5.7",
@@ -10975,7 +11099,7 @@ dependencies = [
 
 [[package]]
 name = "store-api"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "api",
  "aquamarine",
@@ -11144,7 +11268,7 @@ dependencies = [
 
 [[package]]
 name = "substrait"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "async-trait",
  "bytes",
@@ -11345,7 +11469,7 @@ dependencies = [
 
 [[package]]
 name = "table"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "api",
  "async-trait",
@@ -11610,7 +11734,7 @@ checksum = "3369f5ac52d5eb6ab48c6b4ffdc8efbcad6b89c765749064ba298f2c68a16a76"
 
 [[package]]
 name = "tests-fuzz"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "arbitrary",
  "async-trait",
@@ -11652,7 +11776,7 @@ dependencies = [
 
 [[package]]
 name = "tests-integration"
-version = "0.9.2"
+version = "0.9.3"
 dependencies = [
  "api",
  "arrow-flight",
@@ -11685,6 +11809,7 @@ dependencies = [
  "datanode",
  "datatypes",
  "dotenv",
+ "flate2",
  "flow",
  "frontend",
  "futures",
@@ -11708,11 +11833,12 @@ dependencies = [
  "serde_json",
  "servers",
  "session",
+ "similar-asserts",
  "snafu 0.8.4",
  "sql",
  "sqlx",
  "store-api",
- "substrait 0.9.2",
+ "substrait 0.9.3",
  "table",
  "tempfile",
  "time",
@@ -11800,9 +11926,9 @@ dependencies = [
 
 [[package]]
 name = "tikv-jemalloc-ctl"
-version = "0.5.4"
+version = "0.6.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "619bfed27d807b54f7f776b9430d4f8060e66ee138a28632ca898584d462c31c"
+checksum = "f21f216790c8df74ce3ab25b534e0718da5a1916719771d3fec23315c99e468b"
 dependencies = [
  "libc",
  "paste",
@@ -11811,9 +11937,9 @@ dependencies = [
 
 [[package]]
 name = "tikv-jemalloc-sys"
-version = "0.5.4+5.3.0-patched"
+version = "0.6.0+5.3.0-1-ge13ca993e8ccb9ba9847cc330696e02839f328f7"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "9402443cb8fd499b6f327e40565234ff34dbda27460c5b47db0db77443dd85d1"
+checksum = "cd3c60906412afa9c2b5b5a48ca6a5abe5736aec9eb48ad05037a677e52e4e2d"
 dependencies = [
  "cc",
  "libc",
@@ -11821,9 +11947,9 @@ dependencies = [
 
 [[package]]
 name = "tikv-jemallocator"
-version = "0.5.4"
+version = "0.6.0"
 source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "965fe0c26be5c56c94e38ba547249074803efd52adfb66de62107d95aab3eaca"
+checksum = "4cec5ff18518d81584f477e9bfdf957f5bb0979b0bac3af4ca30b5b3ae2d2865"
 dependencies = [
  "libc",
  "tikv-jemalloc-sys",
@@ -12431,6 +12557,16 @@ dependencies = [
  "web-time 0.2.4",
 ]
 
+[[package]]
+name = "tracing-serde"
+version = "0.1.3"
+source = "registry+https://github.com/rust-lang/crates.io-index"
+checksum = "bc6b213177105856957181934e4920de57730fc69bf42c37ee5bb664d406d9e1"
+dependencies = [
+ "serde",
+ "tracing-core",
+]
+
 [[package]]
 name = "tracing-subscriber"
 version = "0.3.18"
@@ -12441,12 +12577,15 @@ dependencies = [
  "nu-ansi-term",
  "once_cell",
  "regex",
+ "serde",
+ "serde_json",
  "sharded-slab",
  "smallvec",
  "thread_local",
  "tracing",
  "tracing-core",
  "tracing-log 0.2.0",
+ "tracing-serde",
 ]
 
 [[package]]

Cargo.toml

@@ -64,7 +64,7 @@ members = [
 resolver = "2"
 
 [workspace.package]
-version = "0.9.2"
+version = "0.9.3"
 edition = "2021"
 license = "Apache-2.0"
 
@@ -77,6 +77,7 @@ clippy.readonly_write_lock = "allow"
 rust.unknown_lints = "deny"
 # Remove this after https://github.com/PyO3/pyo3/issues/4094
 rust.non_local_definitions = "allow"
+rust.unexpected_cfgs = { level = "warn", check-cfg = ['cfg(tokio_unstable)'] }
 
 [workspace.dependencies]
 # We turn off default-features for some dependencies here so the workspaces which inherit them can
@@ -119,10 +120,11 @@ etcd-client = { version = "0.13" }
 fst = "0.4.7"
 futures = "0.3"
 futures-util = "0.3"
-greptime-proto = { git = "https://github.com/GreptimeTeam/greptime-proto.git", rev = "c437b55725b7f5224fe9d46db21072b4a682ee4b" }
+greptime-proto = { git = "https://github.com/GreptimeTeam/greptime-proto.git", rev = "973f49cde88a582fb65755cc572ebcf6fb93ccf7" }
 humantime = "2.1"
 humantime-serde = "1.1"
 itertools = "0.10"
+jsonb = { git = "https://github.com/CookiePieWw/jsonb.git", rev = "d0166c130fce903bf6c58643417a3173a6172d31", default-features = false }
 lazy_static = "1.4"
 meter-core = { git = "https://github.com/GreptimeTeam/greptime-meter.git", rev = "80eb97c24c88af4dd9a86f8bbaf50e741d4eb8cd" }
 mockall = "0.11.4"
@@ -134,6 +136,7 @@ opentelemetry-proto = { version = "0.5", features = [
     "gen-tonic",
     "metrics",
     "trace",
+    "with-serde",
 ] }
 parquet = { version = "51.0.0", default-features = false, features = ["arrow", "async", "object_store"] }
 paste = "1.0"
@@ -151,8 +154,7 @@ reqwest = { version = "0.12", default-features = false, features = [
     "stream",
     "multipart",
 ] }
-# SCRAM-SHA-512 requires https://github.com/dequbed/rsasl/pull/48, https://github.com/influxdata/rskafka/pull/247 
-rskafka = { git = "https://github.com/WenyXu/rskafka.git", rev = "940c6030012c5b746fad819fb72e3325b26e39de", features = [
+rskafka = { git = "https://github.com/influxdata/rskafka.git", rev = "75535b5ad9bae4a5dbb582c82e44dfd81ec10105", features = [
     "transport-tls",
 ] }
 rstest = "0.21"
@@ -167,6 +169,7 @@ shadow-rs = "0.31"
 smallvec = { version = "1", features = ["serde"] }
 snafu = "0.8"
 sysinfo = "0.30"
+similar-asserts = "1.6.0"
 # on branch v0.44.x
 sqlparser = { git = "https://github.com/GreptimeTeam/sqlparser-rs.git", rev = "54a267ac89c09b11c0c88934690530807185d3e7", features = [
     "visitor",
@@ -251,7 +254,7 @@ debug = 1
 
 [profile.nightly]
 inherits = "release"
-strip = true
+strip = "debuginfo"
 lto = "thin"
 debug = false
 incremental = false

Makefile

@@ -8,6 +8,7 @@ CARGO_BUILD_OPTS := --locked
 IMAGE_REGISTRY ?= docker.io
 IMAGE_NAMESPACE ?= greptime
 IMAGE_TAG ?= latest
+DEV_BUILDER_IMAGE_TAG ?= 2024-06-06-b4b105ad-20240827021230
 BUILDX_MULTI_PLATFORM_BUILD ?= false
 BUILDX_BUILDER_NAME ?= gtbuilder
 BASE_IMAGE ?= ubuntu
@@ -77,7 +78,7 @@ build: ## Build debug version greptime.
 build-by-dev-builder: ## Build greptime by dev-builder.
 	docker run --network=host \
 	-v ${PWD}:/greptimedb -v ${CARGO_REGISTRY_CACHE}:/root/.cargo/registry \
-	-w /greptimedb ${IMAGE_REGISTRY}/${IMAGE_NAMESPACE}/dev-builder-${BASE_IMAGE}:latest \
+	-w /greptimedb ${IMAGE_REGISTRY}/${IMAGE_NAMESPACE}/dev-builder-${BASE_IMAGE}:${DEV_BUILDER_IMAGE_TAG} \
 	make build \
 	CARGO_EXTENSION="${CARGO_EXTENSION}" \
 	CARGO_PROFILE=${CARGO_PROFILE} \
@@ -91,7 +92,7 @@ build-by-dev-builder: ## Build greptime by dev-builder.
 build-android-bin: ## Build greptime binary for android.
 	docker run --network=host \
 	-v ${PWD}:/greptimedb -v ${CARGO_REGISTRY_CACHE}:/root/.cargo/registry \
-	-w /greptimedb ${IMAGE_REGISTRY}/${IMAGE_NAMESPACE}/dev-builder-android:latest \
+	-w /greptimedb ${IMAGE_REGISTRY}/${IMAGE_NAMESPACE}/dev-builder-android:${DEV_BUILDER_IMAGE_TAG} \
 	make build \
 	CARGO_EXTENSION="ndk --platform 23 -t aarch64-linux-android" \
 	CARGO_PROFILE=release \
@@ -105,8 +106,8 @@ build-android-bin: ## Build greptime binary for android.
 strip-android-bin: build-android-bin ## Strip greptime binary for android.
 	docker run --network=host \
 	-v ${PWD}:/greptimedb \
-	-w /greptimedb ${IMAGE_REGISTRY}/${IMAGE_NAMESPACE}/dev-builder-android:latest \
-	bash -c '$${NDK_ROOT}/toolchains/llvm/prebuilt/linux-x86_64/bin/llvm-strip /greptimedb/target/aarch64-linux-android/release/greptime'
+	-w /greptimedb ${IMAGE_REGISTRY}/${IMAGE_NAMESPACE}/dev-builder-android:${DEV_BUILDER_IMAGE_TAG} \
+	bash -c '$${NDK_ROOT}/toolchains/llvm/prebuilt/linux-x86_64/bin/llvm-strip --strip-debug /greptimedb/target/aarch64-linux-android/release/greptime'
 
 .PHONY: clean
 clean: ## Clean the project.
@@ -145,7 +146,7 @@ dev-builder: multi-platform-buildx ## Build dev-builder image.
 	docker buildx build --builder ${BUILDX_BUILDER_NAME} \
 	--build-arg="RUST_TOOLCHAIN=${RUST_TOOLCHAIN}" \
 	-f docker/dev-builder/${BASE_IMAGE}/Dockerfile \
-	-t ${IMAGE_REGISTRY}/${IMAGE_NAMESPACE}/dev-builder-${BASE_IMAGE}:${IMAGE_TAG} ${BUILDX_MULTI_PLATFORM_BUILD_OPTS} .
+	-t ${IMAGE_REGISTRY}/${IMAGE_NAMESPACE}/dev-builder-${BASE_IMAGE}:${DEV_BUILDER_IMAGE_TAG} ${BUILDX_MULTI_PLATFORM_BUILD_OPTS} .
 
 .PHONY: multi-platform-buildx
 multi-platform-buildx: ## Create buildx multi-platform builder.
@@ -190,6 +191,7 @@ fix-clippy: ## Fix clippy violations.
 .PHONY: fmt-check
 fmt-check: ## Check code format.
 	cargo fmt --all -- --check
+	python3 scripts/check-snafu.py
 
 .PHONY: start-etcd
 start-etcd: ## Start single node etcd for testing purpose.
@@ -203,7 +205,7 @@ stop-etcd: ## Stop single node etcd for testing purpose.
 run-it-in-container: start-etcd ## Run integration tests in dev-builder.
 	docker run --network=host \
 	-v ${PWD}:/greptimedb -v ${CARGO_REGISTRY_CACHE}:/root/.cargo/registry -v /tmp:/tmp \
-	-w /greptimedb ${IMAGE_REGISTRY}/${IMAGE_NAMESPACE}/dev-builder-${BASE_IMAGE}:latest \
+	-w /greptimedb ${IMAGE_REGISTRY}/${IMAGE_NAMESPACE}/dev-builder-${BASE_IMAGE}:${DEV_BUILDER_IMAGE_TAG} \
 	make test sqlness-test BUILD_JOBS=${BUILD_JOBS}
 
 .PHONY: start-cluster
@@ -219,7 +221,7 @@ config-docs: ## Generate configuration documentation from toml files.
 	docker run --rm \
     -v ${PWD}:/greptimedb \
     -w /greptimedb/config \
-    toml2docs/toml2docs:v0.1.1 \
+    toml2docs/toml2docs:v0.1.3 \
     -p '##' \
     -t ./config-docs-template.md \
     -o ./config.md

README.md

@@ -74,7 +74,7 @@ Our core developers have been building time-series data platforms for years. Bas
 
 * **Compatible with InfluxDB, Prometheus and more protocols**
 
-  Widely adopted database protocols and APIs, including MySQL, PostgreSQL, and Prometheus Remote Storage, etc. [Read more](https://docs.greptime.com/user-guide/clients/overview).
+  Widely adopted database protocols and APIs, including MySQL, PostgreSQL, and Prometheus Remote Storage, etc. [Read more](https://docs.greptime.com/user-guide/protocols/overview).
 
 ## Try GreptimeDB
 
@@ -150,7 +150,7 @@ Our official Grafana dashboard is available at [grafana](grafana/README.md) dire
 
 ## Project Status
 
-The current version has not yet reached the standards for General Availability. 
+The current version has not yet reached the standards for General Availability.
 According to our Greptime 2024 Roadmap, we aim to achieve a production-level version with the release of v1.0 by the end of 2024. [Join Us](https://github.com/GreptimeTeam/greptimedb/issues/3412)
 
 We welcome you to test and use GreptimeDB. Some users have already adopted it in their production environments. If you're interested in trying it out, please use the latest stable release available.
@@ -172,6 +172,13 @@ In addition, you may:
 - Connect us with [Linkedin](https://www.linkedin.com/company/greptime/)
 - Follow us on [Twitter](https://twitter.com/greptime)
 
+## Commerial Support
+
+If you are running GreptimeDB OSS in your organization, we offer additional
+enterprise addons, installation service, training and consulting. [Contact
+us](https://greptime.com/contactus) and we will reach out to you with more
+detail of our commerial license.
+
 ## License
 
 GreptimeDB uses the [Apache License 2.0](https://apache.org/licenses/LICENSE-2.0.txt) to strike a balance between

config/config.md

@@ -14,7 +14,9 @@
 | --- | -----| ------- | ----------- |
 | `mode` | String | `standalone` | The running mode of the datanode. It can be `standalone` or `distributed`. |
 | `enable_telemetry` | Bool | `true` | Enable telemetry to collect anonymous usage data. |
-| `default_timezone` | String | `None` | The default timezone of the server. |
+| `default_timezone` | String | Unset | The default timezone of the server. |
+| `init_regions_in_background` | Bool | `false` | Initialize all regions in the background during the startup.<br/>By default, it provides services after all regions have been initialized. |
+| `init_regions_parallelism` | Integer | `16` | Parallelism of initializing regions. |
 | `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. |
@@ -27,8 +29,8 @@
 | `grpc.runtime_size` | Integer | `8` | The number of server worker threads. |
 | `grpc.tls` | -- | -- | gRPC server TLS options, see `mysql.tls` section. |
 | `grpc.tls.mode` | String | `disable` | TLS mode. |
-| `grpc.tls.cert_path` | String | `None` | Certificate file path. |
-| `grpc.tls.key_path` | String | `None` | Private key file path. |
+| `grpc.tls.cert_path` | String | Unset | Certificate file path. |
+| `grpc.tls.key_path` | String | Unset | Private key file path. |
 | `grpc.tls.watch` | Bool | `false` | Watch for Certificate and key file change and auto reload.<br/>For now, gRPC tls config does not support auto reload. |
 | `mysql` | -- | -- | MySQL server options. |
 | `mysql.enable` | Bool | `true` | Whether to enable. |
@@ -36,8 +38,8 @@
 | `mysql.runtime_size` | Integer | `2` | The number of server worker threads. |
 | `mysql.tls` | -- | -- | -- |
 | `mysql.tls.mode` | String | `disable` | TLS mode, refer to https://www.postgresql.org/docs/current/libpq-ssl.html<br/>- `disable` (default value)<br/>- `prefer`<br/>- `require`<br/>- `verify-ca`<br/>- `verify-full` |
-| `mysql.tls.cert_path` | String | `None` | Certificate file path. |
-| `mysql.tls.key_path` | String | `None` | Private key file path. |
+| `mysql.tls.cert_path` | String | Unset | Certificate file path. |
+| `mysql.tls.key_path` | String | Unset | Private key file path. |
 | `mysql.tls.watch` | Bool | `false` | Watch for Certificate and key file change and auto reload |
 | `postgres` | -- | -- | PostgresSQL server options. |
 | `postgres.enable` | Bool | `true` | Whether to enable |
@@ -45,8 +47,8 @@
 | `postgres.runtime_size` | Integer | `2` | The number of server worker threads. |
 | `postgres.tls` | -- | -- | PostgresSQL server TLS options, see `mysql.tls` section. |
 | `postgres.tls.mode` | String | `disable` | TLS mode. |
-| `postgres.tls.cert_path` | String | `None` | Certificate file path. |
-| `postgres.tls.key_path` | String | `None` | Private key file path. |
+| `postgres.tls.cert_path` | String | Unset | Certificate file path. |
+| `postgres.tls.key_path` | String | Unset | Private key file path. |
 | `postgres.tls.watch` | Bool | `false` | Watch for Certificate and key file change and auto reload |
 | `opentsdb` | -- | -- | OpenTSDB protocol options. |
 | `opentsdb.enable` | Bool | `true` | Whether to enable OpenTSDB put in HTTP API. |
@@ -57,7 +59,7 @@
 | `prom_store.with_metric_engine` | Bool | `true` | Whether to store the data from Prometheus remote write in metric engine. |
 | `wal` | -- | -- | The WAL options. |
 | `wal.provider` | String | `raft_engine` | The provider of the WAL.<br/>- `raft_engine`: the wal is stored in the local file system by raft-engine.<br/>- `kafka`: it's remote wal that data is stored in Kafka. |
-| `wal.dir` | String | `None` | The directory to store the WAL files.<br/>**It's only used when the provider is `raft_engine`**. |
+| `wal.dir` | String | Unset | The directory to store the WAL files.<br/>**It's only used when the provider is `raft_engine`**. |
 | `wal.file_size` | String | `256MB` | The size of the WAL segment file.<br/>**It's only used when the provider is `raft_engine`**. |
 | `wal.purge_threshold` | String | `4GB` | The threshold of the WAL size to trigger a flush.<br/>**It's only used when the provider is `raft_engine`**. |
 | `wal.purge_interval` | String | `10m` | The interval to trigger a flush.<br/>**It's only used when the provider is `raft_engine`**. |
@@ -66,10 +68,12 @@
 | `wal.enable_log_recycle` | Bool | `true` | Whether to reuse logically truncated log files.<br/>**It's only used when the provider is `raft_engine`**. |
 | `wal.prefill_log_files` | Bool | `false` | Whether to pre-create log files on start up.<br/>**It's only used when the provider is `raft_engine`**. |
 | `wal.sync_period` | String | `10s` | Duration for fsyncing log files.<br/>**It's only used when the provider is `raft_engine`**. |
+| `wal.recovery_parallelism` | Integer | `2` | Parallelism during WAL recovery. |
 | `wal.broker_endpoints` | Array | -- | The Kafka broker endpoints.<br/>**It's only used when the provider is `kafka`**. |
-| `wal.num_topics` | Integer | `64` | Number of topics to be created upon start.<br/>**It's only used when the provider is `kafka`**. |
+| `wal.auto_create_topics` | Bool | `true` | Automatically create topics for WAL.<br/>Set to `true` to automatically create topics for WAL.<br/>Otherwise, use topics named `topic_name_prefix_[0..num_topics)` |
+| `wal.num_topics` | Integer | `64` | Number of topics.<br/>**It's only used when the provider is `kafka`**. |
 | `wal.selector_type` | String | `round_robin` | Topic selector type.<br/>Available selector types:<br/>- `round_robin` (default)<br/>**It's only used when the provider is `kafka`**. |
-| `wal.topic_name_prefix` | String | `greptimedb_wal_topic` | A Kafka topic is constructed by concatenating `topic_name_prefix` and `topic_id`.<br/>**It's only used when the provider is `kafka`**. |
+| `wal.topic_name_prefix` | String | `greptimedb_wal_topic` | A Kafka topic is constructed by concatenating `topic_name_prefix` and `topic_id`.<br/>i.g., greptimedb_wal_topic_0, greptimedb_wal_topic_1.<br/>**It's only used when the provider is `kafka`**. |
 | `wal.replication_factor` | Integer | `1` | Expected number of replicas of each partition.<br/>**It's only used when the provider is `kafka`**. |
 | `wal.create_topic_timeout` | String | `30s` | Above which a topic creation operation will be cancelled.<br/>**It's only used when the provider is `kafka`**. |
 | `wal.max_batch_bytes` | String | `1MB` | The max size of a single producer batch.<br/>Warning: Kafka has a default limit of 1MB per message in a topic.<br/>**It's only used when the provider is `kafka`**. |
@@ -87,22 +91,22 @@
 | `storage` | -- | -- | The data storage options. |
 | `storage.data_home` | String | `/tmp/greptimedb/` | 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 | `None` | Cache configuration for object storage such as 'S3' etc.<br/>The local file cache directory. |
-| `storage.cache_capacity` | String | `None` | The local file cache capacity in bytes. |
-| `storage.bucket` | String | `None` | The S3 bucket name.<br/>**It's only used when the storage type is `S3`, `Oss` and `Gcs`**. |
-| `storage.root` | String | `None` | The S3 data will be stored in the specified prefix, for example, `s3://${bucket}/${root}`.<br/>**It's only used when the storage type is `S3`, `Oss` and `Azblob`**. |
-| `storage.access_key_id` | String | `None` | The access key id of the aws account.<br/>It's **highly recommended** to use AWS IAM roles instead of hardcoding the access key id and secret key.<br/>**It's only used when the storage type is `S3` and `Oss`**. |
-| `storage.secret_access_key` | String | `None` | The secret access key of the aws account.<br/>It's **highly recommended** to use AWS IAM roles instead of hardcoding the access key id and secret key.<br/>**It's only used when the storage type is `S3`**. |
-| `storage.access_key_secret` | String | `None` | The secret access key of the aliyun account.<br/>**It's only used when the storage type is `Oss`**. |
-| `storage.account_name` | String | `None` | The account key of the azure account.<br/>**It's only used when the storage type is `Azblob`**. |
-| `storage.account_key` | String | `None` | The account key of the azure account.<br/>**It's only used when the storage type is `Azblob`**. |
-| `storage.scope` | String | `None` | The scope of the google cloud storage.<br/>**It's only used when the storage type is `Gcs`**. |
-| `storage.credential_path` | String | `None` | The credential path of the google cloud storage.<br/>**It's only used when the storage type is `Gcs`**. |
-| `storage.credential` | String | `None` | The credential of the google cloud storage.<br/>**It's only used when the storage type is `Gcs`**. |
-| `storage.container` | String | `None` | The container of the azure account.<br/>**It's only used when the storage type is `Azblob`**. |
-| `storage.sas_token` | String | `None` | The sas token of the azure account.<br/>**It's only used when the storage type is `Azblob`**. |
-| `storage.endpoint` | String | `None` | The endpoint of the S3 service.<br/>**It's only used when the storage type is `S3`, `Oss`, `Gcs` and `Azblob`**. |
-| `storage.region` | String | `None` | The region of the S3 service.<br/>**It's only used when the storage type is `S3`, `Oss`, `Gcs` and `Azblob`**. |
+| `storage.cache_path` | String | Unset | Cache configuration for object storage such as 'S3' etc.<br/>The local file cache directory. |
+| `storage.cache_capacity` | String | Unset | The local file cache capacity in bytes. |
+| `storage.bucket` | String | Unset | The S3 bucket name.<br/>**It's only used when the storage type is `S3`, `Oss` and `Gcs`**. |
+| `storage.root` | String | Unset | The S3 data will be stored in the specified prefix, for example, `s3://${bucket}/${root}`.<br/>**It's only used when the storage type is `S3`, `Oss` and `Azblob`**. |
+| `storage.access_key_id` | String | Unset | The access key id of the aws account.<br/>It's **highly recommended** to use AWS IAM roles instead of hardcoding the access key id and secret key.<br/>**It's only used when the storage type is `S3` and `Oss`**. |
+| `storage.secret_access_key` | String | Unset | The secret access key of the aws account.<br/>It's **highly recommended** to use AWS IAM roles instead of hardcoding the access key id and secret key.<br/>**It's only used when the storage type is `S3`**. |
+| `storage.access_key_secret` | String | Unset | The secret access key of the aliyun account.<br/>**It's only used when the storage type is `Oss`**. |
+| `storage.account_name` | String | Unset | The account key of the azure account.<br/>**It's only used when the storage type is `Azblob`**. |
+| `storage.account_key` | String | Unset | The account key of the azure account.<br/>**It's only used when the storage type is `Azblob`**. |
+| `storage.scope` | String | Unset | The scope of the google cloud storage.<br/>**It's only used when the storage type is `Gcs`**. |
+| `storage.credential_path` | String | Unset | The credential path of the google cloud storage.<br/>**It's only used when the storage type is `Gcs`**. |
+| `storage.credential` | String | Unset | The credential of the google cloud storage.<br/>**It's only used when the storage type is `Gcs`**. |
+| `storage.container` | String | Unset | The container of the azure account.<br/>**It's only used when the storage type is `Azblob`**. |
+| `storage.sas_token` | String | Unset | The sas token of the azure account.<br/>**It's only used when the storage type is `Azblob`**. |
+| `storage.endpoint` | String | Unset | The endpoint of the S3 service.<br/>**It's only used when the storage type is `S3`, `Oss`, `Gcs` and `Azblob`**. |
+| `storage.region` | String | Unset | The region of the S3 service.<br/>**It's only used when the storage type is `S3`, `Oss`, `Gcs` and `Azblob`**. |
 | `[[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. |
@@ -112,19 +116,21 @@
 | `region_engine.mito.compress_manifest` | Bool | `false` | Whether to compress manifest and checkpoint file by gzip (default false). |
 | `region_engine.mito.max_background_jobs` | Integer | `4` | Max number of running background jobs |
 | `region_engine.mito.auto_flush_interval` | String | `1h` | Interval to auto flush a region if it has not flushed yet. |
-| `region_engine.mito.global_write_buffer_size` | String | `1GB` | Global write buffer size for all regions. If not set, it's default to 1/8 of OS memory with a max limitation of 1GB. |
-| `region_engine.mito.global_write_buffer_reject_size` | String | `2GB` | Global write buffer size threshold to reject write requests. If not set, it's default to 2 times of `global_write_buffer_size` |
-| `region_engine.mito.sst_meta_cache_size` | String | `128MB` | Cache size for SST metadata. Setting it to 0 to disable the cache.<br/>If not set, it's default to 1/32 of OS memory with a max limitation of 128MB. |
-| `region_engine.mito.vector_cache_size` | String | `512MB` | Cache size for vectors and arrow arrays. Setting it to 0 to disable the cache.<br/>If not set, it's default to 1/16 of OS memory with a max limitation of 512MB. |
-| `region_engine.mito.page_cache_size` | String | `512MB` | Cache size for pages of SST row groups. Setting it to 0 to disable the cache.<br/>If not set, it's default to 1/16 of OS memory with a max limitation of 512MB. |
+| `region_engine.mito.global_write_buffer_size` | String | Auto | Global write buffer size for all regions. If not set, it's default to 1/8 of OS memory with a max limitation of 1GB. |
+| `region_engine.mito.global_write_buffer_reject_size` | String | Auto | Global write buffer size threshold to reject write requests. If not set, it's default to 2 times of `global_write_buffer_size`. |
+| `region_engine.mito.sst_meta_cache_size` | String | Auto | Cache size for SST metadata. Setting it to 0 to disable the cache.<br/>If not set, it's default to 1/32 of OS memory with a max limitation of 128MB. |
+| `region_engine.mito.vector_cache_size` | String | Auto | Cache size for vectors and arrow arrays. Setting it to 0 to disable the cache.<br/>If not set, it's default to 1/16 of OS memory with a max limitation of 512MB. |
+| `region_engine.mito.page_cache_size` | String | Auto | Cache size for pages of SST row groups. Setting it to 0 to disable the cache.<br/>If not set, it's default to 1/8 of OS memory. |
+| `region_engine.mito.selector_result_cache_size` | String | Auto | Cache size for time series selector (e.g. `last_value()`). Setting it to 0 to disable the cache.<br/>If not set, it's default to 1/16 of OS memory with a max limitation of 512MB. |
 | `region_engine.mito.enable_experimental_write_cache` | Bool | `false` | Whether to enable the experimental write cache. |
 | `region_engine.mito.experimental_write_cache_path` | String | `""` | File system path for write cache, defaults to `{data_home}/write_cache`. |
 | `region_engine.mito.experimental_write_cache_size` | String | `512MB` | Capacity for write cache. |
-| `region_engine.mito.experimental_write_cache_ttl` | String | `1h` | TTL for write cache. |
+| `region_engine.mito.experimental_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.scan_parallelism` | Integer | `0` | Parallelism to scan a region (default: 1/4 of cpu cores).<br/>- `0`: using the default value (1/4 of cpu cores).<br/>- `1`: scan in current thread.<br/>- `n`: scan in parallelism n. |
 | `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.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. |
 | `region_engine.mito.index.aux_path` | String | `""` | Auxiliary directory path for the index in filesystem, used to store intermediate files for<br/>creating the index and staging files for searching the index, defaults to `{data_home}/index_intermediate`.<br/>The default name for this directory is `index_intermediate` for backward compatibility.<br/><br/>This path contains two subdirectories:<br/>- `__intm`: for storing intermediate files used during creating index.<br/>- `staging`: for storing staging files used during searching index. |
 | `region_engine.mito.index.staging_size` | String | `2GB` | The max capacity of the staging directory. |
@@ -148,23 +154,24 @@
 | `region_engine.mito.memtable.fork_dictionary_bytes` | String | `1GiB` | Max dictionary bytes.<br/>Only available for `partition_tree` memtable. |
 | `region_engine.file` | -- | -- | Enable the file engine. |
 | `logging` | -- | -- | The logging options. |
-| `logging.dir` | String | `/tmp/greptimedb/logs` | The directory to store the log files. |
-| `logging.level` | String | `None` | The log level. Can be `info`/`debug`/`warn`/`error`. |
+| `logging.dir` | String | `/tmp/greptimedb/logs` | The directory to store the log files. If set to empty, logs will not be written to files. |
+| `logging.level` | String | Unset | The log level. Can be `info`/`debug`/`warn`/`error`. |
 | `logging.enable_otlp_tracing` | Bool | `false` | Enable OTLP tracing. |
 | `logging.otlp_endpoint` | String | `http://localhost:4317` | The OTLP tracing endpoint. |
 | `logging.append_stdout` | Bool | `true` | Whether to append logs to stdout. |
+| `logging.log_format` | String | `text` | The log format. Can be `text`/`json`. |
 | `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` | -- |
 | `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.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 |
-| `export_metrics.self_import.db` | String | `None` | -- |
+| `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=information_schema`. |
+| `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.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 | `None` | The tokio console address. |
+| `tracing.tokio_console_addr` | String | Unset | The tokio console address. |
 
 
 ## Distributed Mode
@@ -173,7 +180,7 @@
 
 | Key | Type | Default | Descriptions |
 | --- | -----| ------- | ----------- |
-| `default_timezone` | String | `None` | The default timezone of the server. |
+| `default_timezone` | String | Unset | The default timezone of the server. |
 | `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. |
@@ -190,8 +197,8 @@
 | `grpc.runtime_size` | Integer | `8` | The number of server worker threads. |
 | `grpc.tls` | -- | -- | gRPC server TLS options, see `mysql.tls` section. |
 | `grpc.tls.mode` | String | `disable` | TLS mode. |
-| `grpc.tls.cert_path` | String | `None` | Certificate file path. |
-| `grpc.tls.key_path` | String | `None` | Private key file path. |
+| `grpc.tls.cert_path` | String | Unset | Certificate file path. |
+| `grpc.tls.key_path` | String | Unset | Private key file path. |
 | `grpc.tls.watch` | Bool | `false` | Watch for Certificate and key file change and auto reload.<br/>For now, gRPC tls config does not support auto reload. |
 | `mysql` | -- | -- | MySQL server options. |
 | `mysql.enable` | Bool | `true` | Whether to enable. |
@@ -199,8 +206,8 @@
 | `mysql.runtime_size` | Integer | `2` | The number of server worker threads. |
 | `mysql.tls` | -- | -- | -- |
 | `mysql.tls.mode` | String | `disable` | TLS mode, refer to https://www.postgresql.org/docs/current/libpq-ssl.html<br/>- `disable` (default value)<br/>- `prefer`<br/>- `require`<br/>- `verify-ca`<br/>- `verify-full` |
-| `mysql.tls.cert_path` | String | `None` | Certificate file path. |
-| `mysql.tls.key_path` | String | `None` | Private key file path. |
+| `mysql.tls.cert_path` | String | Unset | Certificate file path. |
+| `mysql.tls.key_path` | String | Unset | Private key file path. |
 | `mysql.tls.watch` | Bool | `false` | Watch for Certificate and key file change and auto reload |
 | `postgres` | -- | -- | PostgresSQL server options. |
 | `postgres.enable` | Bool | `true` | Whether to enable |
@@ -208,8 +215,8 @@
 | `postgres.runtime_size` | Integer | `2` | The number of server worker threads. |
 | `postgres.tls` | -- | -- | PostgresSQL server TLS options, see `mysql.tls` section. |
 | `postgres.tls.mode` | String | `disable` | TLS mode. |
-| `postgres.tls.cert_path` | String | `None` | Certificate file path. |
-| `postgres.tls.key_path` | String | `None` | Private key file path. |
+| `postgres.tls.cert_path` | String | Unset | Certificate file path. |
+| `postgres.tls.key_path` | String | Unset | Private key file path. |
 | `postgres.tls.watch` | Bool | `false` | Watch for Certificate and key file change and auto reload |
 | `opentsdb` | -- | -- | OpenTSDB protocol options. |
 | `opentsdb.enable` | Bool | `true` | Whether to enable OpenTSDB put in HTTP API. |
@@ -233,23 +240,24 @@
 | `datanode.client.connect_timeout` | String | `10s` | -- |
 | `datanode.client.tcp_nodelay` | Bool | `true` | -- |
 | `logging` | -- | -- | The logging options. |
-| `logging.dir` | String | `/tmp/greptimedb/logs` | The directory to store the log files. |
-| `logging.level` | String | `None` | The log level. Can be `info`/`debug`/`warn`/`error`. |
+| `logging.dir` | String | `/tmp/greptimedb/logs` | The directory to store the log files. If set to empty, logs will not be written to files. |
+| `logging.level` | String | Unset | The log level. Can be `info`/`debug`/`warn`/`error`. |
 | `logging.enable_otlp_tracing` | Bool | `false` | Enable OTLP tracing. |
 | `logging.otlp_endpoint` | String | `http://localhost:4317` | The OTLP tracing endpoint. |
 | `logging.append_stdout` | Bool | `true` | Whether to append logs to stdout. |
+| `logging.log_format` | String | `text` | The log format. Can be `text`/`json`. |
 | `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` | -- |
 | `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.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 |
-| `export_metrics.self_import.db` | String | `None` | -- |
+| `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=information_schema`. |
+| `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.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 | `None` | The tokio console address. |
+| `tracing.tokio_console_addr` | String | Unset | The tokio console address. |
 
 
 ### Metasrv
@@ -259,12 +267,13 @@
 | `data_home` | String | `/tmp/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 frontend and datanode to connect to metasrv,  "127.0.0.1:3002" by default for localhost. |
-| `store_addr` | String | `127.0.0.1:2379` | Etcd server address. |
+| `store_addr` | String | `127.0.0.1:2379` | Store server address default to etcd 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_telemetry` | Bool | `true` | Whether to enable greptimedb telemetry. |
 | `store_key_prefix` | String | `""` | If it's not empty, the metasrv will store all data with this key prefix. |
 | `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). |
+| `backend` | String | `EtcdStore` | The datastore for meta server. |
 | `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. |
@@ -285,9 +294,10 @@
 | `wal` | -- | -- | -- |
 | `wal.provider` | String | `raft_engine` | -- |
 | `wal.broker_endpoints` | Array | -- | The broker endpoints of the Kafka cluster. |
-| `wal.num_topics` | Integer | `64` | Number of topics to be created upon start. |
+| `wal.auto_create_topics` | Bool | `true` | Automatically create topics for WAL.<br/>Set to `true` to automatically create topics for WAL.<br/>Otherwise, use topics named `topic_name_prefix_[0..num_topics)` |
+| `wal.num_topics` | Integer | `64` | Number of topics. |
 | `wal.selector_type` | String | `round_robin` | Topic selector type.<br/>Available selector types:<br/>- `round_robin` (default) |
-| `wal.topic_name_prefix` | String | `greptimedb_wal_topic` | A Kafka topic is constructed by concatenating `topic_name_prefix` and `topic_id`. |
+| `wal.topic_name_prefix` | String | `greptimedb_wal_topic` | A Kafka topic is constructed by concatenating `topic_name_prefix` and `topic_id`.<br/>i.g., greptimedb_wal_topic_0, greptimedb_wal_topic_1. |
 | `wal.replication_factor` | Integer | `1` | Expected number of replicas of each partition. |
 | `wal.create_topic_timeout` | String | `30s` | Above which a topic creation operation will be cancelled. |
 | `wal.backoff_init` | String | `500ms` | The initial backoff for kafka clients. |
@@ -295,23 +305,24 @@
 | `wal.backoff_base` | Integer | `2` | Exponential backoff rate, i.e. next backoff = base * current backoff. |
 | `wal.backoff_deadline` | String | `5mins` | Stop reconnecting if the total wait time reaches the deadline. If this config is missing, the reconnecting won't terminate. |
 | `logging` | -- | -- | The logging options. |
-| `logging.dir` | String | `/tmp/greptimedb/logs` | The directory to store the log files. |
-| `logging.level` | String | `None` | The log level. Can be `info`/`debug`/`warn`/`error`. |
+| `logging.dir` | String | `/tmp/greptimedb/logs` | The directory to store the log files. If set to empty, logs will not be written to files. |
+| `logging.level` | String | Unset | The log level. Can be `info`/`debug`/`warn`/`error`. |
 | `logging.enable_otlp_tracing` | Bool | `false` | Enable OTLP tracing. |
 | `logging.otlp_endpoint` | String | `http://localhost:4317` | The OTLP tracing endpoint. |
 | `logging.append_stdout` | Bool | `true` | Whether to append logs to stdout. |
+| `logging.log_format` | String | `text` | The log format. Can be `text`/`json`. |
 | `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` | -- |
 | `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.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 |
-| `export_metrics.self_import.db` | String | `None` | -- |
+| `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=information_schema`. |
+| `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.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 | `None` | The tokio console address. |
+| `tracing.tokio_console_addr` | String | Unset | The tokio console address. |
 
 
 ### Datanode
@@ -319,16 +330,20 @@
 | Key | Type | Default | Descriptions |
 | --- | -----| ------- | ----------- |
 | `mode` | String | `standalone` | The running mode of the datanode. It can be `standalone` or `distributed`. |
-| `node_id` | Integer | `None` | The datanode identifier and should be unique in the cluster. |
+| `node_id` | Integer | Unset | The datanode identifier and should be unique in the cluster. |
 | `require_lease_before_startup` | Bool | `false` | Start services after regions have obtained leases.<br/>It will block the datanode start if it can't receive leases in the heartbeat from metasrv. |
 | `init_regions_in_background` | Bool | `false` | Initialize all regions in the background during the startup.<br/>By default, it provides services after all regions have been initialized. |
 | `enable_telemetry` | Bool | `true` | Enable telemetry to collect anonymous usage data. |
 | `init_regions_parallelism` | Integer | `16` | Parallelism of initializing regions. |
-| `rpc_addr` | String | `None` | Deprecated, use `grpc.addr` instead. |
-| `rpc_hostname` | String | `None` | Deprecated, use `grpc.hostname` instead. |
-| `rpc_runtime_size` | Integer | `None` | Deprecated, use `grpc.runtime_size` instead. |
-| `rpc_max_recv_message_size` | String | `None` | Deprecated, use `grpc.rpc_max_recv_message_size` instead. |
-| `rpc_max_send_message_size` | String | `None` | Deprecated, use `grpc.rpc_max_send_message_size` instead. |
+| `rpc_addr` | String | Unset | Deprecated, use `grpc.addr` instead. |
+| `rpc_hostname` | String | Unset | Deprecated, use `grpc.hostname` instead. |
+| `rpc_runtime_size` | Integer | Unset | Deprecated, use `grpc.runtime_size` instead. |
+| `rpc_max_recv_message_size` | String | Unset | Deprecated, use `grpc.rpc_max_recv_message_size` instead. |
+| `rpc_max_send_message_size` | String | Unset | Deprecated, use `grpc.rpc_max_send_message_size` instead. |
+| `http` | -- | -- | The HTTP server options. |
+| `http.addr` | String | `127.0.0.1:4000` | The address to bind the HTTP server. |
+| `http.timeout` | String | `30s` | 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. |
 | `grpc` | -- | -- | The gRPC server options. |
 | `grpc.addr` | String | `127.0.0.1:3001` | The address to bind the gRPC server. |
 | `grpc.hostname` | String | `127.0.0.1` | The hostname advertised to the metasrv,<br/>and used for connections from outside the host |
@@ -337,8 +352,8 @@
 | `grpc.max_send_message_size` | String | `512MB` | The maximum send message size for gRPC server. |
 | `grpc.tls` | -- | -- | gRPC server TLS options, see `mysql.tls` section. |
 | `grpc.tls.mode` | String | `disable` | TLS mode. |
-| `grpc.tls.cert_path` | String | `None` | Certificate file path. |
-| `grpc.tls.key_path` | String | `None` | Private key file path. |
+| `grpc.tls.cert_path` | String | Unset | Certificate file path. |
+| `grpc.tls.key_path` | String | Unset | Private key file path. |
 | `grpc.tls.watch` | Bool | `false` | Watch for Certificate and key file change and auto reload.<br/>For now, gRPC tls config does not support auto reload. |
 | `runtime` | -- | -- | The runtime options. |
 | `runtime.global_rt_size` | Integer | `8` | The number of threads to execute the runtime for global read operations. |
@@ -358,7 +373,7 @@
 | `meta_client.metadata_cache_tti` | String | `5m` | -- |
 | `wal` | -- | -- | The WAL options. |
 | `wal.provider` | String | `raft_engine` | The provider of the WAL.<br/>- `raft_engine`: the wal is stored in the local file system by raft-engine.<br/>- `kafka`: it's remote wal that data is stored in Kafka. |
-| `wal.dir` | String | `None` | The directory to store the WAL files.<br/>**It's only used when the provider is `raft_engine`**. |
+| `wal.dir` | String | Unset | The directory to store the WAL files.<br/>**It's only used when the provider is `raft_engine`**. |
 | `wal.file_size` | String | `256MB` | The size of the WAL segment file.<br/>**It's only used when the provider is `raft_engine`**. |
 | `wal.purge_threshold` | String | `4GB` | The threshold of the WAL size to trigger a flush.<br/>**It's only used when the provider is `raft_engine`**. |
 | `wal.purge_interval` | String | `10m` | The interval to trigger a flush.<br/>**It's only used when the provider is `raft_engine`**. |
@@ -367,6 +382,7 @@
 | `wal.enable_log_recycle` | Bool | `true` | Whether to reuse logically truncated log files.<br/>**It's only used when the provider is `raft_engine`**. |
 | `wal.prefill_log_files` | Bool | `false` | Whether to pre-create log files on start up.<br/>**It's only used when the provider is `raft_engine`**. |
 | `wal.sync_period` | String | `10s` | Duration for fsyncing log files.<br/>**It's only used when the provider is `raft_engine`**. |
+| `wal.recovery_parallelism` | Integer | `2` | Parallelism during WAL recovery. |
 | `wal.broker_endpoints` | Array | -- | The Kafka broker endpoints.<br/>**It's only used when the provider is `kafka`**. |
 | `wal.max_batch_bytes` | String | `1MB` | The max size of a single producer batch.<br/>Warning: Kafka has a default limit of 1MB per message in a topic.<br/>**It's only used when the provider is `kafka`**. |
 | `wal.consumer_wait_timeout` | String | `100ms` | The consumer wait timeout.<br/>**It's only used when the provider is `kafka`**. |
@@ -379,22 +395,22 @@
 | `storage` | -- | -- | The data storage options. |
 | `storage.data_home` | String | `/tmp/greptimedb/` | 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 | `None` | Cache configuration for object storage such as 'S3' etc.<br/>The local file cache directory. |
-| `storage.cache_capacity` | String | `None` | The local file cache capacity in bytes. |
-| `storage.bucket` | String | `None` | The S3 bucket name.<br/>**It's only used when the storage type is `S3`, `Oss` and `Gcs`**. |
-| `storage.root` | String | `None` | The S3 data will be stored in the specified prefix, for example, `s3://${bucket}/${root}`.<br/>**It's only used when the storage type is `S3`, `Oss` and `Azblob`**. |
-| `storage.access_key_id` | String | `None` | The access key id of the aws account.<br/>It's **highly recommended** to use AWS IAM roles instead of hardcoding the access key id and secret key.<br/>**It's only used when the storage type is `S3` and `Oss`**. |
-| `storage.secret_access_key` | String | `None` | The secret access key of the aws account.<br/>It's **highly recommended** to use AWS IAM roles instead of hardcoding the access key id and secret key.<br/>**It's only used when the storage type is `S3`**. |
-| `storage.access_key_secret` | String | `None` | The secret access key of the aliyun account.<br/>**It's only used when the storage type is `Oss`**. |
-| `storage.account_name` | String | `None` | The account key of the azure account.<br/>**It's only used when the storage type is `Azblob`**. |
-| `storage.account_key` | String | `None` | The account key of the azure account.<br/>**It's only used when the storage type is `Azblob`**. |
-| `storage.scope` | String | `None` | The scope of the google cloud storage.<br/>**It's only used when the storage type is `Gcs`**. |
-| `storage.credential_path` | String | `None` | The credential path of the google cloud storage.<br/>**It's only used when the storage type is `Gcs`**. |
-| `storage.credential` | String | `None` | The credential of the google cloud storage.<br/>**It's only used when the storage type is `Gcs`**. |
-| `storage.container` | String | `None` | The container of the azure account.<br/>**It's only used when the storage type is `Azblob`**. |
-| `storage.sas_token` | String | `None` | The sas token of the azure account.<br/>**It's only used when the storage type is `Azblob`**. |
-| `storage.endpoint` | String | `None` | The endpoint of the S3 service.<br/>**It's only used when the storage type is `S3`, `Oss`, `Gcs` and `Azblob`**. |
-| `storage.region` | String | `None` | The region of the S3 service.<br/>**It's only used when the storage type is `S3`, `Oss`, `Gcs` and `Azblob`**. |
+| `storage.cache_path` | String | Unset | Cache configuration for object storage such as 'S3' etc.<br/>The local file cache directory. |
+| `storage.cache_capacity` | String | Unset | The local file cache capacity in bytes. |
+| `storage.bucket` | String | Unset | The S3 bucket name.<br/>**It's only used when the storage type is `S3`, `Oss` and `Gcs`**. |
+| `storage.root` | String | Unset | The S3 data will be stored in the specified prefix, for example, `s3://${bucket}/${root}`.<br/>**It's only used when the storage type is `S3`, `Oss` and `Azblob`**. |
+| `storage.access_key_id` | String | Unset | The access key id of the aws account.<br/>It's **highly recommended** to use AWS IAM roles instead of hardcoding the access key id and secret key.<br/>**It's only used when the storage type is `S3` and `Oss`**. |
+| `storage.secret_access_key` | String | Unset | The secret access key of the aws account.<br/>It's **highly recommended** to use AWS IAM roles instead of hardcoding the access key id and secret key.<br/>**It's only used when the storage type is `S3`**. |
+| `storage.access_key_secret` | String | Unset | The secret access key of the aliyun account.<br/>**It's only used when the storage type is `Oss`**. |
+| `storage.account_name` | String | Unset | The account key of the azure account.<br/>**It's only used when the storage type is `Azblob`**. |
+| `storage.account_key` | String | Unset | The account key of the azure account.<br/>**It's only used when the storage type is `Azblob`**. |
+| `storage.scope` | String | Unset | The scope of the google cloud storage.<br/>**It's only used when the storage type is `Gcs`**. |
+| `storage.credential_path` | String | Unset | The credential path of the google cloud storage.<br/>**It's only used when the storage type is `Gcs`**. |
+| `storage.credential` | String | Unset | The credential of the google cloud storage.<br/>**It's only used when the storage type is `Gcs`**. |
+| `storage.container` | String | Unset | The container of the azure account.<br/>**It's only used when the storage type is `Azblob`**. |
+| `storage.sas_token` | String | Unset | The sas token of the azure account.<br/>**It's only used when the storage type is `Azblob`**. |
+| `storage.endpoint` | String | Unset | The endpoint of the S3 service.<br/>**It's only used when the storage type is `S3`, `Oss`, `Gcs` and `Azblob`**. |
+| `storage.region` | String | Unset | The region of the S3 service.<br/>**It's only used when the storage type is `S3`, `Oss`, `Gcs` and `Azblob`**. |
 | `[[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. |
@@ -404,19 +420,21 @@
 | `region_engine.mito.compress_manifest` | Bool | `false` | Whether to compress manifest and checkpoint file by gzip (default false). |
 | `region_engine.mito.max_background_jobs` | Integer | `4` | Max number of running background jobs |
 | `region_engine.mito.auto_flush_interval` | String | `1h` | Interval to auto flush a region if it has not flushed yet. |
-| `region_engine.mito.global_write_buffer_size` | String | `1GB` | Global write buffer size for all regions. If not set, it's default to 1/8 of OS memory with a max limitation of 1GB. |
-| `region_engine.mito.global_write_buffer_reject_size` | String | `2GB` | Global write buffer size threshold to reject write requests. If not set, it's default to 2 times of `global_write_buffer_size` |
-| `region_engine.mito.sst_meta_cache_size` | String | `128MB` | Cache size for SST metadata. Setting it to 0 to disable the cache.<br/>If not set, it's default to 1/32 of OS memory with a max limitation of 128MB. |
-| `region_engine.mito.vector_cache_size` | String | `512MB` | Cache size for vectors and arrow arrays. Setting it to 0 to disable the cache.<br/>If not set, it's default to 1/16 of OS memory with a max limitation of 512MB. |
-| `region_engine.mito.page_cache_size` | String | `512MB` | Cache size for pages of SST row groups. Setting it to 0 to disable the cache.<br/>If not set, it's default to 1/16 of OS memory with a max limitation of 512MB. |
+| `region_engine.mito.global_write_buffer_size` | String | Auto | Global write buffer size for all regions. If not set, it's default to 1/8 of OS memory with a max limitation of 1GB. |
+| `region_engine.mito.global_write_buffer_reject_size` | String | Auto | Global write buffer size threshold to reject write requests. If not set, it's default to 2 times of `global_write_buffer_size` |
+| `region_engine.mito.sst_meta_cache_size` | String | Auto | Cache size for SST metadata. Setting it to 0 to disable the cache.<br/>If not set, it's default to 1/32 of OS memory with a max limitation of 128MB. |
+| `region_engine.mito.vector_cache_size` | String | Auto | Cache size for vectors and arrow arrays. Setting it to 0 to disable the cache.<br/>If not set, it's default to 1/16 of OS memory with a max limitation of 512MB. |
+| `region_engine.mito.page_cache_size` | String | Auto | Cache size for pages of SST row groups. Setting it to 0 to disable the cache.<br/>If not set, it's default to 1/8 of OS memory. |
+| `region_engine.mito.selector_result_cache_size` | String | Auto | Cache size for time series selector (e.g. `last_value()`). Setting it to 0 to disable the cache.<br/>If not set, it's default to 1/16 of OS memory with a max limitation of 512MB. |
 | `region_engine.mito.enable_experimental_write_cache` | Bool | `false` | Whether to enable the experimental write cache. |
 | `region_engine.mito.experimental_write_cache_path` | String | `""` | File system path for write cache, defaults to `{data_home}/write_cache`. |
 | `region_engine.mito.experimental_write_cache_size` | String | `512MB` | Capacity for write cache. |
-| `region_engine.mito.experimental_write_cache_ttl` | String | `1h` | TTL for write cache. |
+| `region_engine.mito.experimental_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.scan_parallelism` | Integer | `0` | Parallelism to scan a region (default: 1/4 of cpu cores).<br/>- `0`: using the default value (1/4 of cpu cores).<br/>- `1`: scan in current thread.<br/>- `n`: scan in parallelism n. |
 | `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.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. |
 | `region_engine.mito.index.aux_path` | String | `""` | Auxiliary directory path for the index in filesystem, used to store intermediate files for<br/>creating the index and staging files for searching the index, defaults to `{data_home}/index_intermediate`.<br/>The default name for this directory is `index_intermediate` for backward compatibility.<br/><br/>This path contains two subdirectories:<br/>- `__intm`: for storing intermediate files used during creating index.<br/>- `staging`: for storing staging files used during searching index. |
 | `region_engine.mito.index.staging_size` | String | `2GB` | The max capacity of the staging directory. |
@@ -438,23 +456,24 @@
 | `region_engine.mito.memtable.fork_dictionary_bytes` | String | `1GiB` | Max dictionary bytes.<br/>Only available for `partition_tree` memtable. |
 | `region_engine.file` | -- | -- | Enable the file engine. |
 | `logging` | -- | -- | The logging options. |
-| `logging.dir` | String | `/tmp/greptimedb/logs` | The directory to store the log files. |
-| `logging.level` | String | `None` | The log level. Can be `info`/`debug`/`warn`/`error`. |
+| `logging.dir` | String | `/tmp/greptimedb/logs` | The directory to store the log files. If set to empty, logs will not be written to files. |
+| `logging.level` | String | Unset | The log level. Can be `info`/`debug`/`warn`/`error`. |
 | `logging.enable_otlp_tracing` | Bool | `false` | Enable OTLP tracing. |
 | `logging.otlp_endpoint` | String | `http://localhost:4317` | The OTLP tracing endpoint. |
 | `logging.append_stdout` | Bool | `true` | Whether to append logs to stdout. |
+| `logging.log_format` | String | `text` | The log format. Can be `text`/`json`. |
 | `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` | -- |
 | `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.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 |
-| `export_metrics.self_import.db` | String | `None` | -- |
+| `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=information_schema`. |
+| `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.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 | `None` | The tokio console address. |
+| `tracing.tokio_console_addr` | String | Unset | The tokio console address. |
 
 
 ### Flownode
@@ -462,7 +481,7 @@
 | Key | Type | Default | Descriptions |
 | --- | -----| ------- | ----------- |
 | `mode` | String | `distributed` | The running mode of the flownode. It can be `standalone` or `distributed`. |
-| `node_id` | Integer | `None` | The flownode identifier and should be unique in the cluster. |
+| `node_id` | Integer | Unset | The flownode identifier and should be unique in the cluster. |
 | `grpc` | -- | -- | The gRPC server options. |
 | `grpc.addr` | String | `127.0.0.1:6800` | The address to bind the gRPC server. |
 | `grpc.hostname` | String | `127.0.0.1` | The hostname advertised to the metasrv,<br/>and used for connections from outside the host |
@@ -483,12 +502,13 @@
 | `heartbeat.interval` | String | `3s` | Interval for sending heartbeat messages to the metasrv. |
 | `heartbeat.retry_interval` | String | `3s` | Interval for retrying to send heartbeat messages to the metasrv. |
 | `logging` | -- | -- | The logging options. |
-| `logging.dir` | String | `/tmp/greptimedb/logs` | The directory to store the log files. |
-| `logging.level` | String | `None` | The log level. Can be `info`/`debug`/`warn`/`error`. |
+| `logging.dir` | String | `/tmp/greptimedb/logs` | The directory to store the log files. If set to empty, logs will not be written to files. |
+| `logging.level` | String | Unset | The log level. Can be `info`/`debug`/`warn`/`error`. |
 | `logging.enable_otlp_tracing` | Bool | `false` | Enable OTLP tracing. |
 | `logging.otlp_endpoint` | String | `http://localhost:4317` | The OTLP tracing endpoint. |
 | `logging.append_stdout` | Bool | `true` | Whether to append logs to stdout. |
+| `logging.log_format` | String | `text` | The log format. Can be `text`/`json`. |
 | `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` | -- |
 | `tracing` | -- | -- | The tracing options. Only effect when compiled with `tokio-console` feature. |
-| `tracing.tokio_console_addr` | String | `None` | The tokio console address. |
+| `tracing.tokio_console_addr` | String | Unset | The tokio console address. |

config/datanode.example.toml

@@ -2,7 +2,7 @@
 mode = "standalone"
 
 ## The datanode identifier and should be unique in the cluster.
-## +toml2docs:none-default
+## @toml2docs:none-default
 node_id = 42
 
 ## Start services after regions have obtained leases.
@@ -20,25 +20,37 @@ enable_telemetry = true
 init_regions_parallelism = 16
 
 ## Deprecated, use `grpc.addr` instead.
-## +toml2docs:none-default
+## @toml2docs:none-default
 rpc_addr = "127.0.0.1:3001"
 
 ## Deprecated, use `grpc.hostname` instead.
-## +toml2docs:none-default
+## @toml2docs:none-default
 rpc_hostname = "127.0.0.1"
 
 ## Deprecated, use `grpc.runtime_size` instead.
-## +toml2docs:none-default
+## @toml2docs:none-default
 rpc_runtime_size = 8
 
 ## Deprecated, use `grpc.rpc_max_recv_message_size` instead.
-## +toml2docs:none-default
+## @toml2docs:none-default
 rpc_max_recv_message_size = "512MB"
 
 ## Deprecated, use `grpc.rpc_max_send_message_size` instead.
-## +toml2docs:none-default
+## @toml2docs:none-default
 rpc_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 = "30s"
+## 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"
+
 ## The gRPC server options.
 [grpc]
 ## The address to bind the gRPC server.
@@ -59,11 +71,11 @@ max_send_message_size = "512MB"
 mode = "disable"
 
 ## Certificate file path.
-## +toml2docs:none-default
+## @toml2docs:none-default
 cert_path = ""
 
 ## Private key file path.
-## +toml2docs:none-default
+## @toml2docs:none-default
 key_path = ""
 
 ## Watch for Certificate and key file change and auto reload.
@@ -71,11 +83,11 @@ key_path = ""
 watch = false
 
 ## The runtime options.
-[runtime]
+#+ [runtime]
 ## The number of threads to execute the runtime for global read operations.
-global_rt_size = 8
+#+ global_rt_size = 8
 ## The number of threads to execute the runtime for global write operations.
-compact_rt_size = 4
+#+ compact_rt_size = 4
 
 ## The heartbeat options.
 [heartbeat]
@@ -123,7 +135,7 @@ provider = "raft_engine"
 
 ## The directory to store the WAL files.
 ## **It's only used when the provider is `raft_engine`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 dir = "/tmp/greptimedb/wal"
 
 ## The size of the WAL segment file.
@@ -158,6 +170,9 @@ prefill_log_files = false
 ## **It's only used when the provider is `raft_engine`**.
 sync_period = "10s"
 
+## Parallelism during WAL recovery.
+recovery_parallelism = 2
+
 ## The Kafka broker endpoints.
 ## **It's only used when the provider is `kafka`**.
 broker_endpoints = ["127.0.0.1:9092"]
@@ -267,90 +282,104 @@ type = "File"
 
 ## Cache configuration for object storage such as 'S3' etc.
 ## The local file cache directory.
-## +toml2docs:none-default
+## @toml2docs:none-default
 cache_path = "/path/local_cache"
 
 ## The local file cache capacity in bytes.
-## +toml2docs:none-default
+## @toml2docs:none-default
 cache_capacity = "256MB"
 
 ## The S3 bucket name.
 ## **It's only used when the storage type is `S3`, `Oss` and `Gcs`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 bucket = "greptimedb"
 
 ## The S3 data will be stored in the specified prefix, for example, `s3://${bucket}/${root}`.
 ## **It's only used when the storage type is `S3`, `Oss` and `Azblob`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 root = "greptimedb"
 
 ## The access key id of the aws account.
 ## It's **highly recommended** to use AWS IAM roles instead of hardcoding the access key id and secret key.
 ## **It's only used when the storage type is `S3` and `Oss`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 access_key_id = "test"
 
 ## The secret access key of the aws account.
 ## It's **highly recommended** to use AWS IAM roles instead of hardcoding the access key id and secret key.
 ## **It's only used when the storage type is `S3`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 secret_access_key = "test"
 
 ## The secret access key of the aliyun account.
 ## **It's only used when the storage type is `Oss`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 access_key_secret = "test"
 
 ## The account key of the azure account.
 ## **It's only used when the storage type is `Azblob`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 account_name = "test"
 
 ## The account key of the azure account.
 ## **It's only used when the storage type is `Azblob`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 account_key = "test"
 
 ## The scope of the google cloud storage.
 ## **It's only used when the storage type is `Gcs`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 scope = "test"
 
 ## The credential path of the google cloud storage.
 ## **It's only used when the storage type is `Gcs`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 credential_path = "test"
 
 ## The credential of the google cloud storage.
 ## **It's only used when the storage type is `Gcs`**.
-## +toml2docs:none-default
-credential= "base64-credential"
+## @toml2docs:none-default
+credential = "base64-credential"
 
 ## The container of the azure account.
 ## **It's only used when the storage type is `Azblob`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 container = "greptimedb"
 
 ## The sas token of the azure account.
 ## **It's only used when the storage type is `Azblob`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 sas_token = ""
 
 ## The endpoint of the S3 service.
 ## **It's only used when the storage type is `S3`, `Oss`, `Gcs` and `Azblob`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 endpoint = "https://s3.amazonaws.com"
 
 ## The region of the S3 service.
 ## **It's only used when the storage type is `S3`, `Oss`, `Gcs` and `Azblob`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 region = "us-west-2"
 
 # Custom storage options
 # [[storage.providers]]
+# name = "S3"
 # type = "S3"
+# bucket = "greptimedb"
+# root = "data"
+# access_key_id = "test"
+# secret_access_key = "123456"
+# endpoint = "https://s3.amazonaws.com"
+# region = "us-west-2"
 # [[storage.providers]]
+# name = "Gcs"
 # type = "Gcs"
+# bucket = "greptimedb"
+# root = "data"
+# scope = "test"
+# credential_path = "123456"
+# credential = "base64-credential"
+# endpoint = "https://storage.googleapis.com"
 
 ## The region engine options. You can configure multiple region engines.
 [[region_engine]]
@@ -359,7 +388,7 @@ region = "us-west-2"
 [region_engine.mito]
 
 ## Number of region workers.
-num_workers = 8
+#+ num_workers = 8
 
 ## Request channel size of each worker.
 worker_channel_size = 128
@@ -380,22 +409,32 @@ max_background_jobs = 4
 auto_flush_interval = "1h"
 
 ## Global write buffer size for all regions. If not set, it's default to 1/8 of OS memory with a max limitation of 1GB.
-global_write_buffer_size = "1GB"
+## @toml2docs:none-default="Auto"
+#+ global_write_buffer_size = "1GB"
 
 ## Global write buffer size threshold to reject write requests. If not set, it's default to 2 times of `global_write_buffer_size`
-global_write_buffer_reject_size = "2GB"
+## @toml2docs:none-default="Auto"
+#+ global_write_buffer_reject_size = "2GB"
 
 ## Cache size for SST metadata. Setting it to 0 to disable the cache.
 ## If not set, it's default to 1/32 of OS memory with a max limitation of 128MB.
-sst_meta_cache_size = "128MB"
+## @toml2docs:none-default="Auto"
+#+ sst_meta_cache_size = "128MB"
 
 ## Cache size for vectors and arrow arrays. Setting it to 0 to disable the cache.
 ## If not set, it's default to 1/16 of OS memory with a max limitation of 512MB.
-vector_cache_size = "512MB"
+## @toml2docs:none-default="Auto"
+#+ vector_cache_size = "512MB"
 
 ## Cache size for pages of SST row groups. Setting it to 0 to disable the cache.
+## If not set, it's default to 1/8 of OS memory.
+## @toml2docs:none-default="Auto"
+#+ page_cache_size = "512MB"
+
+## Cache size for time series selector (e.g. `last_value()`). Setting it to 0 to disable the cache.
 ## If not set, it's default to 1/16 of OS memory with a max limitation of 512MB.
-page_cache_size = "512MB"
+## @toml2docs:none-default="Auto"
+#+ selector_result_cache_size = "512MB"
 
 ## Whether to enable the experimental write cache.
 enable_experimental_write_cache = false
@@ -407,7 +446,8 @@ experimental_write_cache_path = ""
 experimental_write_cache_size = "512MB"
 
 ## TTL for write cache.
-experimental_write_cache_ttl = "1h"
+## @toml2docs:none-default
+experimental_write_cache_ttl = "8h"
 
 ## Buffer size for SST writing.
 sst_write_buffer_size = "8MB"
@@ -424,6 +464,10 @@ parallel_scan_channel_size = 32
 ## Whether to allow stale WAL entries read during replay.
 allow_stale_entries = false
 
+## Minimum time interval between two compactions.
+## To align with the old behavior, the default value is 0 (no restrictions).
+min_compaction_interval = "0m"
+
 ## The options for index in Mito engine.
 [region_engine.mito.index]
 
@@ -514,11 +558,11 @@ fork_dictionary_bytes = "1GiB"
 
 ## The logging options.
 [logging]
-## The directory to store the log files.
+## The directory to store the log files. If set to empty, logs will not be written to files.
 dir = "/tmp/greptimedb/logs"
 
 ## The log level. Can be `info`/`debug`/`warn`/`error`.
-## +toml2docs:none-default
+## @toml2docs:none-default
 level = "info"
 
 ## Enable OTLP tracing.
@@ -530,6 +574,9 @@ otlp_endpoint = "http://localhost:4317"
 ## Whether to append logs to stdout.
 append_stdout = true
 
+## The log format. Can be `text`/`json`.
+log_format = "text"
+
 ## The percentage of tracing will be sampled and exported.
 ## Valid range `[0, 1]`, 1 means all traces are sampled, 0 means all traces are not sampled, the default value is 1.
 ## ratio > 1 are treated as 1. Fractions < 0 are treated as 0
@@ -547,12 +594,13 @@ enable = false
 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 = "information_schema"
+## @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=information_schema`.
+## The url 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.
@@ -561,5 +609,5 @@ headers = { }
 ## The tracing options. Only effect when compiled with `tokio-console` feature.
 [tracing]
 ## The tokio console address.
-## +toml2docs:none-default
+## @toml2docs:none-default
 tokio_console_addr = "127.0.0.1"

config/flownode.example.toml

@@ -2,7 +2,7 @@
 mode = "distributed"
 
 ## The flownode identifier and should be unique in the cluster.
-## +toml2docs:none-default
+## @toml2docs:none-default
 node_id = 14
 
 ## The gRPC server options.
@@ -59,11 +59,11 @@ retry_interval = "3s"
 
 ## The logging options.
 [logging]
-## The directory to store the log files.
+## The directory to store the log files. If set to empty, logs will not be written to files.
 dir = "/tmp/greptimedb/logs"
 
 ## The log level. Can be `info`/`debug`/`warn`/`error`.
-## +toml2docs:none-default
+## @toml2docs:none-default
 level = "info"
 
 ## Enable OTLP tracing.
@@ -75,6 +75,9 @@ otlp_endpoint = "http://localhost:4317"
 ## Whether to append logs to stdout.
 append_stdout = true
 
+## The log format. Can be `text`/`json`.
+log_format = "text"
+
 ## The percentage of tracing will be sampled and exported.
 ## Valid range `[0, 1]`, 1 means all traces are sampled, 0 means all traces are not sampled, the default value is 1.
 ## ratio > 1 are treated as 1. Fractions < 0 are treated as 0
@@ -84,6 +87,6 @@ default_ratio = 1.0
 ## The tracing options. Only effect when compiled with `tokio-console` feature.
 [tracing]
 ## The tokio console address.
-## +toml2docs:none-default
+## @toml2docs:none-default
 tokio_console_addr = "127.0.0.1"
 

config/frontend.example.toml

@@ -1,13 +1,13 @@
 ## The default timezone of the server.
-## +toml2docs:none-default
+## @toml2docs:none-default
 default_timezone = "UTC"
 
 ## The runtime options.
-[runtime]
+#+ [runtime]
 ## The number of threads to execute the runtime for global read operations.
-global_rt_size = 8
+#+ global_rt_size = 8
 ## The number of threads to execute the runtime for global write operations.
-compact_rt_size = 4
+#+ compact_rt_size = 4
 
 ## The heartbeat options.
 [heartbeat]
@@ -44,11 +44,11 @@ runtime_size = 8
 mode = "disable"
 
 ## Certificate file path.
-## +toml2docs:none-default
+## @toml2docs:none-default
 cert_path = ""
 
 ## Private key file path.
-## +toml2docs:none-default
+## @toml2docs:none-default
 key_path = ""
 
 ## Watch for Certificate and key file change and auto reload.
@@ -76,11 +76,11 @@ runtime_size = 2
 mode = "disable"
 
 ## Certificate file path.
-## +toml2docs:none-default
+## @toml2docs:none-default
 cert_path = ""
 
 ## Private key file path.
-## +toml2docs:none-default
+## @toml2docs:none-default
 key_path = ""
 
 ## Watch for Certificate and key file change and auto reload
@@ -101,11 +101,11 @@ runtime_size = 2
 mode = "disable"
 
 ## Certificate file path.
-## +toml2docs:none-default
+## @toml2docs:none-default
 cert_path = ""
 
 ## Private key file path.
-## +toml2docs:none-default
+## @toml2docs:none-default
 key_path = ""
 
 ## Watch for Certificate and key file change and auto reload
@@ -166,11 +166,11 @@ tcp_nodelay = true
 
 ## The logging options.
 [logging]
-## The directory to store the log files.
+## The directory to store the log files. If set to empty, logs will not be written to files.
 dir = "/tmp/greptimedb/logs"
 
 ## The log level. Can be `info`/`debug`/`warn`/`error`.
-## +toml2docs:none-default
+## @toml2docs:none-default
 level = "info"
 
 ## Enable OTLP tracing.
@@ -182,6 +182,9 @@ otlp_endpoint = "http://localhost:4317"
 ## Whether to append logs to stdout.
 append_stdout = true
 
+## The log format. Can be `text`/`json`.
+log_format = "text"
+
 ## The percentage of tracing will be sampled and exported.
 ## Valid range `[0, 1]`, 1 means all traces are sampled, 0 means all traces are not sampled, the default value is 1.
 ## ratio > 1 are treated as 1. Fractions < 0 are treated as 0
@@ -199,12 +202,13 @@ enable = false
 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 = "information_schema"
+## @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=information_schema`.
+## The url 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.
@@ -213,5 +217,5 @@ headers = { }
 ## The tracing options. Only effect when compiled with `tokio-console` feature.
 [tracing]
 ## The tokio console address.
-## +toml2docs:none-default
+## @toml2docs:none-default
 tokio_console_addr = "127.0.0.1"

config/metasrv.example.toml

@@ -7,7 +7,7 @@ bind_addr = "127.0.0.1:3002"
 ## The communication server address for frontend and datanode to connect to metasrv,  "127.0.0.1:3002" by default for localhost.
 server_addr = "127.0.0.1:3002"
 
-## Etcd server address.
+## Store server address default to etcd store.
 store_addr = "127.0.0.1:2379"
 
 ## Datanode selector type.
@@ -32,12 +32,15 @@ store_key_prefix = ""
 ## - Using shared storage (e.g., s3).
 enable_region_failover = false
 
+## The datastore for meta server.
+backend = "EtcdStore"
+
 ## The runtime options.
-[runtime]
+#+ [runtime]
 ## The number of threads to execute the runtime for global read operations.
-global_rt_size = 8
+#+ global_rt_size = 8
 ## The number of threads to execute the runtime for global write operations.
-compact_rt_size = 4
+#+ compact_rt_size = 4
 
 ## Procedure storage options.
 [procedure]
@@ -96,7 +99,12 @@ provider = "raft_engine"
 ## The broker endpoints of the Kafka cluster.
 broker_endpoints = ["127.0.0.1:9092"]
 
-## Number of topics to be created upon start.
+## Automatically create topics for WAL.
+## Set to `true` to automatically create topics for WAL.
+## Otherwise, use topics named `topic_name_prefix_[0..num_topics)`
+auto_create_topics = true
+
+## Number of topics.
 num_topics = 64
 
 ## Topic selector type.
@@ -105,6 +113,7 @@ num_topics = 64
 selector_type = "round_robin"
 
 ## A Kafka topic is constructed by concatenating `topic_name_prefix` and `topic_id`.
+## i.g., greptimedb_wal_topic_0, greptimedb_wal_topic_1.
 topic_name_prefix = "greptimedb_wal_topic"
 
 ## Expected number of replicas of each partition.
@@ -144,11 +153,11 @@ backoff_deadline = "5mins"
 
 ## The logging options.
 [logging]
-## The directory to store the log files.
+## The directory to store the log files. If set to empty, logs will not be written to files.
 dir = "/tmp/greptimedb/logs"
 
 ## The log level. Can be `info`/`debug`/`warn`/`error`.
-## +toml2docs:none-default
+## @toml2docs:none-default
 level = "info"
 
 ## Enable OTLP tracing.
@@ -160,6 +169,9 @@ otlp_endpoint = "http://localhost:4317"
 ## Whether to append logs to stdout.
 append_stdout = true
 
+## The log format. Can be `text`/`json`.
+log_format = "text"
+
 ## The percentage of tracing will be sampled and exported.
 ## Valid range `[0, 1]`, 1 means all traces are sampled, 0 means all traces are not sampled, the default value is 1.
 ## ratio > 1 are treated as 1. Fractions < 0 are treated as 0
@@ -177,12 +189,13 @@ enable = false
 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 = "information_schema"
+## @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=information_schema`.
+## The url 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.
@@ -191,5 +204,5 @@ headers = { }
 ## The tracing options. Only effect when compiled with `tokio-console` feature.
 [tracing]
 ## The tokio console address.
-## +toml2docs:none-default
+## @toml2docs:none-default
 tokio_console_addr = "127.0.0.1"

config/standalone.example.toml

@@ -5,15 +5,22 @@ mode = "standalone"
 enable_telemetry = true
 
 ## The default timezone of the server.
-## +toml2docs:none-default
+## @toml2docs:none-default
 default_timezone = "UTC"
 
+## Initialize all regions in the background during the startup.
+## By default, it provides services after all regions have been initialized.
+init_regions_in_background = false
+
+## Parallelism of initializing regions.
+init_regions_parallelism = 16
+
 ## The runtime options.
-[runtime]
+#+ [runtime]
 ## The number of threads to execute the runtime for global read operations.
-global_rt_size = 8
+#+ global_rt_size = 8
 ## The number of threads to execute the runtime for global write operations.
-compact_rt_size = 4
+#+ compact_rt_size = 4
 
 ## The HTTP server options.
 [http]
@@ -39,11 +46,11 @@ runtime_size = 8
 mode = "disable"
 
 ## Certificate file path.
-## +toml2docs:none-default
+## @toml2docs:none-default
 cert_path = ""
 
 ## Private key file path.
-## +toml2docs:none-default
+## @toml2docs:none-default
 key_path = ""
 
 ## Watch for Certificate and key file change and auto reload.
@@ -71,11 +78,11 @@ runtime_size = 2
 mode = "disable"
 
 ## Certificate file path.
-## +toml2docs:none-default
+## @toml2docs:none-default
 cert_path = ""
 
 ## Private key file path.
-## +toml2docs:none-default
+## @toml2docs:none-default
 key_path = ""
 
 ## Watch for Certificate and key file change and auto reload
@@ -96,11 +103,11 @@ runtime_size = 2
 mode = "disable"
 
 ## Certificate file path.
-## +toml2docs:none-default
+## @toml2docs:none-default
 cert_path = ""
 
 ## Private key file path.
-## +toml2docs:none-default
+## @toml2docs:none-default
 key_path = ""
 
 ## Watch for Certificate and key file change and auto reload
@@ -132,7 +139,7 @@ provider = "raft_engine"
 
 ## The directory to store the WAL files.
 ## **It's only used when the provider is `raft_engine`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 dir = "/tmp/greptimedb/wal"
 
 ## The size of the WAL segment file.
@@ -167,11 +174,19 @@ prefill_log_files = false
 ## **It's only used when the provider is `raft_engine`**.
 sync_period = "10s"
 
+## Parallelism during WAL recovery.
+recovery_parallelism = 2
+
 ## The Kafka broker endpoints.
 ## **It's only used when the provider is `kafka`**.
 broker_endpoints = ["127.0.0.1:9092"]
 
-## Number of topics to be created upon start.
+## Automatically create topics for WAL.
+## Set to `true` to automatically create topics for WAL.
+## Otherwise, use topics named `topic_name_prefix_[0..num_topics)`
+auto_create_topics = true
+
+## Number of topics.
 ## **It's only used when the provider is `kafka`**.
 num_topics = 64
 
@@ -182,6 +197,7 @@ num_topics = 64
 selector_type = "round_robin"
 
 ## A Kafka topic is constructed by concatenating `topic_name_prefix` and `topic_id`.
+## i.g., greptimedb_wal_topic_0, greptimedb_wal_topic_1.
 ## **It's only used when the provider is `kafka`**.
 topic_name_prefix = "greptimedb_wal_topic"
 
@@ -304,90 +320,104 @@ type = "File"
 
 ## Cache configuration for object storage such as 'S3' etc.
 ## The local file cache directory.
-## +toml2docs:none-default
+## @toml2docs:none-default
 cache_path = "/path/local_cache"
 
 ## The local file cache capacity in bytes.
-## +toml2docs:none-default
+## @toml2docs:none-default
 cache_capacity = "256MB"
 
 ## The S3 bucket name.
 ## **It's only used when the storage type is `S3`, `Oss` and `Gcs`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 bucket = "greptimedb"
 
 ## The S3 data will be stored in the specified prefix, for example, `s3://${bucket}/${root}`.
 ## **It's only used when the storage type is `S3`, `Oss` and `Azblob`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 root = "greptimedb"
 
 ## The access key id of the aws account.
 ## It's **highly recommended** to use AWS IAM roles instead of hardcoding the access key id and secret key.
 ## **It's only used when the storage type is `S3` and `Oss`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 access_key_id = "test"
 
 ## The secret access key of the aws account.
 ## It's **highly recommended** to use AWS IAM roles instead of hardcoding the access key id and secret key.
 ## **It's only used when the storage type is `S3`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 secret_access_key = "test"
 
 ## The secret access key of the aliyun account.
 ## **It's only used when the storage type is `Oss`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 access_key_secret = "test"
 
 ## The account key of the azure account.
 ## **It's only used when the storage type is `Azblob`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 account_name = "test"
 
 ## The account key of the azure account.
 ## **It's only used when the storage type is `Azblob`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 account_key = "test"
 
 ## The scope of the google cloud storage.
 ## **It's only used when the storage type is `Gcs`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 scope = "test"
 
 ## The credential path of the google cloud storage.
 ## **It's only used when the storage type is `Gcs`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 credential_path = "test"
 
 ## The credential of the google cloud storage.
 ## **It's only used when the storage type is `Gcs`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 credential = "base64-credential"
 
 ## The container of the azure account.
 ## **It's only used when the storage type is `Azblob`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 container = "greptimedb"
 
 ## The sas token of the azure account.
 ## **It's only used when the storage type is `Azblob`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 sas_token = ""
 
 ## The endpoint of the S3 service.
 ## **It's only used when the storage type is `S3`, `Oss`, `Gcs` and `Azblob`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 endpoint = "https://s3.amazonaws.com"
 
 ## The region of the S3 service.
 ## **It's only used when the storage type is `S3`, `Oss`, `Gcs` and `Azblob`**.
-## +toml2docs:none-default
+## @toml2docs:none-default
 region = "us-west-2"
 
 # Custom storage options
 # [[storage.providers]]
+# name = "S3"
 # type = "S3"
+# bucket = "greptimedb"
+# root = "data"
+# access_key_id = "test"
+# secret_access_key = "123456"
+# endpoint = "https://s3.amazonaws.com"
+# region = "us-west-2"
 # [[storage.providers]]
+# name = "Gcs"
 # type = "Gcs"
+# bucket = "greptimedb"
+# root = "data"
+# scope = "test"
+# credential_path = "123456"
+# credential = "base64-credential"
+# endpoint = "https://storage.googleapis.com"
 
 ## The region engine options. You can configure multiple region engines.
 [[region_engine]]
@@ -396,7 +426,7 @@ region = "us-west-2"
 [region_engine.mito]
 
 ## Number of region workers.
-num_workers = 8
+#+ num_workers = 8
 
 ## Request channel size of each worker.
 worker_channel_size = 128
@@ -417,22 +447,32 @@ max_background_jobs = 4
 auto_flush_interval = "1h"
 
 ## Global write buffer size for all regions. If not set, it's default to 1/8 of OS memory with a max limitation of 1GB.
-global_write_buffer_size = "1GB"
+## @toml2docs:none-default="Auto"
+#+ global_write_buffer_size = "1GB"
 
-## Global write buffer size threshold to reject write requests. If not set, it's default to 2 times of `global_write_buffer_size`
-global_write_buffer_reject_size = "2GB"
+## Global write buffer size threshold to reject write requests. If not set, it's default to 2 times of `global_write_buffer_size`.
+## @toml2docs:none-default="Auto"
+#+ global_write_buffer_reject_size = "2GB"
 
 ## Cache size for SST metadata. Setting it to 0 to disable the cache.
 ## If not set, it's default to 1/32 of OS memory with a max limitation of 128MB.
-sst_meta_cache_size = "128MB"
+## @toml2docs:none-default="Auto"
+#+ sst_meta_cache_size = "128MB"
 
 ## Cache size for vectors and arrow arrays. Setting it to 0 to disable the cache.
 ## If not set, it's default to 1/16 of OS memory with a max limitation of 512MB.
-vector_cache_size = "512MB"
+## @toml2docs:none-default="Auto"
+#+ vector_cache_size = "512MB"
 
 ## Cache size for pages of SST row groups. Setting it to 0 to disable the cache.
+## If not set, it's default to 1/8 of OS memory.
+## @toml2docs:none-default="Auto"
+#+ page_cache_size = "512MB"
+
+## Cache size for time series selector (e.g. `last_value()`). Setting it to 0 to disable the cache.
 ## If not set, it's default to 1/16 of OS memory with a max limitation of 512MB.
-page_cache_size = "512MB"
+## @toml2docs:none-default="Auto"
+#+ selector_result_cache_size = "512MB"
 
 ## Whether to enable the experimental write cache.
 enable_experimental_write_cache = false
@@ -444,7 +484,8 @@ experimental_write_cache_path = ""
 experimental_write_cache_size = "512MB"
 
 ## TTL for write cache.
-experimental_write_cache_ttl = "1h"
+## @toml2docs:none-default
+experimental_write_cache_ttl = "8h"
 
 ## Buffer size for SST writing.
 sst_write_buffer_size = "8MB"
@@ -461,6 +502,10 @@ parallel_scan_channel_size = 32
 ## Whether to allow stale WAL entries read during replay.
 allow_stale_entries = false
 
+## Minimum time interval between two compactions.
+## To align with the old behavior, the default value is 0 (no restrictions).
+min_compaction_interval = "0m"
+
 ## The options for index in Mito engine.
 [region_engine.mito.index]
 
@@ -557,11 +602,11 @@ fork_dictionary_bytes = "1GiB"
 
 ## The logging options.
 [logging]
-## The directory to store the log files.
+## The directory to store the log files. If set to empty, logs will not be written to files.
 dir = "/tmp/greptimedb/logs"
 
 ## The log level. Can be `info`/`debug`/`warn`/`error`.
-## +toml2docs:none-default
+## @toml2docs:none-default
 level = "info"
 
 ## Enable OTLP tracing.
@@ -573,6 +618,9 @@ otlp_endpoint = "http://localhost:4317"
 ## Whether to append logs to stdout.
 append_stdout = true
 
+## The log format. Can be `text`/`json`.
+log_format = "text"
+
 ## The percentage of tracing will be sampled and exported.
 ## Valid range `[0, 1]`, 1 means all traces are sampled, 0 means all traces are not sampled, the default value is 1.
 ## ratio > 1 are treated as 1. Fractions < 0 are treated as 0
@@ -589,13 +637,14 @@ enable = false
 ## The interval of export metrics.
 write_interval = "30s"
 
-## For `standalone` mode, `self_import` is recommend to collect metrics generated by itself
+## For `standalone` mode, `self_import` is recommended to collect metrics generated by itself
+## You must create the database before enabling it.
 [export_metrics.self_import]
-## +toml2docs:none-default
-db = "information_schema"
+## @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=information_schema`.
+## The url 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.
@@ -604,5 +653,5 @@ headers = { }
 ## The tracing options. Only effect when compiled with `tokio-console` feature.
 [tracing]
 ## The tokio console address.
-## +toml2docs:none-default
+## @toml2docs:none-default
 tokio_console_addr = "127.0.0.1"

docs/benchmarks/log/README.md

@@ -0,0 +1,51 @@
+# Log benchmark configuration
+This repo holds the configuration we used to benchmark GreptimeDB, Clickhouse and Elastic Search.
+
+Here are the versions of databases we used in the benchmark
+
+| name          | version    |
+| :------------ | :--------- |
+| GreptimeDB    | v0.9.2     |
+| Clickhouse    | 24.9.1.219 |
+| Elasticsearch | 8.15.0     |
+
+## Structured model vs Unstructured model
+We divide test into two parts, using structured model and unstructured model accordingly. You can also see the difference in create table clause.
+
+__Structured model__
+
+The log data is pre-processed into columns by vector. For example an insert request looks like following
+```SQL
+INSERT INTO test_table (bytes, http_version, ip, method, path, status, user, timestamp) VALUES ()
+```
+The goal is to test string/text support for each database. In real scenarios it means the datasource(or log data producers) have separate fields defined, or have already processed the raw input.
+
+__Unstructured model__
+
+The log data is inserted as a long string, and then we build fulltext index upon these strings. For example an insert request looks like following
+```SQL
+INSERT INTO test_table (message, timestamp) VALUES ()
+```
+The goal is to test fuzzy search performance for each database. In real scenarios it means the log is produced by some kind of middleware and inserted directly into the database.
+
+## Creating tables
+See [here](./create_table.sql) for GreptimeDB and Clickhouse's create table clause.
+The mapping of Elastic search is created automatically.
+
+## Vector Configuration
+We use vector to generate random log data and send inserts to databases.
+Please refer to [structured config](./structured_vector.toml) and [unstructured config](./unstructured_vector.toml) for detailed configuration.
+
+## SQLs and payloads
+Please refer to [SQL query](./query.sql) for GreptimeDB and Clickhouse, and [query payload](./query.md) for Elastic search.
+
+## Steps to reproduce
+0. Decide whether to run structured model test or unstructured mode test.
+1. Build vector binary(see vector's config file for specific branch) and databases binaries accordingly.
+2. Create table in GreptimeDB and Clickhouse in advance.
+3. Run vector to insert data.
+4. When data insertion is finished, run queries against each database. Note: you'll need to update timerange value after data insertion.
+
+## Addition
+- You can tune GreptimeDB's configuration to get better performance.
+- You can setup GreptimeDB to use S3 as storage, see [here](https://docs.greptime.com/user-guide/operations/configuration/#storage-options).

docs/benchmarks/log/create_table.sql

@@ -0,0 +1,56 @@
+-- GreptimeDB create table clause
+-- structured test, use vector to pre-process log data into fields
+CREATE TABLE IF NOT EXISTS `test_table` (
+    `bytes` Int64 NULL,
+    `http_version` STRING NULL,
+    `ip` STRING NULL,
+    `method` STRING NULL,
+    `path` STRING NULL,
+    `status` SMALLINT UNSIGNED NULL,
+    `user` STRING NULL,
+    `timestamp` TIMESTAMP(3) NOT NULL,
+    PRIMARY KEY (`user`, `path`, `status`),
+    TIME INDEX (`timestamp`)
+)
+ENGINE=mito
+WITH(
+    append_mode = 'true'
+);
+
+-- unstructured test, build fulltext index on message column
+CREATE TABLE IF NOT EXISTS `test_table` (
+    `message` STRING NULL FULLTEXT WITH(analyzer = 'English', case_sensitive = 'false'),
+    `timestamp` TIMESTAMP(3) NOT NULL,
+    TIME INDEX (`timestamp`)
+)
+ENGINE=mito
+WITH(
+    append_mode = 'true'
+);
+
+-- Clickhouse create table clause
+-- structured test
+CREATE TABLE IF NOT EXISTS test_table
+(
+    bytes UInt64 NOT NULL,
+    http_version String NOT NULL,
+    ip String NOT NULL,
+    method String NOT NULL,
+    path String NOT NULL,
+    status UInt8 NOT NULL,
+    user String NOT NULL,
+    timestamp String NOT NULL,
+)
+ENGINE = MergeTree()
+ORDER BY (user, path, status);
+
+-- unstructured test
+SET allow_experimental_full_text_index = true;
+CREATE TABLE IF NOT EXISTS test_table
+(
+    message String,
+    timestamp String,
+    INDEX inv_idx(message) TYPE full_text(0) GRANULARITY 1
+)
+ENGINE = MergeTree()
+ORDER BY tuple();

docs/benchmarks/log/query.md

@@ -0,0 +1,199 @@
+# Query URL and payload for Elastic Search
+## Count
+URL: `http://127.0.0.1:9200/_count`
+
+## Query by timerange
+URL: `http://127.0.0.1:9200/_search`
+
+You can use the following payload to get the full timerange first.
+```JSON
+{"size":0,"aggs":{"max_timestamp":{"max":{"field":"timestamp"}},"min_timestamp":{"min":{"field":"timestamp"}}}}
+```
+
+And then use this payload to query by timerange.
+```JSON
+{
+  "from": 0,
+  "size": 1000,
+  "query": {
+    "range": {
+      "timestamp": {
+        "gte": "2024-08-16T04:30:44.000Z",
+        "lte": "2024-08-16T04:51:52.000Z"
+      }
+    }
+  }
+}
+```
+
+## Query by condition
+URL: `http://127.0.0.1:9200/_search`
+### Structured payload
+```JSON
+{
+  "from": 0,
+  "size": 10000,
+  "query": {
+    "bool": {
+      "must": [
+        {
+          "term": {
+            "user.keyword": "CrucifiX"
+          }
+        },
+        {
+          "term": {
+            "method.keyword": "OPTION"
+          }
+        },
+        {
+          "term": {
+            "path.keyword": "/user/booperbot124"
+          }
+        },
+        {
+          "term": {
+            "http_version.keyword": "HTTP/1.1"
+          }
+        },
+        {
+          "term": {
+            "status": "401"
+          }
+        }
+      ]
+    }
+  }
+}
+```
+### Unstructured payload
+```JSON
+{
+  "from": 0,
+  "size": 10000,
+  "query": {
+    "bool": {
+      "must": [
+        {
+          "match_phrase": {
+            "message": "CrucifiX"
+          }
+        },
+        {
+          "match_phrase": {
+            "message": "OPTION"
+          }
+        },
+        {
+          "match_phrase": {
+            "message": "/user/booperbot124"
+          }
+        },
+        {
+          "match_phrase": {
+            "message": "HTTP/1.1"
+          }
+        },
+        {
+          "match_phrase": {
+            "message": "401"
+          }
+        }
+      ]
+    }
+  }
+}
+```
+
+## Query by condition and timerange
+URL: `http://127.0.0.1:9200/_search`
+### Structured payload
+```JSON
+{
+  "size": 10000,
+  "query": {
+    "bool": {
+      "must": [
+        {
+          "term": {
+            "user.keyword": "CrucifiX"
+          }
+        },
+        {
+          "term": {
+            "method.keyword": "OPTION"
+          }
+        },
+        {
+          "term": {
+            "path.keyword": "/user/booperbot124"
+          }
+        },
+        {
+          "term": {
+            "http_version.keyword": "HTTP/1.1"
+          }
+        },
+        {
+          "term": {
+            "status": "401"
+          }
+        },
+        {
+          "range": {
+            "timestamp": {
+              "gte": "2024-08-19T07:03:37.383Z",
+              "lte": "2024-08-19T07:24:58.883Z"
+            }
+          }
+        }
+      ]
+    }
+  }
+}
+```
+### Unstructured payload
+```JSON
+{
+  "size": 10000,
+  "query": {
+    "bool": {
+      "must": [
+        {
+          "match_phrase": {
+            "message": "CrucifiX"
+          }
+        },
+        {
+          "match_phrase": {
+            "message": "OPTION"
+          }
+        },
+        {
+          "match_phrase": {
+            "message": "/user/booperbot124"
+          }
+        },
+        {
+          "match_phrase": {
+            "message": "HTTP/1.1"
+          }
+        },
+        {
+          "match_phrase": {
+            "message": "401"
+          }
+        },
+        {
+          "range": {
+            "timestamp": {
+              "gte": "2024-08-19T05:16:17.099Z",
+              "lte": "2024-08-19T05:46:02.722Z"
+            }
+          }
+        }
+      ]
+    }
+  }
+}
+```

docs/benchmarks/log/query.sql

@@ -0,0 +1,50 @@
+-- Structured query for GreptimeDB and Clickhouse
+
+-- query count
+select count(*) from test_table;
+
+-- query by timerange. Note: place the timestamp range in the where clause
+-- GreptimeDB
+-- you can use `select max(timestamp)::bigint from test_table;` and `select min(timestamp)::bigint from test_table;`
+-- to get the full timestamp range
+select * from test_table where timestamp between 1723710843619 and 1723711367588;
+-- Clickhouse
+-- you can use `select max(timestamp) from test_table;` and `select min(timestamp) from test_table;`
+-- to get the full timestamp range
+select * from test_table where timestamp between '2024-08-16T03:58:46Z' and '2024-08-16T04:03:50Z';
+
+-- query by condition
+SELECT * FROM test_table WHERE user = 'CrucifiX' and method = 'OPTION' and path = '/user/booperbot124' and http_version = 'HTTP/1.1' and status = 401;
+
+-- query by condition and timerange
+-- GreptimeDB
+SELECT * FROM test_table WHERE user = "CrucifiX" and method = "OPTION" and path = "/user/booperbot124" and http_version = "HTTP/1.1" and status = 401 
+and timestamp between 1723774396760 and 1723774788760;
+-- Clickhouse
+SELECT * FROM test_table WHERE user = 'CrucifiX' and method = 'OPTION' and path = '/user/booperbot124' and http_version = 'HTTP/1.1' and status = 401 
+and timestamp between '2024-08-16T03:58:46Z' and '2024-08-16T04:03:50Z';
+
+-- Unstructured query for GreptimeDB and Clickhouse
+
+
+-- query by condition
+-- GreptimeDB
+SELECT * FROM test_table WHERE MATCHES(message, "+CrucifiX +OPTION +/user/booperbot124 +HTTP/1.1 +401");
+-- Clickhouse
+SELECT * FROM test_table WHERE (message LIKE '%CrucifiX%') 
+AND (message LIKE '%OPTION%') 
+AND (message LIKE '%/user/booperbot124%') 
+AND (message LIKE '%HTTP/1.1%') 
+AND (message LIKE '%401%');
+
+-- query by condition and timerange
+-- GreptimeDB
+SELECT * FROM test_table WHERE MATCHES(message, "+CrucifiX +OPTION +/user/booperbot124 +HTTP/1.1 +401") 
+and timestamp between 1723710843619 and 1723711367588;
+-- Clickhouse
+SELECT * FROM test_table WHERE (message LIKE '%CrucifiX%') 
+AND (message LIKE '%OPTION%') 
+AND (message LIKE '%/user/booperbot124%') 
+AND (message LIKE '%HTTP/1.1%') 
+AND (message LIKE '%401%') 
+AND timestamp between '2024-08-15T10:25:26.524000000Z' AND '2024-08-15T10:31:31.746000000Z';

docs/benchmarks/log/structured_vector.toml

@@ -0,0 +1,57 @@
+# Please note we use patched branch to build vector
+# https://github.com/shuiyisong/vector/tree/chore/greptime_log_ingester_logitem
+
+[sources.demo_logs]
+type = "demo_logs"
+format = "apache_common"
+# interval value = 1 / rps
+# say you want to insert at 20k/s, that is 1 / 20000 = 0.00005
+# set to 0 to run as fast as possible
+interval = 0
+# total rows to insert
+count = 100000000
+lines = [ "line1" ]
+
+[transforms.parse_logs]
+type = "remap"
+inputs = ["demo_logs"]
+source = '''
+. = parse_regex!(.message, r'^(?P<ip>\S+) - (?P<user>\S+) \[(?P<timestamp>[^\]]+)\] "(?P<method>\S+) (?P<path>\S+) (?P<http_version>\S+)" (?P<status>\d+) (?P<bytes>\d+)$')
+
+# Convert timestamp to a standard format
+.timestamp = parse_timestamp!(.timestamp, format: "%d/%b/%Y:%H:%M:%S %z")
+
+# Convert status and bytes to integers
+.status = to_int!(.status)
+.bytes = to_int!(.bytes)
+'''
+
+[sinks.sink_greptime_logs]
+type = "greptimedb_logs"
+# The table to insert into
+table = "test_table"
+pipeline_name = "demo_pipeline"
+compression = "none"
+inputs = [ "parse_logs" ]
+endpoint = "http://127.0.0.1:4000"
+# Batch size for each insertion
+batch.max_events = 4000
+
+[sinks.clickhouse]
+type = "clickhouse"
+inputs = [ "parse_logs" ]
+database = "default"
+endpoint = "http://127.0.0.1:8123"
+format = "json_each_row"
+# The table to insert into
+table = "test_table"
+
+[sinks.sink_elasticsearch]
+type = "elasticsearch"
+inputs = [ "parse_logs" ]
+api_version = "auto"
+compression = "none"
+doc_type = "_doc"
+endpoints = [ "http://127.0.0.1:9200" ]
+id_key = "id"
+mode = "bulk"

docs/benchmarks/log/unstructured_vector.toml

@@ -0,0 +1,43 @@
+# Please note we use patched branch to build vector
+# https://github.com/shuiyisong/vector/tree/chore/greptime_log_ingester_ft
+
+[sources.demo_logs]
+type = "demo_logs"
+format = "apache_common"
+# interval value = 1 / rps
+# say you want to insert at 20k/s, that is 1 / 20000 = 0.00005
+# set to 0 to run as fast as possible
+interval = 0
+# total rows to insert
+count = 100000000
+lines = [ "line1" ]
+
+[sinks.sink_greptime_logs]
+type = "greptimedb_logs"
+# The table to insert into
+table = "test_table"
+pipeline_name = "demo_pipeline"
+compression = "none"
+inputs = [ "demo_logs" ]
+endpoint = "http://127.0.0.1:4000"
+# Batch size for each insertion
+batch.max_events = 500
+
+[sinks.clickhouse]
+type = "clickhouse"
+inputs = [ "demo_logs" ]
+database = "default"
+endpoint = "http://127.0.0.1:8123"
+format = "json_each_row"
+# The table to insert into
+table = "test_table"
+
+[sinks.sink_elasticsearch]
+type = "elasticsearch"
+inputs = [ "demo_logs" ]
+api_version = "auto"
+compression = "none"
+doc_type = "_doc"
+endpoints = [ "http://127.0.0.1:9200" ]
+id_key = "id"
+mode = "bulk"

grafana/greptimedb-cluster.json

@@ -157,26 +157,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -326,26 +306,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -497,26 +457,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -668,26 +608,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -852,26 +772,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -1023,26 +923,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -1194,26 +1074,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -1365,26 +1225,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -1536,26 +1376,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -1707,26 +1527,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -1878,26 +1678,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -2062,26 +1842,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -2233,26 +1993,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -2417,26 +2157,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -2569,26 +2289,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -2751,26 +2451,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -3091,28 +2771,7 @@
           },
           "unit": "s"
         },
-        "overrides": [
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
-          }
-        ]
+        "overrides": []
       },
       "gridPos": {
         "h": 8,
@@ -3242,26 +2901,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -3429,26 +3068,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -3598,26 +3217,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -3740,28 +3339,7 @@
           },
           "unit": "s"
         },
-        "overrides": [
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
-          }
-        ]
+        "overrides": []
       },
       "gridPos": {
         "h": 8,
@@ -4101,26 +3679,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -4270,26 +3828,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -4439,26 +3977,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -4608,26 +4126,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -4777,26 +4275,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -4946,26 +4424,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -5115,26 +4573,6 @@
                 }
               }
             ]
-          },
-          {
-            "matcher": {
-              "id": "byValue",
-              "options": {
-                "op": "gte",
-                "reducer": "allIsZero",
-                "value": 0
-              }
-            },
-            "properties": [
-              {
-                "id": "custom.hideFrom",
-                "value": {
-                  "legend": true,
-                  "tooltip": true,
-                  "viz": true
-                }
-              }
-            ]
           }
         ]
       },
@@ -5241,4 +4679,4 @@
   "uid": "ea35efe5-918e-44fa-9743-e9aa1a340a3f",
   "version": 11,
   "weekStart": ""
-}
+}

rust-toolchain.toml

@@ -1,2 +1,3 @@
 [toolchain]
-channel = "nightly-2024-04-20"
+channel = "nightly-2024-06-06"
+

scripts/check-builder-rust-version.sh

@@ -0,0 +1,42 @@
+#!/usr/bin/env bash
+
+set -e
+
+RUST_TOOLCHAIN_VERSION_FILE="rust-toolchain.toml"
+DEV_BUILDER_UBUNTU_REGISTRY="docker.io"
+DEV_BUILDER_UBUNTU_NAMESPACE="greptime"
+DEV_BUILDER_UBUNTU_NAME="dev-builder-ubuntu"
+
+function check_rust_toolchain_version() {
+  DEV_BUILDER_IMAGE_TAG=$(grep "DEV_BUILDER_IMAGE_TAG ?= " Makefile | cut -d= -f2 | sed 's/^[ \t]*//')
+  if [ -z "$DEV_BUILDER_IMAGE_TAG" ]; then
+    echo "Error: No DEV_BUILDER_IMAGE_TAG found in Makefile"
+    exit 1
+  fi
+
+  DEV_BUILDER_UBUNTU_IMAGE="$DEV_BUILDER_UBUNTU_REGISTRY/$DEV_BUILDER_UBUNTU_NAMESPACE/$DEV_BUILDER_UBUNTU_NAME:$DEV_BUILDER_IMAGE_TAG"
+
+  CURRENT_VERSION=$(grep -Eo '[0-9]{4}-[0-9]{2}-[0-9]{2}' "$RUST_TOOLCHAIN_VERSION_FILE")
+  if [ -z "$CURRENT_VERSION" ]; then
+    echo "Error: No rust toolchain version found in $RUST_TOOLCHAIN_VERSION_FILE"
+    exit 1
+  fi
+
+  RUST_TOOLCHAIN_VERSION_IN_BUILDER=$(docker run "$DEV_BUILDER_UBUNTU_IMAGE" rustc --version | grep  -Eo '[0-9]{4}-[0-9]{2}-[0-9]{2}')
+  if [ -z "$RUST_TOOLCHAIN_VERSION_IN_BUILDER" ]; then
+    echo "Error: No rustc version found in $DEV_BUILDER_UBUNTU_IMAGE"
+    exit 1
+  fi
+
+  # Compare the version and the difference should be less than 1 day.
+  current_rust_toolchain_seconds=$(date -d "$CURRENT_VERSION" +%s)
+  rust_toolchain_in_dev_builder_ubuntu_seconds=$(date -d "$RUST_TOOLCHAIN_VERSION_IN_BUILDER" +%s)
+  date_diff=$(( (current_rust_toolchain_seconds - rust_toolchain_in_dev_builder_ubuntu_seconds) / 86400 ))
+
+  if [ $date_diff -gt 1 ]; then
+    echo "Error: The rust toolchain '$RUST_TOOLCHAIN_VERSION_IN_BUILDER' in builder '$DEV_BUILDER_UBUNTU_IMAGE' maybe outdated, please update it to '$CURRENT_VERSION'"
+    exit 1
+  fi
+}
+
+check_rust_toolchain_version

scripts/check-snafu.py

@@ -0,0 +1,69 @@
+# 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 os
+import re
+
+
+def find_rust_files(directory):
+    error_files = []
+    other_rust_files = []
+    for root, _, files in os.walk(directory):
+        for file in files:
+            if file == "error.rs":
+                error_files.append(os.path.join(root, file))
+            elif file.endswith(".rs"):
+                other_rust_files.append(os.path.join(root, file))
+    return error_files, other_rust_files
+
+
+def extract_branch_names(file_content):
+    pattern = re.compile(r"#\[snafu\(display\([^\)]*\)\)\]\s*(\w+)\s*\{")
+    return pattern.findall(file_content)
+
+
+def check_snafu_in_files(branch_name, rust_files):
+    branch_name_snafu = f"{branch_name}Snafu"
+    for rust_file in rust_files:
+        with open(rust_file, "r") as file:
+            content = file.read()
+            if branch_name_snafu in content:
+                return True
+    return False
+
+
+def main():
+    error_files, other_rust_files = find_rust_files(".")
+    branch_names = []
+
+    for error_file in error_files:
+        with open(error_file, "r") as file:
+            content = file.read()
+            branch_names.extend(extract_branch_names(content))
+
+    unused_snafu = [
+        branch_name
+        for branch_name in branch_names
+        if not check_snafu_in_files(branch_name, other_rust_files)
+    ]
+
+    for name in unused_snafu:
+        print(name)
+
+    if unused_snafu:
+        raise SystemExit(1)
+
+
+if __name__ == "__main__":
+    main()

src/api/src/helper.rs

@@ -42,7 +42,8 @@ use greptime_proto::v1::greptime_request::Request;
 use greptime_proto::v1::query_request::Query;
 use greptime_proto::v1::value::ValueData;
 use greptime_proto::v1::{
-    ColumnDataTypeExtension, DdlRequest, DecimalTypeExtension, QueryRequest, Row, SemanticType,
+    ColumnDataTypeExtension, DdlRequest, DecimalTypeExtension, JsonTypeExtension, QueryRequest,
+    Row, SemanticType,
 };
 use paste::paste;
 use snafu::prelude::*;
@@ -103,7 +104,17 @@ impl From<ColumnDataTypeWrapper> for ConcreteDataType {
             ColumnDataType::Uint64 => ConcreteDataType::uint64_datatype(),
             ColumnDataType::Float32 => ConcreteDataType::float32_datatype(),
             ColumnDataType::Float64 => ConcreteDataType::float64_datatype(),
-            ColumnDataType::Binary => ConcreteDataType::binary_datatype(),
+            ColumnDataType::Binary => {
+                if let Some(TypeExt::JsonType(_)) = datatype_wrapper
+                    .datatype_ext
+                    .as_ref()
+                    .and_then(|datatype_ext| datatype_ext.type_ext.as_ref())
+                {
+                    ConcreteDataType::json_datatype()
+                } else {
+                    ConcreteDataType::binary_datatype()
+                }
+            }
             ColumnDataType::String => ConcreteDataType::string_datatype(),
             ColumnDataType::Date => ConcreteDataType::date_datatype(),
             ColumnDataType::Datetime => ConcreteDataType::datetime_datatype(),
@@ -236,7 +247,7 @@ impl TryFrom<ConcreteDataType> for ColumnDataTypeWrapper {
             ConcreteDataType::UInt64(_) => ColumnDataType::Uint64,
             ConcreteDataType::Float32(_) => ColumnDataType::Float32,
             ConcreteDataType::Float64(_) => ColumnDataType::Float64,
-            ConcreteDataType::Binary(_) => ColumnDataType::Binary,
+            ConcreteDataType::Binary(_) | ConcreteDataType::Json(_) => ColumnDataType::Binary,
             ConcreteDataType::String(_) => ColumnDataType::String,
             ConcreteDataType::Date(_) => ColumnDataType::Date,
             ConcreteDataType::DateTime(_) => ColumnDataType::Datetime,
@@ -276,6 +287,16 @@ impl TryFrom<ConcreteDataType> for ColumnDataTypeWrapper {
                         })),
                     })
             }
+            ColumnDataType::Binary => {
+                if datatype == ConcreteDataType::json_datatype() {
+                    // Json is the same as  binary in proto. The extension marks the binary in proto is actually a json.
+                    Some(ColumnDataTypeExtension {
+                        type_ext: Some(TypeExt::JsonType(JsonTypeExtension::JsonBinary.into())),
+                    })
+                } else {
+                    None
+                }
+            }
             _ => None,
         };
         Ok(Self {
@@ -649,7 +670,8 @@ pub fn pb_values_to_vector_ref(data_type: &ConcreteDataType, values: Values) ->
         ConcreteDataType::Null(_)
         | ConcreteDataType::List(_)
         | ConcreteDataType::Dictionary(_)
-        | ConcreteDataType::Duration(_) => {
+        | ConcreteDataType::Duration(_)
+        | ConcreteDataType::Json(_) => {
             unreachable!()
         }
     }
@@ -813,7 +835,8 @@ pub fn pb_values_to_values(data_type: &ConcreteDataType, values: Values) -> Vec<
         ConcreteDataType::Null(_)
         | ConcreteDataType::List(_)
         | ConcreteDataType::Dictionary(_)
-        | ConcreteDataType::Duration(_) => {
+        | ConcreteDataType::Duration(_)
+        | ConcreteDataType::Json(_) => {
             unreachable!()
         }
     }
@@ -831,7 +854,13 @@ pub fn is_column_type_value_eq(
     expect_type: &ConcreteDataType,
 ) -> bool {
     ColumnDataTypeWrapper::try_new(type_value, type_extension)
-        .map(|wrapper| ConcreteDataType::from(wrapper) == *expect_type)
+        .map(|wrapper| {
+            let datatype = ConcreteDataType::from(wrapper);
+            (datatype == *expect_type)
+            // Json type leverage binary type in pb, so this is valid.
+                || (datatype == ConcreteDataType::binary_datatype()
+                    && *expect_type == ConcreteDataType::json_datatype())
+        })
         .unwrap_or(false)
 }
 

src/api/src/region.rs

@@ -21,14 +21,14 @@ use greptime_proto::v1::region::RegionResponse as RegionResponseV1;
 #[derive(Debug)]
 pub struct RegionResponse {
     pub affected_rows: AffectedRows,
-    pub extension: HashMap<String, Vec<u8>>,
+    pub extensions: HashMap<String, Vec<u8>>,
 }
 
 impl RegionResponse {
     pub fn from_region_response(region_response: RegionResponseV1) -> Self {
         Self {
             affected_rows: region_response.affected_rows as _,
-            extension: region_response.extension,
+            extensions: region_response.extensions,
         }
     }
 
@@ -36,7 +36,7 @@ impl RegionResponse {
     pub fn new(affected_rows: AffectedRows) -> Self {
         Self {
             affected_rows,
-            extension: Default::default(),
+            extensions: Default::default(),
         }
     }
 }

src/auth/src/tests.rs

@@ -13,9 +13,11 @@
 // limitations under the License.
 
 use common_base::secrets::ExposeSecret;
+use common_error::ext::BoxedError;
+use snafu::{OptionExt, ResultExt};
 
 use crate::error::{
-    AccessDeniedSnafu, Result, UnsupportedPasswordTypeSnafu, UserNotFoundSnafu,
+    AccessDeniedSnafu, AuthBackendSnafu, Result, UnsupportedPasswordTypeSnafu, UserNotFoundSnafu,
     UserPasswordMismatchSnafu,
 };
 use crate::user_info::DefaultUserInfo;
@@ -49,6 +51,19 @@ impl MockUserProvider {
         info.schema.clone_into(&mut self.schema);
         info.username.clone_into(&mut self.username);
     }
+
+    // this is a deliberate function to ref AuthBackendSnafu
+    // so that it won't get deleted in the future
+    pub fn ref_auth_backend_snafu(&self) -> Result<()> {
+        let none_option = None;
+
+        none_option
+            .context(UserNotFoundSnafu {
+                username: "no_user".to_string(),
+            })
+            .map_err(BoxedError::new)
+            .context(AuthBackendSnafu)
+    }
 }
 
 #[async_trait::async_trait]

src/auth/tests/mod.rs

@@ -18,6 +18,7 @@ use std::sync::Arc;
 
 use api::v1::greptime_request::Request;
 use auth::error::Error::InternalState;
+use auth::error::InternalStateSnafu;
 use auth::{PermissionChecker, PermissionCheckerRef, PermissionReq, PermissionResp, UserInfoRef};
 use sql::statements::show::{ShowDatabases, ShowKind};
 use sql::statements::statement::Statement;
@@ -33,9 +34,10 @@ impl PermissionChecker for DummyPermissionChecker {
         match req {
             PermissionReq::GrpcRequest(_) => Ok(PermissionResp::Allow),
             PermissionReq::SqlStatement(_) => Ok(PermissionResp::Reject),
-            _ => Err(InternalState {
+            _ => InternalStateSnafu {
                 msg: "testing".to_string(),
-            }),
+            }
+            .fail(),
         }
     }
 }

src/catalog/src/error.rs

@@ -97,13 +97,6 @@ pub enum Error {
         source: table::error::Error,
     },
 
-    #[snafu(display("System catalog is not valid: {}", msg))]
-    SystemCatalog {
-        msg: String,
-        #[snafu(implicit)]
-        location: Location,
-    },
-
     #[snafu(display("Cannot find catalog by name: {}", catalog_name))]
     CatalogNotFound {
         catalog_name: String,
@@ -186,13 +179,6 @@ pub enum Error {
         source: common_query::error::Error,
     },
 
-    #[snafu(display("Failed to perform metasrv operation"))]
-    Metasrv {
-        #[snafu(implicit)]
-        location: Location,
-        source: meta_client::error::Error,
-    },
-
     #[snafu(display("Invalid table info in catalog"))]
     InvalidTableInfoInCatalog {
         #[snafu(implicit)]
@@ -288,8 +274,6 @@ impl ErrorExt for Error {
 
             Error::FlowInfoNotFound { .. } => StatusCode::FlowNotFound,
 
-            Error::SystemCatalog { .. } => StatusCode::StorageUnavailable,
-
             Error::UpgradeWeakCatalogManagerRef { .. } => StatusCode::Internal,
 
             Error::CreateRecordBatch { source, .. } => source.status_code(),
@@ -303,7 +287,6 @@ impl ErrorExt for Error {
 
             Error::CreateTable { source, .. } => source.status_code(),
 
-            Error::Metasrv { source, .. } => source.status_code(),
             Error::DecodePlan { source, .. } => source.status_code(),
             Error::InvalidTableInfoInCatalog { source, .. } => source.status_code(),
 
@@ -338,27 +321,6 @@ mod tests {
 
     use super::*;
 
-    #[test]
-    pub fn test_error_status_code() {
-        assert_eq!(
-            StatusCode::TableAlreadyExists,
-            Error::TableExists {
-                table: "some_table".to_string(),
-                location: Location::generate(),
-            }
-            .status_code()
-        );
-
-        assert_eq!(
-            StatusCode::StorageUnavailable,
-            Error::SystemCatalog {
-                msg: String::default(),
-                location: Location::generate(),
-            }
-            .status_code()
-        );
-    }
-
     #[test]
     pub fn test_errors_to_datafusion_error() {
         let e: DataFusionError = Error::TableExists {

src/catalog/src/kvbackend/client.rs

@@ -20,8 +20,8 @@ use std::time::Duration;
 
 use common_error::ext::BoxedError;
 use common_meta::cache_invalidator::KvCacheInvalidator;
-use common_meta::error::Error::{CacheNotGet, GetKvCache};
-use common_meta::error::{CacheNotGetSnafu, Error, ExternalSnafu, Result};
+use common_meta::error::Error::CacheNotGet;
+use common_meta::error::{CacheNotGetSnafu, Error, ExternalSnafu, GetKvCacheSnafu, Result};
 use common_meta::kv_backend::{KvBackend, KvBackendRef, TxnService};
 use common_meta::rpc::store::{
     BatchDeleteRequest, BatchDeleteResponse, BatchGetRequest, BatchGetResponse, BatchPutRequest,
@@ -282,8 +282,11 @@ impl KvBackend for CachedMetaKvBackend {
                 _ => Err(e),
             },
         }
-        .map_err(|e| GetKvCache {
-            err_msg: e.to_string(),
+        .map_err(|e| {
+            GetKvCacheSnafu {
+                err_msg: e.to_string(),
+            }
+            .build()
         });
 
         // "cache.invalidate_key" and "cache.try_get_with_by_ref" are not mutually exclusive. So we need

src/catalog/src/kvbackend/manager.rs

@@ -36,6 +36,7 @@ use futures_util::{StreamExt, TryStreamExt};
 use meta_client::client::MetaClient;
 use moka::sync::Cache;
 use partition::manager::{PartitionRuleManager, PartitionRuleManagerRef};
+use session::context::{Channel, QueryContext};
 use snafu::prelude::*;
 use table::dist_table::DistTable;
 use table::table::numbers::{NumbersTable, NUMBERS_TABLE_NAME};
@@ -152,7 +153,11 @@ impl CatalogManager for KvBackendCatalogManager {
         Ok(keys)
     }
 
-    async fn schema_names(&self, catalog: &str) -> Result<Vec<String>> {
+    async fn schema_names(
+        &self,
+        catalog: &str,
+        query_ctx: Option<&QueryContext>,
+    ) -> Result<Vec<String>> {
         let stream = self
             .table_metadata_manager
             .schema_manager()
@@ -163,12 +168,17 @@ impl CatalogManager for KvBackendCatalogManager {
             .map_err(BoxedError::new)
             .context(ListSchemasSnafu { catalog })?;
 
-        keys.extend(self.system_catalog.schema_names());
+        keys.extend(self.system_catalog.schema_names(query_ctx));
 
         Ok(keys.into_iter().collect())
     }
 
-    async fn table_names(&self, catalog: &str, schema: &str) -> Result<Vec<String>> {
+    async fn table_names(
+        &self,
+        catalog: &str,
+        schema: &str,
+        query_ctx: Option<&QueryContext>,
+    ) -> Result<Vec<String>> {
         let stream = self
             .table_metadata_manager
             .table_name_manager()
@@ -181,7 +191,7 @@ impl CatalogManager for KvBackendCatalogManager {
             .into_iter()
             .map(|(k, _)| k)
             .collect::<Vec<_>>();
-        tables.extend_from_slice(&self.system_catalog.table_names(schema));
+        tables.extend_from_slice(&self.system_catalog.table_names(schema, query_ctx));
 
         Ok(tables.into_iter().collect())
     }
@@ -194,8 +204,13 @@ impl CatalogManager for KvBackendCatalogManager {
             .context(TableMetadataManagerSnafu)
     }
 
-    async fn schema_exists(&self, catalog: &str, schema: &str) -> Result<bool> {
-        if self.system_catalog.schema_exists(schema) {
+    async fn schema_exists(
+        &self,
+        catalog: &str,
+        schema: &str,
+        query_ctx: Option<&QueryContext>,
+    ) -> Result<bool> {
+        if self.system_catalog.schema_exists(schema, query_ctx) {
             return Ok(true);
         }
 
@@ -206,8 +221,14 @@ impl CatalogManager for KvBackendCatalogManager {
             .context(TableMetadataManagerSnafu)
     }
 
-    async fn table_exists(&self, catalog: &str, schema: &str, table: &str) -> Result<bool> {
-        if self.system_catalog.table_exists(schema, table) {
+    async fn table_exists(
+        &self,
+        catalog: &str,
+        schema: &str,
+        table: &str,
+        query_ctx: Option<&QueryContext>,
+    ) -> Result<bool> {
+        if self.system_catalog.table_exists(schema, table, query_ctx) {
             return Ok(true);
         }
 
@@ -225,10 +246,12 @@ impl CatalogManager for KvBackendCatalogManager {
         catalog_name: &str,
         schema_name: &str,
         table_name: &str,
+        query_ctx: Option<&QueryContext>,
     ) -> Result<Option<TableRef>> {
-        if let Some(table) = self
-            .system_catalog
-            .table(catalog_name, schema_name, table_name)
+        let channel = query_ctx.map_or(Channel::Unknown, |ctx| ctx.channel());
+        if let Some(table) =
+            self.system_catalog
+                .table(catalog_name, schema_name, table_name, query_ctx)
         {
             return Ok(Some(table));
         }
@@ -236,23 +259,45 @@ impl CatalogManager for KvBackendCatalogManager {
         let table_cache: TableCacheRef = self.cache_registry.get().context(CacheNotFoundSnafu {
             name: "table_cache",
         })?;
-
-        table_cache
+        if let Some(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)
+            .context(GetTableCacheSnafu)?
+        {
+            return Ok(Some(table));
+        }
+
+        if channel == Channel::Postgres {
+            // falldown to pg_catalog
+            if let Some(table) =
+                self.system_catalog
+                    .table(catalog_name, PG_CATALOG_NAME, table_name, query_ctx)
+            {
+                return Ok(Some(table));
+            }
+        }
+
+        return Ok(None);
     }
 
-    fn tables<'a>(&'a self, catalog: &'a str, schema: &'a str) -> BoxStream<'a, Result<TableRef>> {
+    fn tables<'a>(
+        &'a self,
+        catalog: &'a str,
+        schema: &'a str,
+        query_ctx: Option<&'a QueryContext>,
+    ) -> BoxStream<'a, Result<TableRef>> {
         let sys_tables = try_stream!({
             // System tables
-            let sys_table_names = self.system_catalog.table_names(schema);
+            let sys_table_names = self.system_catalog.table_names(schema, query_ctx);
             for table_name in sys_table_names {
-                if let Some(table) = self.system_catalog.table(catalog, schema, &table_name) {
+                if let Some(table) =
+                    self.system_catalog
+                        .table(catalog, schema, &table_name, query_ctx)
+                {
                     yield table;
                 }
             }
@@ -313,25 +358,34 @@ struct SystemCatalog {
     catalog_cache: Cache<String, Arc<InformationSchemaProvider>>,
     pg_catalog_cache: Cache<String, Arc<PGCatalogProvider>>,
 
-    // system_schema_provier for default catalog
+    // system_schema_provider for default catalog
     information_schema_provider: Arc<InformationSchemaProvider>,
     pg_catalog_provider: Arc<PGCatalogProvider>,
     backend: KvBackendRef,
 }
 
 impl SystemCatalog {
-    // TODO(j0hn50n133): remove the duplicated hard-coded table names logic
-    fn schema_names(&self) -> Vec<String> {
-        vec![
-            INFORMATION_SCHEMA_NAME.to_string(),
-            PG_CATALOG_NAME.to_string(),
-        ]
+    fn schema_names(&self, query_ctx: Option<&QueryContext>) -> Vec<String> {
+        let channel = query_ctx.map_or(Channel::Unknown, |ctx| ctx.channel());
+        match channel {
+            // pg_catalog only visible under postgres protocol
+            Channel::Postgres => vec![
+                INFORMATION_SCHEMA_NAME.to_string(),
+                PG_CATALOG_NAME.to_string(),
+            ],
+            _ => {
+                vec![INFORMATION_SCHEMA_NAME.to_string()]
+            }
+        }
     }
 
-    fn table_names(&self, schema: &str) -> Vec<String> {
+    fn table_names(&self, schema: &str, query_ctx: Option<&QueryContext>) -> Vec<String> {
+        let channel = query_ctx.map_or(Channel::Unknown, |ctx| ctx.channel());
         match schema {
             INFORMATION_SCHEMA_NAME => self.information_schema_provider.table_names(),
-            PG_CATALOG_NAME => self.pg_catalog_provider.table_names(),
+            PG_CATALOG_NAME if channel == Channel::Postgres => {
+                self.pg_catalog_provider.table_names()
+            }
             DEFAULT_SCHEMA_NAME => {
                 vec![NUMBERS_TABLE_NAME.to_string()]
             }
@@ -339,23 +393,35 @@ impl SystemCatalog {
         }
     }
 
-    fn schema_exists(&self, schema: &str) -> bool {
-        schema == INFORMATION_SCHEMA_NAME || schema == PG_CATALOG_NAME
+    fn schema_exists(&self, schema: &str, query_ctx: Option<&QueryContext>) -> bool {
+        let channel = query_ctx.map_or(Channel::Unknown, |ctx| ctx.channel());
+        match channel {
+            Channel::Postgres => schema == PG_CATALOG_NAME || schema == INFORMATION_SCHEMA_NAME,
+            _ => schema == INFORMATION_SCHEMA_NAME,
+        }
     }
 
-    fn table_exists(&self, schema: &str, table: &str) -> bool {
+    fn table_exists(&self, schema: &str, table: &str, query_ctx: Option<&QueryContext>) -> bool {
+        let channel = query_ctx.map_or(Channel::Unknown, |ctx| ctx.channel());
         if schema == INFORMATION_SCHEMA_NAME {
             self.information_schema_provider.table(table).is_some()
         } else if schema == DEFAULT_SCHEMA_NAME {
             table == NUMBERS_TABLE_NAME
-        } else if schema == PG_CATALOG_NAME {
+        } else if schema == PG_CATALOG_NAME && channel == Channel::Postgres {
             self.pg_catalog_provider.table(table).is_some()
         } else {
             false
         }
     }
 
-    fn table(&self, catalog: &str, schema: &str, table_name: &str) -> Option<TableRef> {
+    fn table(
+        &self,
+        catalog: &str,
+        schema: &str,
+        table_name: &str,
+        query_ctx: Option<&QueryContext>,
+    ) -> Option<TableRef> {
+        let channel = query_ctx.map_or(Channel::Unknown, |ctx| ctx.channel());
         if schema == INFORMATION_SCHEMA_NAME {
             let information_schema_provider =
                 self.catalog_cache.get_with_by_ref(catalog, move || {
@@ -366,7 +432,7 @@ impl SystemCatalog {
                     ))
                 });
             information_schema_provider.table(table_name)
-        } else if schema == PG_CATALOG_NAME {
+        } else if schema == PG_CATALOG_NAME && channel == Channel::Postgres {
             if catalog == DEFAULT_CATALOG_NAME {
                 self.pg_catalog_provider.table(table_name)
             } else {

src/catalog/src/lib.rs

@@ -20,8 +20,10 @@ use std::fmt::{Debug, Formatter};
 use std::sync::Arc;
 
 use api::v1::CreateTableExpr;
+use common_catalog::consts::{INFORMATION_SCHEMA_NAME, PG_CATALOG_NAME};
 use futures::future::BoxFuture;
 use futures_util::stream::BoxStream;
+use session::context::QueryContext;
 use table::metadata::TableId;
 use table::TableRef;
 
@@ -44,15 +46,35 @@ pub trait CatalogManager: Send + Sync {
 
     async fn catalog_names(&self) -> Result<Vec<String>>;
 
-    async fn schema_names(&self, catalog: &str) -> Result<Vec<String>>;
+    async fn schema_names(
+        &self,
+        catalog: &str,
+        query_ctx: Option<&QueryContext>,
+    ) -> Result<Vec<String>>;
 
-    async fn table_names(&self, catalog: &str, schema: &str) -> Result<Vec<String>>;
+    async fn table_names(
+        &self,
+        catalog: &str,
+        schema: &str,
+        query_ctx: Option<&QueryContext>,
+    ) -> Result<Vec<String>>;
 
     async fn catalog_exists(&self, catalog: &str) -> Result<bool>;
 
-    async fn schema_exists(&self, catalog: &str, schema: &str) -> Result<bool>;
+    async fn schema_exists(
+        &self,
+        catalog: &str,
+        schema: &str,
+        query_ctx: Option<&QueryContext>,
+    ) -> Result<bool>;
 
-    async fn table_exists(&self, catalog: &str, schema: &str, table: &str) -> Result<bool>;
+    async fn table_exists(
+        &self,
+        catalog: &str,
+        schema: &str,
+        table: &str,
+        query_ctx: Option<&QueryContext>,
+    ) -> Result<bool>;
 
     /// Returns the table by catalog, schema and table name.
     async fn table(
@@ -60,10 +82,25 @@ pub trait CatalogManager: Send + Sync {
         catalog: &str,
         schema: &str,
         table_name: &str,
+        query_ctx: Option<&QueryContext>,
     ) -> Result<Option<TableRef>>;
 
     /// Returns all tables with a stream by catalog and schema.
-    fn tables<'a>(&'a self, catalog: &'a str, schema: &'a str) -> BoxStream<'a, Result<TableRef>>;
+    fn tables<'a>(
+        &'a self,
+        catalog: &'a str,
+        schema: &'a str,
+        query_ctx: Option<&'a QueryContext>,
+    ) -> BoxStream<'a, Result<TableRef>>;
+
+    /// Check if `schema` is a reserved schema name
+    fn is_reserved_schema_name(&self, schema: &str) -> bool {
+        // We have to check whether a schema name is reserved before create schema.
+        // We need this rather than use schema_exists directly because `pg_catalog` is
+        // only visible via postgres protocol. So if we don't check, a mysql client may
+        // create a schema named `pg_catalog` which is somehow malformed.
+        schema == INFORMATION_SCHEMA_NAME || schema == PG_CATALOG_NAME
+    }
 }
 
 pub type CatalogManagerRef = Arc<dyn CatalogManager>;

src/catalog/src/memory/manager.rs

@@ -26,6 +26,7 @@ use common_catalog::consts::{
 use common_meta::key::flow::FlowMetadataManager;
 use common_meta::kv_backend::memory::MemoryKvBackend;
 use futures_util::stream::BoxStream;
+use session::context::QueryContext;
 use snafu::OptionExt;
 use table::TableRef;
 
@@ -53,7 +54,11 @@ impl CatalogManager for MemoryCatalogManager {
         Ok(self.catalogs.read().unwrap().keys().cloned().collect())
     }
 
-    async fn schema_names(&self, catalog: &str) -> Result<Vec<String>> {
+    async fn schema_names(
+        &self,
+        catalog: &str,
+        _query_ctx: Option<&QueryContext>,
+    ) -> Result<Vec<String>> {
         Ok(self
             .catalogs
             .read()
@@ -67,7 +72,12 @@ impl CatalogManager for MemoryCatalogManager {
             .collect())
     }
 
-    async fn table_names(&self, catalog: &str, schema: &str) -> Result<Vec<String>> {
+    async fn table_names(
+        &self,
+        catalog: &str,
+        schema: &str,
+        _query_ctx: Option<&QueryContext>,
+    ) -> Result<Vec<String>> {
         Ok(self
             .catalogs
             .read()
@@ -87,11 +97,22 @@ impl CatalogManager for MemoryCatalogManager {
         self.catalog_exist_sync(catalog)
     }
 
-    async fn schema_exists(&self, catalog: &str, schema: &str) -> Result<bool> {
+    async fn schema_exists(
+        &self,
+        catalog: &str,
+        schema: &str,
+        _query_ctx: Option<&QueryContext>,
+    ) -> Result<bool> {
         self.schema_exist_sync(catalog, schema)
     }
 
-    async fn table_exists(&self, catalog: &str, schema: &str, table: &str) -> Result<bool> {
+    async fn table_exists(
+        &self,
+        catalog: &str,
+        schema: &str,
+        table: &str,
+        _query_ctx: Option<&QueryContext>,
+    ) -> Result<bool> {
         let catalogs = self.catalogs.read().unwrap();
         Ok(catalogs
             .get(catalog)
@@ -108,6 +129,7 @@ impl CatalogManager for MemoryCatalogManager {
         catalog: &str,
         schema: &str,
         table_name: &str,
+        _query_ctx: Option<&QueryContext>,
     ) -> Result<Option<TableRef>> {
         let result = try {
             self.catalogs
@@ -121,7 +143,12 @@ impl CatalogManager for MemoryCatalogManager {
         Ok(result)
     }
 
-    fn tables<'a>(&'a self, catalog: &'a str, schema: &'a str) -> BoxStream<'a, Result<TableRef>> {
+    fn tables<'a>(
+        &'a self,
+        catalog: &'a str,
+        schema: &'a str,
+        _query_ctx: Option<&QueryContext>,
+    ) -> BoxStream<'a, Result<TableRef>> {
         let catalogs = self.catalogs.read().unwrap();
 
         let Some(schemas) = catalogs.get(catalog) else {
@@ -371,11 +398,12 @@ mod tests {
                 DEFAULT_CATALOG_NAME,
                 DEFAULT_SCHEMA_NAME,
                 NUMBERS_TABLE_NAME,
+                None,
             )
             .await
             .unwrap()
             .unwrap();
-        let stream = catalog_list.tables(DEFAULT_CATALOG_NAME, DEFAULT_SCHEMA_NAME);
+        let stream = catalog_list.tables(DEFAULT_CATALOG_NAME, DEFAULT_SCHEMA_NAME, None);
         let tables = stream.try_collect::<Vec<_>>().await.unwrap();
         assert_eq!(tables.len(), 1);
         assert_eq!(
@@ -384,7 +412,12 @@ mod tests {
         );
 
         assert!(catalog_list
-            .table(DEFAULT_CATALOG_NAME, DEFAULT_SCHEMA_NAME, "not_exists")
+            .table(
+                DEFAULT_CATALOG_NAME,
+                DEFAULT_SCHEMA_NAME,
+                "not_exists",
+                None
+            )
             .await
             .unwrap()
             .is_none());
@@ -411,7 +444,7 @@ mod tests {
         };
         catalog.register_table_sync(register_table_req).unwrap();
         assert!(catalog
-            .table(DEFAULT_CATALOG_NAME, DEFAULT_SCHEMA_NAME, table_name)
+            .table(DEFAULT_CATALOG_NAME, DEFAULT_SCHEMA_NAME, table_name, None)
             .await
             .unwrap()
             .is_some());
@@ -423,7 +456,7 @@ mod tests {
         };
         catalog.deregister_table_sync(deregister_table_req).unwrap();
         assert!(catalog
-            .table(DEFAULT_CATALOG_NAME, DEFAULT_SCHEMA_NAME, table_name)
+            .table(DEFAULT_CATALOG_NAME, DEFAULT_SCHEMA_NAME, table_name, None)
             .await
             .unwrap()
             .is_none());

src/catalog/src/system_schema/information_schema/columns.rs

@@ -257,8 +257,8 @@ impl InformationSchemaColumnsBuilder {
             .context(UpgradeWeakCatalogManagerRefSnafu)?;
         let predicates = Predicates::from_scan_request(&request);
 
-        for schema_name in catalog_manager.schema_names(&catalog_name).await? {
-            let mut stream = catalog_manager.tables(&catalog_name, &schema_name);
+        for schema_name in catalog_manager.schema_names(&catalog_name, None).await? {
+            let mut stream = catalog_manager.tables(&catalog_name, &schema_name, None);
 
             while let Some(table) = stream.try_next().await? {
                 let keys = &table.table_info().meta.primary_key_indices;

src/catalog/src/system_schema/information_schema/key_column_usage.rs

@@ -212,8 +212,8 @@ impl InformationSchemaKeyColumnUsageBuilder {
             .context(UpgradeWeakCatalogManagerRefSnafu)?;
         let predicates = Predicates::from_scan_request(&request);
 
-        for schema_name in catalog_manager.schema_names(&catalog_name).await? {
-            let mut stream = catalog_manager.tables(&catalog_name, &schema_name);
+        for schema_name in catalog_manager.schema_names(&catalog_name, None).await? {
+            let mut stream = catalog_manager.tables(&catalog_name, &schema_name, None);
 
             while let Some(table) = stream.try_next().await? {
                 let mut primary_constraints = vec![];

src/catalog/src/system_schema/information_schema/partitions.rs

@@ -240,9 +240,9 @@ impl InformationSchemaPartitionsBuilder {
 
         let predicates = Predicates::from_scan_request(&request);
 
-        for schema_name in catalog_manager.schema_names(&catalog_name).await? {
+        for schema_name in catalog_manager.schema_names(&catalog_name, None).await? {
             let table_info_stream = catalog_manager
-                .tables(&catalog_name, &schema_name)
+                .tables(&catalog_name, &schema_name, None)
                 .try_filter_map(|t| async move {
                     let table_info = t.table_info();
                     if table_info.table_type == TableType::Temporary {

src/catalog/src/system_schema/information_schema/region_peers.rs

@@ -176,9 +176,9 @@ impl InformationSchemaRegionPeersBuilder {
 
         let predicates = Predicates::from_scan_request(&request);
 
-        for schema_name in catalog_manager.schema_names(&catalog_name).await? {
+        for schema_name in catalog_manager.schema_names(&catalog_name, None).await? {
             let table_id_stream = catalog_manager
-                .tables(&catalog_name, &schema_name)
+                .tables(&catalog_name, &schema_name, None)
                 .try_filter_map(|t| async move {
                     let table_info = t.table_info();
                     if table_info.table_type == TableType::Temporary {

src/catalog/src/system_schema/information_schema/schemata.rs

@@ -171,7 +171,7 @@ impl InformationSchemaSchemataBuilder {
         let table_metadata_manager = utils::table_meta_manager(&self.catalog_manager)?;
         let predicates = Predicates::from_scan_request(&request);
 
-        for schema_name in catalog_manager.schema_names(&catalog_name).await? {
+        for schema_name in catalog_manager.schema_names(&catalog_name, None).await? {
             let opts = if let Some(table_metadata_manager) = &table_metadata_manager {
                 table_metadata_manager
                     .schema_manager()

src/catalog/src/system_schema/information_schema/table_constraints.rs

@@ -176,8 +176,8 @@ impl InformationSchemaTableConstraintsBuilder {
             .context(UpgradeWeakCatalogManagerRefSnafu)?;
         let predicates = Predicates::from_scan_request(&request);
 
-        for schema_name in catalog_manager.schema_names(&catalog_name).await? {
-            let mut stream = catalog_manager.tables(&catalog_name, &schema_name);
+        for schema_name in catalog_manager.schema_names(&catalog_name, None).await? {
+            let mut stream = catalog_manager.tables(&catalog_name, &schema_name, None);
 
             while let Some(table) = stream.try_next().await? {
                 let keys = &table.table_info().meta.primary_key_indices;

src/catalog/src/system_schema/information_schema/tables.rs

@@ -234,8 +234,8 @@ impl InformationSchemaTablesBuilder {
             .context(UpgradeWeakCatalogManagerRefSnafu)?;
         let predicates = Predicates::from_scan_request(&request);
 
-        for schema_name in catalog_manager.schema_names(&catalog_name).await? {
-            let mut stream = catalog_manager.tables(&catalog_name, &schema_name);
+        for schema_name in catalog_manager.schema_names(&catalog_name, None).await? {
+            let mut stream = catalog_manager.tables(&catalog_name, &schema_name, None);
 
             while let Some(table) = stream.try_next().await? {
                 let table_info = table.table_info();

src/catalog/src/system_schema/information_schema/views.rs

@@ -192,8 +192,8 @@ impl InformationSchemaViewsBuilder {
             .context(CastManagerSnafu)?
             .view_info_cache()?;
 
-        for schema_name in catalog_manager.schema_names(&catalog_name).await? {
-            let mut stream = catalog_manager.tables(&catalog_name, &schema_name);
+        for schema_name in catalog_manager.schema_names(&catalog_name, None).await? {
+            let mut stream = catalog_manager.tables(&catalog_name, &schema_name, None);
 
             while let Some(table) = stream.try_next().await? {
                 let table_info = table.table_info();

src/catalog/src/system_schema/pg_catalog.rs

@@ -18,15 +18,16 @@ mod pg_namespace;
 mod table_names;
 
 use std::collections::HashMap;
-use std::sync::{Arc, Weak};
+use std::sync::{Arc, LazyLock, Weak};
 
-use common_catalog::consts::{self, PG_CATALOG_NAME};
+use common_catalog::consts::{self, DEFAULT_CATALOG_NAME, DEFAULT_SCHEMA_NAME, PG_CATALOG_NAME};
 use datatypes::schema::ColumnSchema;
 use lazy_static::lazy_static;
 use paste::paste;
 use pg_catalog_memory_table::get_schema_columns;
 use pg_class::PGClass;
 use pg_namespace::PGNamespace;
+use session::context::{Channel, QueryContext};
 use table::TableRef;
 pub use table_names::*;
 
@@ -142,3 +143,12 @@ impl SystemSchemaProviderInner for PGCatalogProvider {
         &self.catalog_name
     }
 }
+
+/// Provide query context to call the [`CatalogManager`]'s method.
+static PG_QUERY_CTX: LazyLock<QueryContext> = LazyLock::new(|| {
+    QueryContext::with_channel(DEFAULT_CATALOG_NAME, DEFAULT_SCHEMA_NAME, Channel::Postgres)
+});
+
+fn query_ctx() -> Option<&'static QueryContext> {
+    Some(&PG_QUERY_CTX)
+}

src/catalog/src/system_schema/pg_catalog/pg_class.rs

@@ -32,7 +32,7 @@ use store_api::storage::ScanRequest;
 use table::metadata::TableType;
 
 use super::pg_namespace::oid_map::PGNamespaceOidMapRef;
-use super::{OID_COLUMN_NAME, PG_CLASS};
+use super::{query_ctx, OID_COLUMN_NAME, PG_CLASS};
 use crate::error::{
     CreateRecordBatchSnafu, InternalSnafu, Result, UpgradeWeakCatalogManagerRefSnafu,
 };
@@ -202,8 +202,11 @@ impl PGClassBuilder {
             .upgrade()
             .context(UpgradeWeakCatalogManagerRefSnafu)?;
         let predicates = Predicates::from_scan_request(&request);
-        for schema_name in catalog_manager.schema_names(&catalog_name).await? {
-            let mut stream = catalog_manager.tables(&catalog_name, &schema_name);
+        for schema_name in catalog_manager
+            .schema_names(&catalog_name, query_ctx())
+            .await?
+        {
+            let mut stream = catalog_manager.tables(&catalog_name, &schema_name, query_ctx());
             while let Some(table) = stream.try_next().await? {
                 let table_info = table.table_info();
                 self.add_class(

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

@@ -31,7 +31,7 @@ use datatypes::vectors::{StringVectorBuilder, UInt32VectorBuilder, VectorRef};
 use snafu::{OptionExt, ResultExt};
 use store_api::storage::ScanRequest;
 
-use super::{PGNamespaceOidMapRef, OID_COLUMN_NAME, PG_NAMESPACE};
+use super::{query_ctx, PGNamespaceOidMapRef, OID_COLUMN_NAME, PG_NAMESPACE};
 use crate::error::{
     CreateRecordBatchSnafu, InternalSnafu, Result, UpgradeWeakCatalogManagerRefSnafu,
 };
@@ -180,7 +180,10 @@ impl PGNamespaceBuilder {
             .upgrade()
             .context(UpgradeWeakCatalogManagerRefSnafu)?;
         let predicates = Predicates::from_scan_request(&request);
-        for schema_name in catalog_manager.schema_names(&catalog_name).await? {
+        for schema_name in catalog_manager
+            .schema_names(&catalog_name, query_ctx())
+            .await?
+        {
             self.add_namespace(&predicates, &schema_name);
         }
         self.finish()

src/catalog/src/table_source.rs

@@ -23,7 +23,7 @@ use datafusion::datasource::view::ViewTable;
 use datafusion::datasource::{provider_as_source, TableProvider};
 use datafusion::logical_expr::TableSource;
 use itertools::Itertools;
-use session::context::QueryContext;
+use session::context::QueryContextRef;
 use snafu::{ensure, OptionExt, ResultExt};
 use table::metadata::TableType;
 use table::table::adapter::DfTableProviderAdapter;
@@ -45,6 +45,7 @@ pub struct DfTableSourceProvider {
     disallow_cross_catalog_query: bool,
     default_catalog: String,
     default_schema: String,
+    query_ctx: QueryContextRef,
     plan_decoder: SubstraitPlanDecoderRef,
     enable_ident_normalization: bool,
 }
@@ -53,7 +54,7 @@ impl DfTableSourceProvider {
     pub fn new(
         catalog_manager: CatalogManagerRef,
         disallow_cross_catalog_query: bool,
-        query_ctx: &QueryContext,
+        query_ctx: QueryContextRef,
         plan_decoder: SubstraitPlanDecoderRef,
         enable_ident_normalization: bool,
     ) -> Self {
@@ -63,6 +64,7 @@ impl DfTableSourceProvider {
             resolved_tables: HashMap::new(),
             default_catalog: query_ctx.current_catalog().to_owned(),
             default_schema: query_ctx.current_schema(),
+            query_ctx,
             plan_decoder,
             enable_ident_normalization,
         }
@@ -71,8 +73,7 @@ impl DfTableSourceProvider {
     pub fn resolve_table_ref(&self, table_ref: TableReference) -> Result<ResolvedTableReference> {
         if self.disallow_cross_catalog_query {
             match &table_ref {
-                TableReference::Bare { .. } => (),
-                TableReference::Partial { .. } => {}
+                TableReference::Bare { .. } | TableReference::Partial { .. } => {}
                 TableReference::Full {
                     catalog, schema, ..
                 } => {
@@ -107,7 +108,7 @@ impl DfTableSourceProvider {
 
         let table = self
             .catalog_manager
-            .table(catalog_name, schema_name, table_name)
+            .table(catalog_name, schema_name, table_name, Some(&self.query_ctx))
             .await?
             .with_context(|| TableNotExistSnafu {
                 table: format_full_table_name(catalog_name, schema_name, table_name),
@@ -210,12 +211,12 @@ mod tests {
 
     #[test]
     fn test_validate_table_ref() {
-        let query_ctx = &QueryContext::with("greptime", "public");
+        let query_ctx = Arc::new(QueryContext::with("greptime", "public"));
 
         let table_provider = DfTableSourceProvider::new(
             MemoryCatalogManager::with_default_setup(),
             true,
-            query_ctx,
+            query_ctx.clone(),
             DummyDecoder::arc(),
             true,
         );
@@ -308,7 +309,7 @@ mod tests {
 
     #[tokio::test]
     async fn test_resolve_view() {
-        let query_ctx = &QueryContext::with("greptime", "public");
+        let query_ctx = Arc::new(QueryContext::with("greptime", "public"));
         let backend = Arc::new(MemoryKvBackend::default());
         let layered_cache_builder = LayeredCacheRegistryBuilder::default()
             .add_cache_registry(CacheRegistryBuilder::default().build());
@@ -344,8 +345,13 @@ mod tests {
             .await
             .unwrap();
 
-        let mut table_provider =
-            DfTableSourceProvider::new(catalog_manager, true, query_ctx, MockDecoder::arc(), true);
+        let mut table_provider = DfTableSourceProvider::new(
+            catalog_manager,
+            true,
+            query_ctx.clone(),
+            MockDecoder::arc(),
+            true,
+        );
 
         // View not found
         let table_ref = TableReference::bare("not_exists_view");

src/catalog/src/table_source/dummy_catalog.rs

@@ -112,7 +112,7 @@ impl SchemaProvider for DummySchemaProvider {
     async fn table(&self, name: &str) -> datafusion::error::Result<Option<Arc<dyn TableProvider>>> {
         let table = self
             .catalog_manager
-            .table(&self.catalog_name, &self.schema_name, name)
+            .table(&self.catalog_name, &self.schema_name, name, None)
             .await?
             .with_context(|| TableNotExistSnafu {
                 table: format_full_table_name(&self.catalog_name, &self.schema_name, name),

src/client/src/database.rs

@@ -37,7 +37,8 @@ use tonic::metadata::AsciiMetadataKey;
 use tonic::transport::Channel;
 
 use crate::error::{
-    ConvertFlightDataSnafu, Error, IllegalFlightMessagesSnafu, InvalidAsciiSnafu, ServerSnafu,
+    ConvertFlightDataSnafu, Error, FlightGetSnafu, IllegalFlightMessagesSnafu, InvalidAsciiSnafu,
+    ServerSnafu,
 };
 use crate::{from_grpc_response, Client, Result};
 
@@ -91,7 +92,7 @@ impl Database {
     ///
     /// - the name of database when using GreptimeDB standalone or cluster
     /// - the name provided by GreptimeCloud or other multi-tenant GreptimeDB
-    /// environment
+    ///   environment
     pub fn new_with_dbname(dbname: impl Into<String>, client: Client) -> Self {
         Self {
             catalog: String::default(),
@@ -225,16 +226,18 @@ impl Database {
 
         let mut client = self.client.make_flight_client()?;
 
-        let response = client.mut_inner().do_get(request).await.map_err(|e| {
+        let response = client.mut_inner().do_get(request).await.or_else(|e| {
             let tonic_code = e.code();
             let e: Error = e.into();
             let code = e.status_code();
             let msg = e.to_string();
-            let error = Error::FlightGet {
-                tonic_code,
-                addr: client.addr().to_string(),
-                source: BoxedError::new(ServerSnafu { code, msg }.build()),
-            };
+            let error =
+                Err(BoxedError::new(ServerSnafu { code, msg }.build())).with_context(|_| {
+                    FlightGetSnafu {
+                        addr: client.addr().to_string(),
+                        tonic_code,
+                    }
+                });
             error!(
                 "Failed to do Flight get, addr: {}, code: {}, source: {:?}",
                 client.addr(),

src/client/src/error.rs

@@ -39,13 +39,6 @@ pub enum Error {
         source: BoxedError,
     },
 
-    #[snafu(display("Failure occurs during handling request"))]
-    HandleRequest {
-        #[snafu(implicit)]
-        location: Location,
-        source: BoxedError,
-    },
-
     #[snafu(display("Failed to convert FlightData"))]
     ConvertFlightData {
         #[snafu(implicit)]
@@ -116,13 +109,6 @@ pub enum Error {
         location: Location,
     },
 
-    #[snafu(display("Failed to send request with streaming: {}", err_msg))]
-    ClientStreaming {
-        err_msg: String,
-        #[snafu(implicit)]
-        location: Location,
-    },
-
     #[snafu(display("Failed to parse ascii string: {}", value))]
     InvalidAscii {
         value: String,
@@ -138,12 +124,10 @@ impl ErrorExt for Error {
         match self {
             Error::IllegalFlightMessages { .. }
             | Error::MissingField { .. }
-            | Error::IllegalDatabaseResponse { .. }
-            | Error::ClientStreaming { .. } => StatusCode::Internal,
+            | Error::IllegalDatabaseResponse { .. } => StatusCode::Internal,
 
             Error::Server { code, .. } => *code,
             Error::FlightGet { source, .. }
-            | Error::HandleRequest { source, .. }
             | Error::RegionServer { source, .. }
             | Error::FlowServer { source, .. } => source.status_code(),
             Error::CreateChannel { source, .. }

src/client/src/flow.rs

@@ -16,9 +16,9 @@ use api::v1::flow::{FlowRequest, FlowResponse};
 use api::v1::region::InsertRequests;
 use common_error::ext::BoxedError;
 use common_meta::node_manager::Flownode;
-use snafu::{location, ResultExt};
+use snafu::ResultExt;
 
-use crate::error::Result;
+use crate::error::{FlowServerSnafu, Result};
 use crate::Client;
 
 #[derive(Debug)]
@@ -57,15 +57,10 @@ impl FlowRequester {
         let response = client
             .handle_create_remove(request)
             .await
-            .map_err(|e| {
+            .or_else(|e| {
                 let code = e.code();
                 let err: crate::error::Error = e.into();
-                crate::error::Error::FlowServer {
-                    addr,
-                    code,
-                    source: BoxedError::new(err),
-                    location: location!(),
-                }
+                Err(BoxedError::new(err)).context(FlowServerSnafu { addr, code })
             })?
             .into_inner();
         Ok(response)
@@ -88,15 +83,10 @@ impl FlowRequester {
         let response = client
             .handle_mirror_request(requests)
             .await
-            .map_err(|e| {
+            .or_else(|e| {
                 let code = e.code();
                 let err: crate::error::Error = e.into();
-                crate::error::Error::FlowServer {
-                    addr,
-                    code,
-                    source: BoxedError::new(err),
-                    location: location!(),
-                }
+                Err(BoxedError::new(err)).context(FlowServerSnafu { addr, code })
             })?
             .into_inner();
         Ok(response)

src/client/src/region.rs

@@ -38,8 +38,8 @@ use substrait::{DFLogicalSubstraitConvertor, SubstraitPlan};
 use tokio_stream::StreamExt;
 
 use crate::error::{
-    self, ConvertFlightDataSnafu, IllegalDatabaseResponseSnafu, IllegalFlightMessagesSnafu,
-    MissingFieldSnafu, Result, ServerSnafu,
+    self, ConvertFlightDataSnafu, FlightGetSnafu, IllegalDatabaseResponseSnafu,
+    IllegalFlightMessagesSnafu, MissingFieldSnafu, Result, ServerSnafu,
 };
 use crate::{metrics, Client, Error};
 
@@ -103,11 +103,14 @@ impl RegionRequester {
                 let e: error::Error = e.into();
                 let code = e.status_code();
                 let msg = e.to_string();
-                let error = Error::FlightGet {
-                    tonic_code,
-                    addr: flight_client.addr().to_string(),
-                    source: BoxedError::new(ServerSnafu { code, msg }.build()),
-                };
+                let error = ServerSnafu { code, msg }
+                    .fail::<()>()
+                    .map_err(BoxedError::new)
+                    .with_context(|_| FlightGetSnafu {
+                        tonic_code,
+                        addr: flight_client.addr().to_string(),
+                    })
+                    .unwrap_err();
                 error!(
                     e; "Failed to do Flight get, addr: {}, code: {}",
                     flight_client.addr(),

src/cmd/Cargo.toml

@@ -51,7 +51,7 @@ file-engine.workspace = true
 flow.workspace = true
 frontend = { workspace = true, default-features = false }
 futures.workspace = true
-human-panic = "1.2.2"
+human-panic = "2.0"
 lazy_static.workspace = true
 meta-client.workspace = true
 meta-srv.workspace = true
@@ -70,6 +70,7 @@ serde.workspace = true
 serde_json.workspace = true
 servers.workspace = true
 session.workspace = true
+similar-asserts.workspace = true
 snafu.workspace = true
 store-api.workspace = true
 substrait.workspace = true
@@ -80,7 +81,7 @@ tonic.workspace = true
 tracing-appender = "0.2"
 
 [target.'cfg(not(windows))'.dependencies]
-tikv-jemallocator = "0.5"
+tikv-jemallocator = "0.6"
 
 [dev-dependencies]
 client = { workspace = true, features = ["testing"] }

src/cmd/src/bin/greptime.rs

@@ -139,13 +139,10 @@ async fn start(cli: Command) -> Result<()> {
 }
 
 fn setup_human_panic() {
-    let metadata = human_panic::Metadata {
-        version: env!("CARGO_PKG_VERSION").into(),
-        name: "GreptimeDB".into(),
-        authors: Default::default(),
-        homepage: "https://github.com/GreptimeTeam/greptimedb/discussions".into(),
-    };
-    human_panic::setup_panic!(metadata);
+    human_panic::setup_panic!(
+        human_panic::Metadata::new("GreptimeDB", env!("CARGO_PKG_VERSION"))
+            .homepage("https://github.com/GreptimeTeam/greptimedb/discussions")
+    );
 
     common_telemetry::set_panic_hook();
 }

src/cmd/src/cli.rs

@@ -21,6 +21,8 @@ mod export;
 mod helper;
 
 // Wait for https://github.com/GreptimeTeam/greptimedb/issues/2373
+mod database;
+mod import;
 #[allow(unused)]
 mod repl;
 
@@ -32,6 +34,7 @@ pub use repl::Repl;
 use tracing_appender::non_blocking::WorkerGuard;
 
 use self::export::ExportCommand;
+use crate::cli::import::ImportCommand;
 use crate::error::Result;
 use crate::options::GlobalOptions;
 use crate::App;
@@ -114,6 +117,7 @@ enum SubCommand {
     // Attach(AttachCommand),
     Bench(BenchTableMetadataCommand),
     Export(ExportCommand),
+    Import(ImportCommand),
 }
 
 impl SubCommand {
@@ -122,6 +126,7 @@ impl SubCommand {
             // SubCommand::Attach(cmd) => cmd.build().await,
             SubCommand::Bench(cmd) => cmd.build(guard).await,
             SubCommand::Export(cmd) => cmd.build(guard).await,
+            SubCommand::Import(cmd) => cmd.build(guard).await,
         }
     }
 }

src/cmd/src/cli/database.rs

@@ -0,0 +1,119 @@
+// 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 base64::engine::general_purpose;
+use base64::Engine;
+use common_catalog::consts::{DEFAULT_CATALOG_NAME, DEFAULT_SCHEMA_NAME};
+use serde_json::Value;
+use servers::http::greptime_result_v1::GreptimedbV1Response;
+use servers::http::GreptimeQueryOutput;
+use snafu::ResultExt;
+
+use crate::error::{HttpQuerySqlSnafu, Result, SerdeJsonSnafu};
+
+pub(crate) struct DatabaseClient {
+    addr: String,
+    catalog: String,
+    auth_header: Option<String>,
+}
+
+impl DatabaseClient {
+    pub fn new(addr: String, catalog: String, auth_basic: Option<String>) -> Self {
+        let auth_header = if let Some(basic) = auth_basic {
+            let encoded = general_purpose::STANDARD.encode(basic);
+            Some(format!("basic {}", encoded))
+        } else {
+            None
+        };
+
+        Self {
+            addr,
+            catalog,
+            auth_header,
+        }
+    }
+
+    pub async fn sql_in_public(&self, sql: &str) -> Result<Option<Vec<Vec<Value>>>> {
+        self.sql(sql, DEFAULT_SCHEMA_NAME).await
+    }
+
+    /// Execute sql query.
+    pub async fn sql(&self, sql: &str, schema: &str) -> Result<Option<Vec<Vec<Value>>>> {
+        let url = format!("http://{}/v1/sql", self.addr);
+        let params = [
+            ("db", format!("{}-{}", self.catalog, schema)),
+            ("sql", sql.to_string()),
+        ];
+        let mut request = reqwest::Client::new()
+            .post(&url)
+            .form(&params)
+            .header("Content-Type", "application/x-www-form-urlencoded");
+        if let Some(ref auth) = self.auth_header {
+            request = request.header("Authorization", auth);
+        }
+
+        let response = request.send().await.with_context(|_| HttpQuerySqlSnafu {
+            reason: format!("bad url: {}", url),
+        })?;
+        let response = response
+            .error_for_status()
+            .with_context(|_| HttpQuerySqlSnafu {
+                reason: format!("query failed: {}", sql),
+            })?;
+
+        let text = response.text().await.with_context(|_| HttpQuerySqlSnafu {
+            reason: "cannot get response text".to_string(),
+        })?;
+
+        let body = serde_json::from_str::<GreptimedbV1Response>(&text).context(SerdeJsonSnafu)?;
+        Ok(body.output().first().and_then(|output| match output {
+            GreptimeQueryOutput::Records(records) => Some(records.rows().clone()),
+            GreptimeQueryOutput::AffectedRows(_) => None,
+        }))
+    }
+}
+
+/// Split at `-`.
+pub(crate) fn split_database(database: &str) -> Result<(String, Option<String>)> {
+    let (catalog, schema) = match database.split_once('-') {
+        Some((catalog, schema)) => (catalog, schema),
+        None => (DEFAULT_CATALOG_NAME, database),
+    };
+
+    if schema == "*" {
+        Ok((catalog.to_string(), None))
+    } else {
+        Ok((catalog.to_string(), Some(schema.to_string())))
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use super::*;
+
+    #[test]
+    fn test_split_database() {
+        let result = split_database("catalog-schema").unwrap();
+        assert_eq!(result, ("catalog".to_string(), Some("schema".to_string())));
+
+        let result = split_database("schema").unwrap();
+        assert_eq!(result, ("greptime".to_string(), Some("schema".to_string())));
+
+        let result = split_database("catalog-*").unwrap();
+        assert_eq!(result, ("catalog".to_string(), None));
+
+        let result = split_database("*").unwrap();
+        assert_eq!(result, ("greptime".to_string(), None));
+    }
+}

src/cmd/src/cli/export.rs

@@ -13,30 +13,23 @@
 // limitations under the License.
 
 use std::collections::HashSet;
-use std::path::Path;
+use std::path::PathBuf;
 use std::sync::Arc;
 
 use async_trait::async_trait;
-use base64::engine::general_purpose;
-use base64::Engine;
 use clap::{Parser, ValueEnum};
-use client::DEFAULT_SCHEMA_NAME;
-use common_catalog::consts::DEFAULT_CATALOG_NAME;
 use common_telemetry::{debug, error, info};
 use serde_json::Value;
-use servers::http::greptime_result_v1::GreptimedbV1Response;
-use servers::http::GreptimeQueryOutput;
-use snafu::ResultExt;
+use snafu::{OptionExt, ResultExt};
 use tokio::fs::File;
 use tokio::io::{AsyncWriteExt, BufWriter};
 use tokio::sync::Semaphore;
 use tokio::time::Instant;
 use tracing_appender::non_blocking::WorkerGuard;
 
-use crate::cli::{Instance, Tool};
-use crate::error::{
-    EmptyResultSnafu, Error, FileIoSnafu, HttpQuerySqlSnafu, Result, SerdeJsonSnafu,
-};
+use crate::cli::database::DatabaseClient;
+use crate::cli::{database, Instance, Tool};
+use crate::error::{EmptyResultSnafu, Error, FileIoSnafu, Result, SchemaNotFoundSnafu};
 
 type TableReference = (String, String, String);
 
@@ -94,26 +87,21 @@ pub struct ExportCommand {
 
 impl ExportCommand {
     pub async fn build(&self, guard: Vec<WorkerGuard>) -> Result<Instance> {
-        let (catalog, schema) = split_database(&self.database)?;
+        let (catalog, schema) = database::split_database(&self.database)?;
 
-        let auth_header = if let Some(basic) = &self.auth_basic {
-            let encoded = general_purpose::STANDARD.encode(basic);
-            Some(format!("basic {}", encoded))
-        } else {
-            None
-        };
+        let database_client =
+            DatabaseClient::new(self.addr.clone(), catalog.clone(), self.auth_basic.clone());
 
         Ok(Instance::new(
             Box::new(Export {
-                addr: self.addr.clone(),
                 catalog,
                 schema,
+                database_client,
                 output_dir: self.output_dir.clone(),
                 parallelism: self.export_jobs,
                 target: self.target.clone(),
                 start_time: self.start_time.clone(),
                 end_time: self.end_time.clone(),
-                auth_header,
             }),
             guard,
         ))
@@ -121,78 +109,59 @@ impl ExportCommand {
 }
 
 pub struct Export {
-    addr: String,
     catalog: String,
     schema: Option<String>,
+    database_client: DatabaseClient,
     output_dir: String,
     parallelism: usize,
     target: ExportTarget,
     start_time: Option<String>,
     end_time: Option<String>,
-    auth_header: Option<String>,
 }
 
 impl Export {
-    /// Execute one single sql query.
-    async fn sql(&self, sql: &str) -> Result<Option<Vec<Vec<Value>>>> {
-        let url = format!(
-            "http://{}/v1/sql?db={}-{}&sql={}",
-            self.addr,
-            self.catalog,
-            self.schema.as_deref().unwrap_or(DEFAULT_SCHEMA_NAME),
-            sql
-        );
+    fn catalog_path(&self) -> PathBuf {
+        PathBuf::from(&self.output_dir).join(&self.catalog)
+    }
 
-        let mut request = reqwest::Client::new()
-            .get(&url)
-            .header("Content-Type", "application/x-www-form-urlencoded");
-        if let Some(ref auth) = self.auth_header {
-            request = request.header("Authorization", auth);
-        }
+    async fn get_db_names(&self) -> Result<Vec<String>> {
+        let db_names = self.all_db_names().await?;
+        let Some(schema) = &self.schema else {
+            return Ok(db_names);
+        };
 
-        let response = request.send().await.with_context(|_| HttpQuerySqlSnafu {
-            reason: format!("bad url: {}", url),
-        })?;
-        let response = response
-            .error_for_status()
-            .with_context(|_| HttpQuerySqlSnafu {
-                reason: format!("query failed: {}", sql),
-            })?;
-
-        let text = response.text().await.with_context(|_| HttpQuerySqlSnafu {
-            reason: "cannot get response text".to_string(),
-        })?;
-
-        let body = serde_json::from_str::<GreptimedbV1Response>(&text).context(SerdeJsonSnafu)?;
-        Ok(body.output().first().and_then(|output| match output {
-            GreptimeQueryOutput::Records(records) => Some(records.rows().clone()),
-            GreptimeQueryOutput::AffectedRows(_) => None,
-        }))
+        // Check if the schema exists
+        db_names
+            .into_iter()
+            .find(|db_name| db_name.to_lowercase() == schema.to_lowercase())
+            .map(|name| vec![name])
+            .context(SchemaNotFoundSnafu {
+                catalog: &self.catalog,
+                schema,
+            })
     }
 
     /// Iterate over all db names.
-    ///
-    /// Newbie: `db_name` is catalog + schema.
-    async fn iter_db_names(&self) -> Result<Vec<(String, String)>> {
-        if let Some(schema) = &self.schema {
-            Ok(vec![(self.catalog.clone(), schema.clone())])
-        } else {
-            let result = self.sql("SHOW DATABASES").await?;
-            let Some(records) = result else {
-                EmptyResultSnafu.fail()?
+    async fn all_db_names(&self) -> Result<Vec<String>> {
+        let records = self
+            .database_client
+            .sql_in_public("SHOW DATABASES")
+            .await?
+            .context(EmptyResultSnafu)?;
+        let mut result = Vec::with_capacity(records.len());
+        for value in records {
+            let Value::String(schema) = &value[0] else {
+                unreachable!()
             };
-            let mut result = Vec::with_capacity(records.len());
-            for value in records {
-                let Value::String(schema) = &value[0] else {
-                    unreachable!()
-                };
-                if schema == common_catalog::consts::INFORMATION_SCHEMA_NAME {
-                    continue;
-                }
-                result.push((self.catalog.clone(), schema.clone()));
+            if schema == common_catalog::consts::INFORMATION_SCHEMA_NAME {
+                continue;
             }
-            Ok(result)
+            if schema == common_catalog::consts::PG_CATALOG_NAME {
+                continue;
+            }
+            result.push(schema.clone());
         }
+        Ok(result)
     }
 
     /// Return a list of [`TableReference`] to be exported.
@@ -201,7 +170,11 @@ impl Export {
         &self,
         catalog: &str,
         schema: &str,
-    ) -> Result<(Vec<TableReference>, Vec<TableReference>)> {
+    ) -> Result<(
+        Vec<TableReference>,
+        Vec<TableReference>,
+        Vec<TableReference>,
+    )> {
         // Puts all metric table first
         let sql = format!(
             "SELECT table_catalog, table_schema, table_name \
@@ -210,15 +183,16 @@ impl Export {
                 and table_catalog = \'{catalog}\' \
                 and table_schema = \'{schema}\'"
         );
-        let result = self.sql(&sql).await?;
-        let Some(records) = result else {
-            EmptyResultSnafu.fail()?
-        };
+        let records = self
+            .database_client
+            .sql_in_public(&sql)
+            .await?
+            .context(EmptyResultSnafu)?;
         let mut metric_physical_tables = HashSet::with_capacity(records.len());
         for value in records {
             let mut t = Vec::with_capacity(3);
             for v in &value {
-                let serde_json::Value::String(value) = v else {
+                let Value::String(value) = v else {
                     unreachable!()
                 };
                 t.push(value);
@@ -226,100 +200,142 @@ impl Export {
             metric_physical_tables.insert((t[0].clone(), t[1].clone(), t[2].clone()));
         }
 
-        // TODO: SQL injection hurts
         let sql = format!(
-            "SELECT table_catalog, table_schema, table_name \
+            "SELECT table_catalog, table_schema, table_name, table_type \
             FROM information_schema.tables \
-            WHERE table_type = \'BASE TABLE\' \
+            WHERE (table_type = \'BASE TABLE\' OR table_type = \'VIEW\') \
                 and table_catalog = \'{catalog}\' \
                 and table_schema = \'{schema}\'",
         );
-        let result = self.sql(&sql).await?;
-        let Some(records) = result else {
-            EmptyResultSnafu.fail()?
-        };
+        let records = self
+            .database_client
+            .sql_in_public(&sql)
+            .await?
+            .context(EmptyResultSnafu)?;
 
-        debug!("Fetched table list: {:?}", records);
+        debug!("Fetched table/view list: {:?}", records);
 
         if records.is_empty() {
-            return Ok((vec![], vec![]));
+            return Ok((vec![], vec![], vec![]));
         }
 
         let mut remaining_tables = Vec::with_capacity(records.len());
+        let mut views = Vec::new();
         for value in records {
-            let mut t = Vec::with_capacity(3);
+            let mut t = Vec::with_capacity(4);
             for v in &value {
-                let serde_json::Value::String(value) = v else {
+                let Value::String(value) = v else {
                     unreachable!()
                 };
                 t.push(value);
             }
             let table = (t[0].clone(), t[1].clone(), t[2].clone());
+            let table_type = t[3].as_str();
             // Ignores the physical table
             if !metric_physical_tables.contains(&table) {
-                remaining_tables.push(table);
+                if table_type == "VIEW" {
+                    views.push(table);
+                } else {
+                    remaining_tables.push(table);
+                }
             }
         }
 
         Ok((
             metric_physical_tables.into_iter().collect(),
             remaining_tables,
+            views,
         ))
     }
 
-    async fn show_create_table(&self, catalog: &str, schema: &str, table: &str) -> Result<String> {
-        let sql = format!(
-            r#"SHOW CREATE TABLE "{}"."{}"."{}""#,
-            catalog, schema, table
-        );
-        let result = self.sql(&sql).await?;
-        let Some(records) = result else {
-            EmptyResultSnafu.fail()?
+    async fn show_create(
+        &self,
+        show_type: &str,
+        catalog: &str,
+        schema: &str,
+        table: Option<&str>,
+    ) -> Result<String> {
+        let sql = match table {
+            Some(table) => format!(
+                r#"SHOW CREATE {} "{}"."{}"."{}""#,
+                show_type, catalog, schema, table
+            ),
+            None => format!(r#"SHOW CREATE {} "{}"."{}""#, show_type, catalog, schema),
         };
-        let Value::String(create_table) = &records[0][1] else {
+        let records = self
+            .database_client
+            .sql_in_public(&sql)
+            .await?
+            .context(EmptyResultSnafu)?;
+        let Value::String(create) = &records[0][1] else {
             unreachable!()
         };
 
-        Ok(format!("{};\n", create_table))
+        Ok(format!("{};\n", create))
+    }
+
+    async fn export_create_database(&self) -> Result<()> {
+        let timer = Instant::now();
+        let db_names = self.get_db_names().await?;
+        let db_count = db_names.len();
+        for schema in db_names {
+            let db_dir = self.catalog_path().join(format!("{schema}/"));
+            tokio::fs::create_dir_all(&db_dir)
+                .await
+                .context(FileIoSnafu)?;
+            let file = db_dir.join("create_database.sql");
+            let mut file = File::create(file).await.context(FileIoSnafu)?;
+            let create_database = self
+                .show_create("DATABASE", &self.catalog, &schema, None)
+                .await?;
+            file.write_all(create_database.as_bytes())
+                .await
+                .context(FileIoSnafu)?;
+        }
+
+        let elapsed = timer.elapsed();
+        info!("Success {db_count} jobs, cost: {elapsed:?}");
+
+        Ok(())
     }
 
     async fn export_create_table(&self) -> Result<()> {
         let timer = Instant::now();
         let semaphore = Arc::new(Semaphore::new(self.parallelism));
-        let db_names = self.iter_db_names().await?;
+        let db_names = self.get_db_names().await?;
         let db_count = db_names.len();
         let mut tasks = Vec::with_capacity(db_names.len());
-        for (catalog, schema) in db_names {
+        for schema in db_names {
             let semaphore_moved = semaphore.clone();
             tasks.push(async move {
                 let _permit = semaphore_moved.acquire().await.unwrap();
-                let (metric_physical_tables, remaining_tables) =
-                    self.get_table_list(&catalog, &schema).await?;
-                let table_count = metric_physical_tables.len() + remaining_tables.len();
-                let output_dir = Path::new(&self.output_dir)
-                    .join(&catalog)
-                    .join(format!("{schema}/"));
-                tokio::fs::create_dir_all(&output_dir)
+                let (metric_physical_tables, remaining_tables, views) =
+                    self.get_table_list(&self.catalog, &schema).await?;
+                let table_count =
+                    metric_physical_tables.len() + remaining_tables.len() + views.len();
+                let db_dir = self.catalog_path().join(format!("{schema}/"));
+                tokio::fs::create_dir_all(&db_dir)
                     .await
                     .context(FileIoSnafu)?;
-                let output_file = Path::new(&output_dir).join("create_tables.sql");
-                let mut file = File::create(output_file).await.context(FileIoSnafu)?;
+                let file = db_dir.join("create_tables.sql");
+                let mut file = File::create(file).await.context(FileIoSnafu)?;
                 for (c, s, t) in metric_physical_tables.into_iter().chain(remaining_tables) {
-                    match self.show_create_table(&c, &s, &t).await {
-                        Err(e) => {
-                            error!(e; r#"Failed to export table "{}"."{}"."{}""#, c, s, t)
-                        }
-                        Ok(create_table) => {
-                            file.write_all(create_table.as_bytes())
-                                .await
-                                .context(FileIoSnafu)?;
-                        }
-                    }
+                    let create_table = self.show_create("TABLE", &c, &s, Some(&t)).await?;
+                    file.write_all(create_table.as_bytes())
+                        .await
+                        .context(FileIoSnafu)?;
+                }
+                for (c, s, v) in views {
+                    let create_view = self.show_create("VIEW", &c, &s, Some(&v)).await?;
+                    file.write_all(create_view.as_bytes())
+                        .await
+                        .context(FileIoSnafu)?;
                 }
 
                 info!(
-                    "Finished exporting {catalog}.{schema} with {table_count} table schemas to path: {}",
-                    output_dir.to_string_lossy()
+                    "Finished exporting {}.{schema} with {table_count} table schemas to path: {}",
+                    self.catalog,
+                    db_dir.to_string_lossy()
                 );
 
                 Ok::<(), Error>(())
@@ -332,14 +348,14 @@ impl Export {
             .filter(|r| match r {
                 Ok(_) => true,
                 Err(e) => {
-                    error!(e; "export job failed");
+                    error!(e; "export schema job failed");
                     false
                 }
             })
             .count();
 
         let elapsed = timer.elapsed();
-        info!("Success {success}/{db_count} jobs, cost: {:?}", elapsed);
+        info!("Success {success}/{db_count} jobs, cost: {elapsed:?}");
 
         Ok(())
     }
@@ -347,17 +363,15 @@ impl Export {
     async fn export_database_data(&self) -> Result<()> {
         let timer = Instant::now();
         let semaphore = Arc::new(Semaphore::new(self.parallelism));
-        let db_names = self.iter_db_names().await?;
+        let db_names = self.get_db_names().await?;
         let db_count = db_names.len();
-        let mut tasks = Vec::with_capacity(db_names.len());
-        for (catalog, schema) in db_names {
+        let mut tasks = Vec::with_capacity(db_count);
+        for schema in db_names {
             let semaphore_moved = semaphore.clone();
             tasks.push(async move {
                 let _permit = semaphore_moved.acquire().await.unwrap();
-                let output_dir = Path::new(&self.output_dir)
-                    .join(&catalog)
-                    .join(format!("{schema}/"));
-                tokio::fs::create_dir_all(&output_dir)
+                let db_dir = self.catalog_path().join(format!("{schema}/"));
+                tokio::fs::create_dir_all(&db_dir)
                     .await
                     .context(FileIoSnafu)?;
 
@@ -379,30 +393,31 @@ impl Export {
 
                 let sql = format!(
                     r#"COPY DATABASE "{}"."{}" TO '{}' {};"#,
-                    catalog,
+                    self.catalog,
                     schema,
-                    output_dir.to_str().unwrap(),
+                    db_dir.to_str().unwrap(),
                     with_options
                 );
 
                 info!("Executing sql: {sql}");
 
-                self.sql(&sql).await?;
+                self.database_client.sql_in_public(&sql).await?;
 
                 info!(
-                    "Finished exporting {catalog}.{schema} data into path: {}",
-                    output_dir.to_string_lossy()
+                    "Finished exporting {}.{schema} data into path: {}",
+                    self.catalog,
+                    db_dir.to_string_lossy()
                 );
 
                 // The export copy from sql
-                let copy_from_file = output_dir.join("copy_from.sql");
+                let copy_from_file = db_dir.join("copy_from.sql");
                 let mut writer =
                     BufWriter::new(File::create(copy_from_file).await.context(FileIoSnafu)?);
                 let copy_database_from_sql = format!(
                     r#"COPY DATABASE "{}"."{}" FROM '{}' WITH (FORMAT='parquet');"#,
-                    catalog,
+                    self.catalog,
                     schema,
-                    output_dir.to_str().unwrap()
+                    db_dir.to_str().unwrap()
                 );
                 writer
                     .write(copy_database_from_sql.as_bytes())
@@ -410,7 +425,7 @@ impl Export {
                     .context(FileIoSnafu)?;
                 writer.flush().await.context(FileIoSnafu)?;
 
-                info!("Finished exporting {catalog}.{schema} copy_from.sql");
+                info!("Finished exporting {}.{schema} copy_from.sql", self.catalog);
 
                 Ok::<(), Error>(())
             })
@@ -429,20 +444,23 @@ impl Export {
             .count();
         let elapsed = timer.elapsed();
 
-        info!("Success {success}/{db_count} jobs, costs: {:?}", elapsed);
+        info!("Success {success}/{db_count} jobs, costs: {elapsed:?}");
 
         Ok(())
     }
 }
 
-#[allow(deprecated)]
 #[async_trait]
 impl Tool for Export {
     async fn do_work(&self) -> Result<()> {
         match self.target {
-            ExportTarget::Schema => self.export_create_table().await,
+            ExportTarget::Schema => {
+                self.export_create_database().await?;
+                self.export_create_table().await
+            }
             ExportTarget::Data => self.export_database_data().await,
             ExportTarget::All => {
+                self.export_create_database().await?;
                 self.export_create_table().await?;
                 self.export_database_data().await
             }
@@ -450,20 +468,6 @@ impl Tool for Export {
     }
 }
 
-/// Split at `-`.
-fn split_database(database: &str) -> Result<(String, Option<String>)> {
-    let (catalog, schema) = match database.split_once('-') {
-        Some((catalog, schema)) => (catalog, schema),
-        None => (DEFAULT_CATALOG_NAME, database),
-    };
-
-    if schema == "*" {
-        Ok((catalog.to_string(), None))
-    } else {
-        Ok((catalog.to_string(), Some(schema.to_string())))
-    }
-}
-
 #[cfg(test)]
 mod tests {
     use clap::Parser;
@@ -471,26 +475,10 @@ mod tests {
     use common_catalog::consts::{DEFAULT_CATALOG_NAME, DEFAULT_SCHEMA_NAME};
     use common_telemetry::logging::LoggingOptions;
 
-    use crate::cli::export::split_database;
     use crate::error::Result as CmdResult;
     use crate::options::GlobalOptions;
     use crate::{cli, standalone, App};
 
-    #[test]
-    fn test_split_database() {
-        let result = split_database("catalog-schema").unwrap();
-        assert_eq!(result, ("catalog".to_string(), Some("schema".to_string())));
-
-        let result = split_database("schema").unwrap();
-        assert_eq!(result, ("greptime".to_string(), Some("schema".to_string())));
-
-        let result = split_database("catalog-*").unwrap();
-        assert_eq!(result, ("catalog".to_string(), None));
-
-        let result = split_database("*").unwrap();
-        assert_eq!(result, ("greptime".to_string(), None));
-    }
-
     #[tokio::test(flavor = "multi_thread")]
     async fn test_export_create_table_with_quoted_names() -> CmdResult<()> {
         let output_dir = tempfile::tempdir().unwrap();

src/cmd/src/cli/import.rs

@@ -0,0 +1,218 @@
+// 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::path::PathBuf;
+use std::sync::Arc;
+
+use async_trait::async_trait;
+use clap::{Parser, ValueEnum};
+use common_catalog::consts::DEFAULT_SCHEMA_NAME;
+use common_telemetry::{error, info, warn};
+use snafu::{OptionExt, ResultExt};
+use tokio::sync::Semaphore;
+use tokio::time::Instant;
+use tracing_appender::non_blocking::WorkerGuard;
+
+use crate::cli::database::DatabaseClient;
+use crate::cli::{database, Instance, Tool};
+use crate::error::{Error, FileIoSnafu, Result, SchemaNotFoundSnafu};
+
+#[derive(Debug, Default, Clone, ValueEnum)]
+enum ImportTarget {
+    /// Import all table schemas into the database.
+    Schema,
+    /// Import all table data into the database.
+    Data,
+    /// Export all table schemas and data at once.
+    #[default]
+    All,
+}
+
+#[derive(Debug, Default, Parser)]
+pub struct ImportCommand {
+    /// Server address to connect
+    #[clap(long)]
+    addr: String,
+
+    /// Directory of the data. E.g.: /tmp/greptimedb-backup
+    #[clap(long)]
+    input_dir: String,
+
+    /// The name of the catalog to import.
+    #[clap(long, default_value = "greptime-*")]
+    database: String,
+
+    /// Parallelism of the import.
+    #[clap(long, short = 'j', default_value = "1")]
+    import_jobs: usize,
+
+    /// Max retry times for each job.
+    #[clap(long, default_value = "3")]
+    max_retry: usize,
+
+    /// Things to export
+    #[clap(long, short = 't', value_enum, default_value = "all")]
+    target: ImportTarget,
+
+    /// The basic authentication for connecting to the server
+    #[clap(long)]
+    auth_basic: Option<String>,
+}
+
+impl ImportCommand {
+    pub async fn build(&self, guard: Vec<WorkerGuard>) -> Result<Instance> {
+        let (catalog, schema) = database::split_database(&self.database)?;
+        let database_client =
+            DatabaseClient::new(self.addr.clone(), catalog.clone(), self.auth_basic.clone());
+
+        Ok(Instance::new(
+            Box::new(Import {
+                catalog,
+                schema,
+                database_client,
+                input_dir: self.input_dir.clone(),
+                parallelism: self.import_jobs,
+                target: self.target.clone(),
+            }),
+            guard,
+        ))
+    }
+}
+
+pub struct Import {
+    catalog: String,
+    schema: Option<String>,
+    database_client: DatabaseClient,
+    input_dir: String,
+    parallelism: usize,
+    target: ImportTarget,
+}
+
+impl Import {
+    async fn import_create_table(&self) -> Result<()> {
+        // Use default db to creates other dbs
+        self.do_sql_job("create_database.sql", Some(DEFAULT_SCHEMA_NAME))
+            .await?;
+        self.do_sql_job("create_tables.sql", None).await
+    }
+
+    async fn import_database_data(&self) -> Result<()> {
+        self.do_sql_job("copy_from.sql", None).await
+    }
+
+    async fn do_sql_job(&self, filename: &str, exec_db: Option<&str>) -> Result<()> {
+        let timer = Instant::now();
+        let semaphore = Arc::new(Semaphore::new(self.parallelism));
+        let db_names = self.get_db_names().await?;
+        let db_count = db_names.len();
+        let mut tasks = Vec::with_capacity(db_count);
+        for schema in db_names {
+            let semaphore_moved = semaphore.clone();
+            tasks.push(async move {
+                let _permit = semaphore_moved.acquire().await.unwrap();
+                let database_input_dir = self.catalog_path().join(&schema);
+                let sql_file = database_input_dir.join(filename);
+                let sql = tokio::fs::read_to_string(sql_file)
+                    .await
+                    .context(FileIoSnafu)?;
+                if sql.is_empty() {
+                    info!("Empty `{filename}` {database_input_dir:?}");
+                } else {
+                    let db = exec_db.unwrap_or(&schema);
+                    self.database_client.sql(&sql, db).await?;
+                    info!("Imported `{filename}` for database {schema}");
+                }
+
+                Ok::<(), Error>(())
+            })
+        }
+
+        let success = futures::future::join_all(tasks)
+            .await
+            .into_iter()
+            .filter(|r| match r {
+                Ok(_) => true,
+                Err(e) => {
+                    error!(e; "import {filename} job failed");
+                    false
+                }
+            })
+            .count();
+        let elapsed = timer.elapsed();
+        info!("Success {success}/{db_count} `{filename}` jobs, cost: {elapsed:?}");
+
+        Ok(())
+    }
+
+    fn catalog_path(&self) -> PathBuf {
+        PathBuf::from(&self.input_dir).join(&self.catalog)
+    }
+
+    async fn get_db_names(&self) -> Result<Vec<String>> {
+        let db_names = self.all_db_names().await?;
+        let Some(schema) = &self.schema else {
+            return Ok(db_names);
+        };
+
+        // Check if the schema exists
+        db_names
+            .into_iter()
+            .find(|db_name| db_name.to_lowercase() == schema.to_lowercase())
+            .map(|name| vec![name])
+            .context(SchemaNotFoundSnafu {
+                catalog: &self.catalog,
+                schema,
+            })
+    }
+
+    // Get all database names in the input directory.
+    // The directory structure should be like:
+    // /tmp/greptimedb-backup
+    // ├── greptime-1
+    // │   ├── db1
+    // │   └── db2
+    async fn all_db_names(&self) -> Result<Vec<String>> {
+        let mut db_names = vec![];
+        let path = self.catalog_path();
+        let mut entries = tokio::fs::read_dir(path).await.context(FileIoSnafu)?;
+        while let Some(entry) = entries.next_entry().await.context(FileIoSnafu)? {
+            let path = entry.path();
+            if path.is_dir() {
+                let db_name = match path.file_name() {
+                    Some(name) => name.to_string_lossy().to_string(),
+                    None => {
+                        warn!("Failed to get the file name of {:?}", path);
+                        continue;
+                    }
+                };
+                db_names.push(db_name);
+            }
+        }
+        Ok(db_names)
+    }
+}
+
+#[async_trait]
+impl Tool for Import {
+    async fn do_work(&self) -> Result<()> {
+        match self.target {
+            ImportTarget::Schema => self.import_create_table().await,
+            ImportTarget::Data => self.import_database_data().await,
+            ImportTarget::All => {
+                self.import_create_table().await?;
+                self.import_database_data().await
+            }
+        }
+    }
+}

src/cmd/src/datanode.rs

@@ -267,7 +267,7 @@ impl StartCommand {
             &opts.component.tracing,
             opts.component.node_id.map(|x| x.to_string()),
         );
-        log_versions(version(), short_version());
+        log_versions(version(), short_version(), APP_NAME);
 
         info!("Datanode start command: {:#?}", self);
         info!("Datanode options: {:#?}", opts);

src/cmd/src/error.rs

@@ -31,13 +31,6 @@ pub enum Error {
         source: common_meta::error::Error,
     },
 
-    #[snafu(display("Failed to iter stream"))]
-    IterStream {
-        #[snafu(implicit)]
-        location: Location,
-        source: common_meta::error::Error,
-    },
-
     #[snafu(display("Failed to init DDL manager"))]
     InitDdlManager {
         #[snafu(implicit)]
@@ -237,13 +230,6 @@ pub enum Error {
         location: Location,
     },
 
-    #[snafu(display("Failed to start catalog manager"))]
-    StartCatalogManager {
-        #[snafu(implicit)]
-        location: Location,
-        source: catalog::error::Error,
-    },
-
     #[snafu(display("Failed to connect to Etcd at {etcd_addr}"))]
     ConnectEtcd {
         etcd_addr: String,
@@ -253,14 +239,6 @@ pub enum Error {
         location: Location,
     },
 
-    #[snafu(display("Failed to connect server at {addr}"))]
-    ConnectServer {
-        addr: String,
-        source: client::error::Error,
-        #[snafu(implicit)]
-        location: Location,
-    },
-
     #[snafu(display("Failed to serde json"))]
     SerdeJson {
         #[snafu(source)]
@@ -278,12 +256,6 @@ pub enum Error {
         location: Location,
     },
 
-    #[snafu(display("Expect data from output, but got another thing"))]
-    NotDataFromOutput {
-        #[snafu(implicit)]
-        location: Location,
-    },
-
     #[snafu(display("Empty result from output"))]
     EmptyResult {
         #[snafu(implicit)]
@@ -346,13 +318,12 @@ pub enum Error {
         source: meta_client::error::Error,
     },
 
-    #[snafu(display("Tonic transport error: {error:?} with msg: {msg:?}"))]
-    TonicTransport {
+    #[snafu(display("Cannot find schema {schema} in catalog {catalog}"))]
+    SchemaNotFound {
+        catalog: String,
+        schema: String,
         #[snafu(implicit)]
         location: Location,
-        #[snafu(source)]
-        error: tonic::transport::Error,
-        msg: Option<String>,
     },
 }
 
@@ -370,18 +341,16 @@ impl ErrorExt for Error {
             Error::BuildMetaServer { source, .. } => source.status_code(),
             Error::UnsupportedSelectorType { source, .. } => source.status_code(),
 
-            Error::IterStream { source, .. }
-            | Error::InitMetadata { source, .. }
-            | Error::InitDdlManager { source, .. } => source.status_code(),
+            Error::InitMetadata { source, .. } | Error::InitDdlManager { source, .. } => {
+                source.status_code()
+            }
 
-            Error::ConnectServer { source, .. } => source.status_code(),
             Error::MissingConfig { .. }
             | Error::LoadLayeredConfig { .. }
             | Error::IllegalConfig { .. }
             | Error::InvalidReplCommand { .. }
             | Error::InitTimezone { .. }
             | Error::ConnectEtcd { .. }
-            | Error::NotDataFromOutput { .. }
             | Error::CreateDir { .. }
             | Error::EmptyResult { .. } => StatusCode::InvalidArguments,
 
@@ -399,7 +368,6 @@ impl ErrorExt for Error {
                 source.status_code()
             }
             Error::SubstraitEncodeLogicalPlan { source, .. } => source.status_code(),
-            Error::StartCatalogManager { source, .. } => source.status_code(),
 
             Error::SerdeJson { .. } | Error::FileIo { .. } | Error::SpawnThread { .. } => {
                 StatusCode::Unexpected
@@ -414,7 +382,7 @@ impl ErrorExt for Error {
                 source.status_code()
             }
             Error::MetaClientInit { source, .. } => source.status_code(),
-            Error::TonicTransport { .. } => StatusCode::Internal,
+            Error::SchemaNotFound { .. } => StatusCode::DatabaseNotFound,
         }
     }
 

src/cmd/src/flownode.rs

@@ -215,7 +215,7 @@ impl StartCommand {
             &opts.component.tracing,
             opts.component.node_id.map(|x| x.to_string()),
         );
-        log_versions(version(), short_version());
+        log_versions(version(), short_version(), APP_NAME);
 
         info!("Flownode start command: {:#?}", self);
         info!("Flownode options: {:#?}", opts);

src/cmd/src/frontend.rs

@@ -261,7 +261,7 @@ impl StartCommand {
             &opts.component.tracing,
             opts.component.node_id.clone(),
         );
-        log_versions(version(), short_version());
+        log_versions(version(), short_version(), APP_NAME);
 
         info!("Frontend start command: {:#?}", self);
         info!("Frontend options: {:#?}", opts);

src/cmd/src/lib.rs

@@ -30,7 +30,7 @@ pub mod standalone;
 
 lazy_static::lazy_static! {
     static ref APP_VERSION: prometheus::IntGaugeVec =
-        prometheus::register_int_gauge_vec!("greptime_app_version", "app version", &["version", "short_version"]).unwrap();
+        prometheus::register_int_gauge_vec!("greptime_app_version", "app version", &["version", "short_version", "app"]).unwrap();
 }
 
 #[async_trait]
@@ -76,10 +76,10 @@ pub trait App: Send {
 /// Log the versions of the application, and the arguments passed to the cli.
 /// `version` should be the same as the output of cli "--version";
 /// and the `short_version` is the short version of the codes, often consist of git branch and commit.
-pub fn log_versions(version: &str, short_version: &str) {
+pub fn log_versions(version: &str, short_version: &str, app: &str) {
     // Report app version as gauge.
     APP_VERSION
-        .with_label_values(&[env!("CARGO_PKG_VERSION"), short_version])
+        .with_label_values(&[env!("CARGO_PKG_VERSION"), short_version, app])
         .inc();
 
     // Log version and argument flags.

src/cmd/src/metasrv.rs

@@ -244,7 +244,7 @@ impl StartCommand {
             &opts.component.tracing,
             None,
         );
-        log_versions(version(), short_version());
+        log_versions(version(), short_version(), APP_NAME);
 
         info!("Metasrv start command: {:#?}", self);
         info!("Metasrv options: {:#?}", opts);

src/cmd/src/standalone.rs

@@ -141,6 +141,8 @@ pub struct StandaloneOptions {
     pub region_engine: Vec<RegionEngineConfig>,
     pub export_metrics: ExportMetricsOption,
     pub tracing: TracingOptions,
+    pub init_regions_in_background: bool,
+    pub init_regions_parallelism: usize,
 }
 
 impl Default for StandaloneOptions {
@@ -168,6 +170,8 @@ impl Default for StandaloneOptions {
                 RegionEngineConfig::File(FileEngineConfig::default()),
             ],
             tracing: TracingOptions::default(),
+            init_regions_in_background: false,
+            init_regions_parallelism: 16,
         }
     }
 }
@@ -178,6 +182,16 @@ impl Configurable for StandaloneOptions {
     }
 }
 
+/// The [`StandaloneOptions`] is only defined in cmd crate,
+/// we don't want to make `frontend` depends on it, so impl [`Into`]
+/// rather than [`From`].
+#[allow(clippy::from_over_into)]
+impl Into<FrontendOptions> for StandaloneOptions {
+    fn into(self) -> FrontendOptions {
+        self.frontend_options()
+    }
+}
+
 impl StandaloneOptions {
     pub fn frontend_options(&self) -> FrontendOptions {
         let cloned_opts = self.clone();
@@ -208,6 +222,9 @@ impl StandaloneOptions {
             storage: cloned_opts.storage,
             region_engine: cloned_opts.region_engine,
             grpc: cloned_opts.grpc,
+            init_regions_in_background: cloned_opts.init_regions_in_background,
+            init_regions_parallelism: cloned_opts.init_regions_parallelism,
+            mode: Mode::Standalone,
             ..Default::default()
         }
     }
@@ -415,7 +432,7 @@ impl StartCommand {
             &opts.component.tracing,
             None,
         );
-        log_versions(version(), short_version());
+        log_versions(version(), short_version(), APP_NAME);
 
         info!("Standalone start command: {:#?}", self);
         info!("Standalone options: {opts:#?}");
@@ -510,7 +527,7 @@ impl StartCommand {
                 .build(),
         );
         let wal_options_allocator = Arc::new(WalOptionsAllocator::new(
-            opts.wal.into(),
+            opts.wal.clone().into(),
             kv_backend.clone(),
         ));
         let table_meta_allocator = Arc::new(TableMetadataAllocator::new(
@@ -533,7 +550,7 @@ impl StartCommand {
         .await?;
 
         let mut frontend = FrontendBuilder::new(
-            fe_opts.clone(),
+            fe_opts,
             kv_backend.clone(),
             layered_cache_registry.clone(),
             catalog_manager.clone(),
@@ -561,7 +578,7 @@ impl StartCommand {
 
         let (tx, _rx) = broadcast::channel(1);
 
-        let servers = Services::new(fe_opts, Arc::new(frontend.clone()), plugins)
+        let servers = Services::new(opts, Arc::new(frontend.clone()), plugins)
             .build()
             .await
             .context(StartFrontendSnafu)?;

src/cmd/tests/load_config_test.rs

@@ -16,12 +16,10 @@ use std::time::Duration;
 
 use cmd::options::GreptimeOptions;
 use cmd::standalone::StandaloneOptions;
-use common_base::readable_size::ReadableSize;
 use common_config::Configurable;
 use common_grpc::channel_manager::{
     DEFAULT_MAX_GRPC_RECV_MESSAGE_SIZE, DEFAULT_MAX_GRPC_SEND_MESSAGE_SIZE,
 };
-use common_runtime::global::RuntimeOptions;
 use common_telemetry::logging::{LoggingOptions, DEFAULT_OTLP_ENDPOINT};
 use common_wal::config::raft_engine::RaftEngineConfig;
 use common_wal::config::DatanodeWalConfig;
@@ -45,10 +43,6 @@ fn test_load_datanode_example_config() {
             .unwrap();
 
     let expected = GreptimeOptions::<DatanodeOptions> {
-        runtime: RuntimeOptions {
-            global_rt_size: 8,
-            compact_rt_size: 4,
-        },
         component: DatanodeOptions {
             node_id: Some(42),
             meta_client: Some(MetaClientOptions {
@@ -65,6 +59,7 @@ fn test_load_datanode_example_config() {
             wal: DatanodeWalConfig::RaftEngine(RaftEngineConfig {
                 dir: Some("/tmp/greptimedb/wal".to_string()),
                 sync_period: Some(Duration::from_secs(10)),
+                recovery_parallelism: 2,
                 ..Default::default()
             }),
             storage: StorageConfig {
@@ -73,15 +68,9 @@ fn test_load_datanode_example_config() {
             },
             region_engine: vec![
                 RegionEngineConfig::Mito(MitoConfig {
-                    num_workers: 8,
                     auto_flush_interval: Duration::from_secs(3600),
                     scan_parallelism: 0,
-                    global_write_buffer_size: ReadableSize::gb(1),
-                    global_write_buffer_reject_size: ReadableSize::gb(2),
-                    sst_meta_cache_size: ReadableSize::mb(128),
-                    vector_cache_size: ReadableSize::mb(512),
-                    page_cache_size: ReadableSize::mb(512),
-                    max_background_jobs: 4,
+                    experimental_write_cache_ttl: Some(Duration::from_secs(60 * 60 * 8)),
                     ..Default::default()
                 }),
                 RegionEngineConfig::File(EngineConfig {}),
@@ -105,9 +94,10 @@ fn test_load_datanode_example_config() {
             rpc_max_send_message_size: Some(DEFAULT_MAX_GRPC_SEND_MESSAGE_SIZE),
             ..Default::default()
         },
+        ..Default::default()
     };
 
-    assert_eq!(options, expected);
+    similar_asserts::assert_eq!(options, expected);
 }
 
 #[test]
@@ -117,10 +107,6 @@ fn test_load_frontend_example_config() {
         GreptimeOptions::<FrontendOptions>::load_layered_options(example_config.to_str(), "")
             .unwrap();
     let expected = GreptimeOptions::<FrontendOptions> {
-        runtime: RuntimeOptions {
-            global_rt_size: 8,
-            compact_rt_size: 4,
-        },
         component: FrontendOptions {
             default_timezone: Some("UTC".to_string()),
             meta_client: Some(MetaClientOptions {
@@ -153,8 +139,9 @@ fn test_load_frontend_example_config() {
             },
             ..Default::default()
         },
+        ..Default::default()
     };
-    assert_eq!(options, expected);
+    similar_asserts::assert_eq!(options, expected);
 }
 
 #[test]
@@ -164,10 +151,6 @@ fn test_load_metasrv_example_config() {
         GreptimeOptions::<MetasrvOptions>::load_layered_options(example_config.to_str(), "")
             .unwrap();
     let expected = GreptimeOptions::<MetasrvOptions> {
-        runtime: RuntimeOptions {
-            global_rt_size: 8,
-            compact_rt_size: 4,
-        },
         component: MetasrvOptions {
             selector: SelectorType::default(),
             data_home: "/tmp/metasrv/".to_string(),
@@ -185,8 +168,9 @@ fn test_load_metasrv_example_config() {
             },
             ..Default::default()
         },
+        ..Default::default()
     };
-    assert_eq!(options, expected);
+    similar_asserts::assert_eq!(options, expected);
 }
 
 #[test]
@@ -196,28 +180,19 @@ fn test_load_standalone_example_config() {
         GreptimeOptions::<StandaloneOptions>::load_layered_options(example_config.to_str(), "")
             .unwrap();
     let expected = GreptimeOptions::<StandaloneOptions> {
-        runtime: RuntimeOptions {
-            global_rt_size: 8,
-            compact_rt_size: 4,
-        },
         component: StandaloneOptions {
             default_timezone: Some("UTC".to_string()),
             wal: DatanodeWalConfig::RaftEngine(RaftEngineConfig {
                 dir: Some("/tmp/greptimedb/wal".to_string()),
                 sync_period: Some(Duration::from_secs(10)),
+                recovery_parallelism: 2,
                 ..Default::default()
             }),
             region_engine: vec![
                 RegionEngineConfig::Mito(MitoConfig {
-                    num_workers: 8,
                     auto_flush_interval: Duration::from_secs(3600),
+                    experimental_write_cache_ttl: Some(Duration::from_secs(60 * 60 * 8)),
                     scan_parallelism: 0,
-                    global_write_buffer_size: ReadableSize::gb(1),
-                    global_write_buffer_reject_size: ReadableSize::gb(2),
-                    sst_meta_cache_size: ReadableSize::mb(128),
-                    vector_cache_size: ReadableSize::mb(512),
-                    page_cache_size: ReadableSize::mb(512),
-                    max_background_jobs: 4,
                     ..Default::default()
                 }),
                 RegionEngineConfig::File(EngineConfig {}),
@@ -239,6 +214,7 @@ fn test_load_standalone_example_config() {
             },
             ..Default::default()
         },
+        ..Default::default()
     };
-    assert_eq!(options, expected);
+    similar_asserts::assert_eq!(options, expected);
 }

src/common/base/Cargo.toml

@@ -9,10 +9,12 @@ workspace = true
 
 [dependencies]
 anymap = "1.0.0-beta.2"
+async-trait.workspace = true
 bitvec = "1.0"
 bytes.workspace = true
 common-error.workspace = true
 common-macro.workspace = true
+futures.workspace = true
 paste = "1.0"
 serde = { version = "1.0", features = ["derive"] }
 snafu.workspace = true

src/common/base/src/buffer.rs

@@ -1,242 +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::any::Any;
-use std::io::{Read, Write};
-
-use bytes::{Buf, BufMut, BytesMut};
-use common_error::ext::ErrorExt;
-use common_macro::stack_trace_debug;
-use paste::paste;
-use snafu::{ensure, Location, ResultExt, Snafu};
-
-#[derive(Snafu)]
-#[snafu(visibility(pub))]
-#[stack_trace_debug]
-pub enum Error {
-    #[snafu(display(
-        "Destination buffer overflow, src_len: {}, dst_len: {}",
-        src_len,
-        dst_len
-    ))]
-    Overflow {
-        src_len: usize,
-        dst_len: usize,
-        #[snafu(implicit)]
-        location: Location,
-    },
-
-    #[snafu(display("Buffer underflow"))]
-    Underflow {
-        #[snafu(implicit)]
-        location: Location,
-    },
-
-    #[snafu(display("IO operation reach EOF"))]
-    Eof {
-        #[snafu(source)]
-        error: std::io::Error,
-        #[snafu(implicit)]
-        location: Location,
-    },
-}
-
-pub type Result<T> = std::result::Result<T, Error>;
-
-impl ErrorExt for Error {
-    fn as_any(&self) -> &dyn Any {
-        self
-    }
-}
-
-macro_rules! impl_read_le {
-    ( $($num_ty: ty), *) => {
-        $(
-            paste!{
-                // TODO(hl): default implementation requires allocating a
-                // temp buffer. maybe use more efficient impls in concrete buffers.
-                // see https://github.com/GrepTimeTeam/greptimedb/pull/97#discussion_r930798941
-                fn [<read_ $num_ty _le>](&mut self) -> Result<$num_ty> {
-                    let mut buf = [0u8; std::mem::size_of::<$num_ty>()];
-                    self.read_to_slice(&mut buf)?;
-                    Ok($num_ty::from_le_bytes(buf))
-                }
-
-                fn [<peek_ $num_ty _le>](&mut self) -> Result<$num_ty> {
-                    let mut buf = [0u8; std::mem::size_of::<$num_ty>()];
-                    self.peek_to_slice(&mut buf)?;
-                    Ok($num_ty::from_le_bytes(buf))
-                }
-            }
-        )*
-    }
-}
-
-macro_rules! impl_write_le {
-    ( $($num_ty: ty), *) => {
-        $(
-            paste!{
-                fn [<write_ $num_ty _le>](&mut self, n: $num_ty) -> Result<()> {
-                    self.write_from_slice(&n.to_le_bytes())?;
-                    Ok(())
-                }
-            }
-        )*
-    }
-}
-
-pub trait Buffer {
-    /// Returns remaining data size for read.
-    fn remaining_size(&self) -> usize;
-
-    /// Returns true if buffer has no data for read.
-    fn is_empty(&self) -> bool {
-        self.remaining_size() == 0
-    }
-
-    /// Peeks data into dst. This method should not change internal cursor,
-    /// invoke `advance_by` if needed.
-    /// # Panics
-    /// This method **may** panic if buffer does not have enough data to be copied to dst.
-    fn peek_to_slice(&self, dst: &mut [u8]) -> Result<()>;
-
-    /// Reads data into dst. This method will change internal cursor.
-    /// # Panics
-    /// This method **may** panic if buffer does not have enough data to be copied to dst.
-    fn read_to_slice(&mut self, dst: &mut [u8]) -> Result<()> {
-        self.peek_to_slice(dst)?;
-        self.advance_by(dst.len());
-        Ok(())
-    }
-
-    /// Advances internal cursor for next read.
-    /// # Panics
-    /// This method **may** panic if the offset after advancing exceeds the length of underlying buffer.
-    fn advance_by(&mut self, by: usize);
-
-    impl_read_le![u8, i8, u16, i16, u32, i32, u64, i64, f32, f64];
-}
-
-macro_rules! impl_buffer_for_bytes {
-    ( $($buf_ty:ty), *) => {
-        $(
-        impl Buffer for $buf_ty {
-            fn remaining_size(&self) -> usize{
-                self.len()
-            }
-
-            fn peek_to_slice(&self, dst: &mut [u8]) -> Result<()> {
-                let dst_len = dst.len();
-                ensure!(self.remaining() >= dst.len(), OverflowSnafu {
-                        src_len: self.remaining_size(),
-                        dst_len,
-                    }
-                );
-                dst.copy_from_slice(&self[0..dst_len]);
-                Ok(())
-            }
-
-            #[inline]
-            fn advance_by(&mut self, by: usize) {
-                self.advance(by);
-            }
-        }
-        )*
-    };
-}
-
-impl_buffer_for_bytes![bytes::Bytes, bytes::BytesMut];
-
-impl Buffer for &[u8] {
-    fn remaining_size(&self) -> usize {
-        self.len()
-    }
-
-    fn peek_to_slice(&self, dst: &mut [u8]) -> Result<()> {
-        let dst_len = dst.len();
-        ensure!(
-            self.len() >= dst.len(),
-            OverflowSnafu {
-                src_len: self.remaining_size(),
-                dst_len,
-            }
-        );
-        dst.copy_from_slice(&self[0..dst_len]);
-        Ok(())
-    }
-
-    fn read_to_slice(&mut self, dst: &mut [u8]) -> Result<()> {
-        ensure!(
-            self.len() >= dst.len(),
-            OverflowSnafu {
-                src_len: self.remaining_size(),
-                dst_len: dst.len(),
-            }
-        );
-        self.read_exact(dst).context(EofSnafu)
-    }
-
-    fn advance_by(&mut self, by: usize) {
-        *self = &self[by..];
-    }
-}
-
-/// Mutable buffer.
-pub trait BufferMut {
-    fn as_slice(&self) -> &[u8];
-
-    fn write_from_slice(&mut self, src: &[u8]) -> Result<()>;
-
-    impl_write_le![i8, u8, i16, u16, i32, u32, i64, u64, f32, f64];
-}
-
-impl BufferMut for BytesMut {
-    fn as_slice(&self) -> &[u8] {
-        self
-    }
-
-    fn write_from_slice(&mut self, src: &[u8]) -> Result<()> {
-        self.put_slice(src);
-        Ok(())
-    }
-}
-
-impl BufferMut for &mut [u8] {
-    fn as_slice(&self) -> &[u8] {
-        self
-    }
-
-    fn write_from_slice(&mut self, src: &[u8]) -> Result<()> {
-        // see std::io::Write::write_all
-        // https://doc.rust-lang.org/src/std/io/impls.rs.html#363
-        self.write_all(src).map_err(|_| {
-            OverflowSnafu {
-                src_len: src.len(),
-                dst_len: self.as_slice().len(),
-            }
-            .build()
-        })
-    }
-}
-
-impl BufferMut for Vec<u8> {
-    fn as_slice(&self) -> &[u8] {
-        self
-    }
-
-    fn write_from_slice(&mut self, src: &[u8]) -> Result<()> {
-        self.extend_from_slice(src);
-        Ok(())
-    }
-}

src/common/base/src/bytes.rs

@@ -44,6 +44,12 @@ impl From<Vec<u8>> for Bytes {
     }
 }
 
+impl From<Bytes> for Vec<u8> {
+    fn from(bytes: Bytes) -> Vec<u8> {
+        bytes.0.into()
+    }
+}
+
 impl Deref for Bytes {
     type Target = [u8];
 

src/common/base/src/lib.rs

@@ -13,9 +13,9 @@
 // limitations under the License.
 
 pub mod bit_vec;
-pub mod buffer;
 pub mod bytes;
 pub mod plugins;
+pub mod range_read;
 #[allow(clippy::all)]
 pub mod readable_size;
 pub mod secrets;

src/common/base/src/range_read.rs

@@ -0,0 +1,105 @@
+// 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::io;
+use std::ops::Range;
+
+use async_trait::async_trait;
+use bytes::{BufMut, Bytes};
+use futures::{AsyncReadExt, AsyncSeekExt};
+
+/// `Metadata` contains the metadata of a source.
+pub struct Metadata {
+    /// The length of the source in bytes.
+    pub content_length: u64,
+}
+
+/// `RangeReader` reads a range of bytes from a source.
+#[async_trait]
+pub trait RangeReader: Send + Unpin {
+    /// Returns the metadata of the source.
+    async fn metadata(&mut self) -> io::Result<Metadata>;
+
+    /// Reads the bytes in the given range.
+    async fn read(&mut self, range: Range<u64>) -> io::Result<Bytes>;
+
+    /// Reads the bytes in the given range into the buffer.
+    ///
+    /// Handles the buffer based on its capacity:
+    /// - If the buffer is insufficient to hold the bytes, it will either:
+    ///   - Allocate additional space (e.g., for `Vec<u8>`)
+    ///   - Panic (e.g., for `&mut [u8]`)
+    async fn read_into(
+        &mut self,
+        range: Range<u64>,
+        buf: &mut (impl BufMut + Send),
+    ) -> io::Result<()> {
+        let bytes = self.read(range).await?;
+        buf.put_slice(&bytes);
+        Ok(())
+    }
+
+    /// Reads the bytes in the given ranges.
+    async fn read_vec(&mut self, ranges: &[Range<u64>]) -> io::Result<Vec<Bytes>> {
+        let mut result = Vec::with_capacity(ranges.len());
+        for range in ranges {
+            result.push(self.read(range.clone()).await?);
+        }
+        Ok(result)
+    }
+}
+
+#[async_trait]
+impl<R: RangeReader + Send + Unpin> RangeReader for &mut R {
+    async fn metadata(&mut self) -> io::Result<Metadata> {
+        (*self).metadata().await
+    }
+    async fn read(&mut self, range: Range<u64>) -> io::Result<Bytes> {
+        (*self).read(range).await
+    }
+    async fn read_into(
+        &mut self,
+        range: Range<u64>,
+        buf: &mut (impl BufMut + Send),
+    ) -> io::Result<()> {
+        (*self).read_into(range, buf).await
+    }
+    async fn read_vec(&mut self, ranges: &[Range<u64>]) -> io::Result<Vec<Bytes>> {
+        (*self).read_vec(ranges).await
+    }
+}
+
+/// `RangeReaderAdapter` bridges `RangeReader` and `AsyncRead + AsyncSeek`.
+pub struct RangeReaderAdapter<R>(pub R);
+
+/// Implements `RangeReader` for a type that implements `AsyncRead + AsyncSeek`.
+///
+/// TODO(zhongzc): It's a temporary solution for porting the codebase from `AsyncRead + AsyncSeek` to `RangeReader`.
+/// Until the codebase is fully ported to `RangeReader`, remove this implementation.
+#[async_trait]
+impl<R: futures::AsyncRead + futures::AsyncSeek + Send + Unpin> RangeReader
+    for RangeReaderAdapter<R>
+{
+    async fn metadata(&mut self) -> io::Result<Metadata> {
+        let content_length = self.0.seek(io::SeekFrom::End(0)).await?;
+        Ok(Metadata { content_length })
+    }
+
+    async fn read(&mut self, range: Range<u64>) -> io::Result<Bytes> {
+        let mut buf = vec![0; (range.end - range.start) as usize];
+        self.0.seek(io::SeekFrom::Start(range.start)).await?;
+        self.0.read_exact(&mut buf).await?;
+        Ok(Bytes::from(buf))
+    }
+}

src/common/base/tests/buffer_tests.rs

@@ -1,182 +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.
-
-#![feature(assert_matches)]
-
-#[cfg(test)]
-mod tests {
-    use std::assert_matches::assert_matches;
-
-    use bytes::{Buf, Bytes, BytesMut};
-    use common_base::buffer::Error::Overflow;
-    use common_base::buffer::{Buffer, BufferMut};
-    use paste::paste;
-
-    #[test]
-    pub fn test_buffer_read_write() {
-        let mut buf = BytesMut::with_capacity(16);
-        buf.write_u64_le(1234u64).unwrap();
-        let result = buf.peek_u64_le().unwrap();
-        assert_eq!(1234u64, result);
-        buf.advance_by(8);
-
-        buf.write_from_slice("hello, world".as_bytes()).unwrap();
-        let mut content = vec![0u8; 5];
-        buf.peek_to_slice(&mut content).unwrap();
-        let read = String::from_utf8_lossy(&content);
-        assert_eq!("hello", read);
-        buf.advance_by(5);
-        // after read, buffer should still have 7 bytes to read.
-        assert_eq!(7, buf.remaining());
-
-        let mut content = vec![0u8; 6];
-        buf.read_to_slice(&mut content).unwrap();
-        let read = String::from_utf8_lossy(&content);
-        assert_eq!(", worl", read);
-        // after read, buffer should still have 1 byte to read.
-        assert_eq!(1, buf.remaining());
-    }
-
-    #[test]
-    pub fn test_buffer_read() {
-        let mut bytes = Bytes::from_static("hello".as_bytes());
-        assert_eq!(5, bytes.remaining_size());
-        assert_eq!(b'h', bytes.peek_u8_le().unwrap());
-        bytes.advance_by(1);
-        assert_eq!(4, bytes.remaining_size());
-    }
-
-    macro_rules! test_primitive_read_write {
-        ( $($num_ty: ty), *) => {
-            $(
-                paste!{
-                    #[test]
-                    fn [<test_read_write_ $num_ty>]() {
-                        assert_eq!($num_ty::MAX,(&mut $num_ty::MAX.to_le_bytes() as &[u8]).[<read_ $num_ty _le>]().unwrap());
-                        assert_eq!($num_ty::MIN,(&mut $num_ty::MIN.to_le_bytes() as &[u8]).[<read_ $num_ty _le>]().unwrap());
-                    }
-                }
-            )*
-        }
-    }
-
-    test_primitive_read_write![u8, u16, u32, u64, i8, i16, i32, i64, f32, f64];
-
-    #[test]
-    pub fn test_read_write_from_slice_buffer() {
-        let mut buf = "hello".as_bytes();
-        assert_eq!(104, buf.peek_u8_le().unwrap());
-        buf.advance_by(1);
-        assert_eq!(101, buf.peek_u8_le().unwrap());
-        buf.advance_by(1);
-        assert_eq!(108, buf.peek_u8_le().unwrap());
-        buf.advance_by(1);
-        assert_eq!(108, buf.peek_u8_le().unwrap());
-        buf.advance_by(1);
-        assert_eq!(111, buf.peek_u8_le().unwrap());
-        buf.advance_by(1);
-        assert_matches!(buf.peek_u8_le(), Err(Overflow { .. }));
-    }
-
-    #[test]
-    pub fn test_read_u8_from_slice_buffer() {
-        let mut buf = "hello".as_bytes();
-        assert_eq!(104, buf.read_u8_le().unwrap());
-        assert_eq!(101, buf.read_u8_le().unwrap());
-        assert_eq!(108, buf.read_u8_le().unwrap());
-        assert_eq!(108, buf.read_u8_le().unwrap());
-        assert_eq!(111, buf.read_u8_le().unwrap());
-        assert_matches!(buf.read_u8_le(), Err(Overflow { .. }));
-    }
-
-    #[test]
-    pub fn test_read_write_numbers() {
-        let mut buf: Vec<u8> = vec![];
-        buf.write_u64_le(1234).unwrap();
-        assert_eq!(1234, (&buf[..]).read_u64_le().unwrap());
-
-        buf.write_u32_le(4242).unwrap();
-        let mut p = &buf[..];
-        assert_eq!(1234, p.read_u64_le().unwrap());
-        assert_eq!(4242, p.read_u32_le().unwrap());
-    }
-
-    macro_rules! test_primitive_vec_read_write {
-        ( $($num_ty: ty), *) => {
-            $(
-                paste!{
-                    #[test]
-                    fn [<test_read_write_ $num_ty _from_vec_buffer>]() {
-                        let mut buf = vec![];
-                        let _ = buf.[<write_ $num_ty _le>]($num_ty::MAX).unwrap();
-                        assert_eq!($num_ty::MAX, buf.as_slice().[<read_ $num_ty _le>]().unwrap());
-                    }
-                }
-            )*
-        }
-    }
-
-    test_primitive_vec_read_write![u8, u16, u32, u64, i8, i16, i32, i64, f32, f64];
-
-    #[test]
-    pub fn test_peek_write_from_vec_buffer() {
-        let mut buf: Vec<u8> = vec![];
-        buf.write_from_slice("hello".as_bytes()).unwrap();
-        let mut slice = buf.as_slice();
-        assert_eq!(104, slice.peek_u8_le().unwrap());
-        slice.advance_by(1);
-        assert_eq!(101, slice.peek_u8_le().unwrap());
-        slice.advance_by(1);
-        assert_eq!(108, slice.peek_u8_le().unwrap());
-        slice.advance_by(1);
-        assert_eq!(108, slice.peek_u8_le().unwrap());
-        slice.advance_by(1);
-        assert_eq!(111, slice.peek_u8_le().unwrap());
-        slice.advance_by(1);
-        assert_matches!(slice.read_u8_le(), Err(Overflow { .. }));
-    }
-
-    macro_rules! test_primitive_bytes_read_write {
-        ( $($num_ty: ty), *) => {
-            $(
-                paste!{
-                    #[test]
-                    fn [<test_read_write_ $num_ty _from_bytes>]() {
-                        let mut bytes = bytes::Bytes::from($num_ty::MAX.to_le_bytes().to_vec());
-                        assert_eq!($num_ty::MAX, bytes.[<read_ $num_ty _le>]().unwrap());
-
-                        let mut bytes = bytes::Bytes::from($num_ty::MIN.to_le_bytes().to_vec());
-                        assert_eq!($num_ty::MIN, bytes.[<read_ $num_ty _le>]().unwrap());
-                    }
-                }
-            )*
-        }
-    }
-
-    test_primitive_bytes_read_write![u8, u16, u32, u64, i8, i16, i32, i64, f32, f64];
-
-    #[test]
-    pub fn test_write_overflow() {
-        let mut buf = [0u8; 4];
-        assert_matches!(
-            (&mut buf[..]).write_from_slice("hell".as_bytes()),
-            Ok { .. }
-        );
-
-        assert_matches!(
-            (&mut buf[..]).write_from_slice("hello".as_bytes()),
-            Err(common_base::buffer::Error::Overflow { .. })
-        );
-    }
-}

src/common/catalog/src/error.rs

@@ -1,46 +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::any::Any;
-
-use common_error::ext::ErrorExt;
-use common_error::status_code::StatusCode;
-use common_macro::stack_trace_debug;
-use snafu::{Location, Snafu};
-
-#[derive(Snafu)]
-#[snafu(visibility(pub))]
-#[stack_trace_debug]
-pub enum Error {
-    #[snafu(display("Invalid full table name: {}", table_name))]
-    InvalidFullTableName {
-        table_name: String,
-        #[snafu(implicit)]
-        location: Location,
-    },
-}
-
-impl ErrorExt for Error {
-    fn status_code(&self) -> StatusCode {
-        match self {
-            Error::InvalidFullTableName { .. } => StatusCode::Unexpected,
-        }
-    }
-
-    fn as_any(&self) -> &dyn Any {
-        self
-    }
-}
-
-pub type Result<T> = std::result::Result<T, Error>;

src/common/catalog/src/lib.rs

@@ -15,7 +15,6 @@
 use consts::DEFAULT_CATALOG_NAME;
 
 pub mod consts;
-pub mod error;
 
 #[inline]
 pub fn format_schema_name(catalog: &str, schema: &str) -> String {
@@ -48,19 +47,19 @@ pub fn build_db_string(catalog: &str, schema: &str) -> String {
 /// The database name may come from different sources:
 ///
 /// - MySQL `schema` name in MySQL protocol login request: it's optional and user
-/// and switch database using `USE` command
+///   and switch database using `USE` command
 /// - Postgres `database` parameter in Postgres wire protocol, required
 /// - HTTP RESTful API: the database parameter, optional
 /// - gRPC: the dbname field in header, optional but has a higher priority than
-/// original catalog/schema
+///   original catalog/schema
 ///
 /// When database name is provided, we attempt to parse catalog and schema from
 /// it. We assume the format `[<catalog>-]<schema>`:
 ///
 /// - If `[<catalog>-]` part is not provided, we use whole database name as
-/// schema name
+///   schema name
 /// - if `[<catalog>-]` is provided, we split database name with `-` and use
-/// `<catalog>` and `<schema>`.
+///   `<catalog>` and `<schema>`.
 pub fn parse_catalog_and_schema_from_db_string(db: &str) -> (String, String) {
     match parse_optional_catalog_and_schema_from_db_string(db) {
         (Some(catalog), schema) => (catalog, schema),

src/common/datasource/src/object_store/fs.rs

@@ -13,6 +13,7 @@
 // limitations under the License.
 
 use object_store::services::Fs;
+use object_store::util::DefaultLoggingInterceptor;
 use object_store::ObjectStore;
 use snafu::ResultExt;
 
@@ -22,13 +23,9 @@ pub fn build_fs_backend(root: &str) -> Result<ObjectStore> {
     let builder = Fs::default();
     let object_store = ObjectStore::new(builder.root(root))
         .context(BuildBackendSnafu)?
-        .layer(
-            object_store::layers::LoggingLayer::default()
-                // Print the expected error only in DEBUG level.
-                // See https://docs.rs/opendal/latest/opendal/layers/struct.LoggingLayer.html#method.with_error_level
-                .with_error_level(Some("debug"))
-                .expect("input error level must be valid"),
-        )
+        .layer(object_store::layers::LoggingLayer::new(
+            DefaultLoggingInterceptor,
+        ))
         .layer(object_store::layers::TracingLayer)
         .layer(object_store::layers::PrometheusMetricsLayer::new(true))
         .finish();

src/common/datasource/src/object_store/s3.rs

@@ -15,6 +15,7 @@
 use std::collections::HashMap;
 
 use object_store::services::S3;
+use object_store::util::DefaultLoggingInterceptor;
 use object_store::ObjectStore;
 use snafu::ResultExt;
 
@@ -84,13 +85,9 @@ pub fn build_s3_backend(
     // TODO(weny): Consider finding a better way to eliminate duplicate code.
     Ok(ObjectStore::new(builder)
         .context(error::BuildBackendSnafu)?
-        .layer(
-            object_store::layers::LoggingLayer::default()
-                // Print the expected error only in DEBUG level.
-                // See https://docs.rs/opendal/latest/opendal/layers/struct.LoggingLayer.html#method.with_error_level
-                .with_error_level(Some("debug"))
-                .expect("input error level must be valid"),
-        )
+        .layer(object_store::layers::LoggingLayer::new(
+            DefaultLoggingInterceptor,
+        ))
         .layer(object_store::layers::TracingLayer)
         .layer(object_store::layers::PrometheusMetricsLayer::new(true))
         .finish())

src/common/function/Cargo.toml

@@ -7,6 +7,10 @@ license.workspace = true
 [lints]
 workspace = true
 
+[features]
+default = ["geo"]
+geo = ["geohash", "h3o"]
+
 [dependencies]
 api.workspace = true
 arc-swap = "1.0"
@@ -23,6 +27,9 @@ common-time.workspace = true
 common-version.workspace = true
 datafusion.workspace = true
 datatypes.workspace = true
+geohash = { version = "0.13", optional = true }
+h3o = { version = "0.6", optional = true }
+jsonb.workspace = true
 num = "0.4"
 num-traits = "0.2"
 once_cell.workspace = true
@@ -39,3 +46,4 @@ table.workspace = true
 [dev-dependencies]
 ron = "0.7"
 serde = { version = "1.0", features = ["derive"] }
+tokio.workspace = true

src/common/function/src/flush_flow.rs

@@ -110,7 +110,7 @@ mod test {
     use session::context::QueryContext;
 
     use super::*;
-    use crate::function::{Function, FunctionContext};
+    use crate::function::{AsyncFunction, FunctionContext};
 
     #[test]
     fn test_flush_flow_metadata() {
@@ -130,8 +130,8 @@ mod test {
         );
     }
 
-    #[test]
-    fn test_missing_flow_service() {
+    #[tokio::test]
+    async fn test_missing_flow_service() {
         let f = FlushFlowFunction;
 
         let args = vec!["flow_name"];
@@ -140,7 +140,7 @@ mod test {
             .map(|arg| Arc::new(StringVector::from_slice(&[arg])) as _)
             .collect::<Vec<_>>();
 
-        let result = f.eval(FunctionContext::default(), &args).unwrap_err();
+        let result = f.eval(FunctionContext::default(), &args).await.unwrap_err();
         assert_eq!(
             "Missing FlowServiceHandler, not expected",
             result.to_string()

src/common/function/src/function.rs

@@ -32,7 +32,7 @@ pub struct FunctionContext {
 
 impl FunctionContext {
     /// Create a mock [`FunctionContext`] for test.
-    #[cfg(any(test, feature = "testing"))]
+    #[cfg(test)]
     pub fn mock() -> Self {
         Self {
             query_ctx: QueryContextBuilder::default().build().into(),
@@ -56,8 +56,10 @@ pub trait Function: fmt::Display + Sync + Send {
     /// Returns the name of the function, should be unique.
     fn name(&self) -> &str;
 
+    /// The returned data type of function execution.
     fn return_type(&self, input_types: &[ConcreteDataType]) -> Result<ConcreteDataType>;
 
+    /// The signature of function.
     fn signature(&self) -> Signature;
 
     /// Evaluate the function, e.g. run/execute the function.
@@ -65,3 +67,22 @@ pub trait Function: fmt::Display + Sync + Send {
 }
 
 pub type FunctionRef = Arc<dyn Function>;
+
+/// Async Scalar function trait
+#[async_trait::async_trait]
+pub trait AsyncFunction: fmt::Display + Sync + Send {
+    /// Returns the name of the function, should be unique.
+    fn name(&self) -> &str;
+
+    /// The returned data type of function execution.
+    fn return_type(&self, input_types: &[ConcreteDataType]) -> Result<ConcreteDataType>;
+
+    /// The signature of function.
+    fn signature(&self) -> Signature;
+
+    /// Evaluate the function, e.g. run/execute the function.
+    /// TODO(dennis): simplify the signature and refactor all the admin functions.
+    async fn eval(&self, _func_ctx: FunctionContext, _columns: &[VectorRef]) -> Result<VectorRef>;
+}
+
+pub type AsyncFunctionRef = Arc<dyn AsyncFunction>;

src/common/function/src/function_registry.rs

@@ -18,10 +18,11 @@ use std::sync::{Arc, RwLock};
 
 use once_cell::sync::Lazy;
 
-use crate::function::FunctionRef;
+use crate::function::{AsyncFunctionRef, FunctionRef};
 use crate::scalars::aggregate::{AggregateFunctionMetaRef, AggregateFunctions};
 use crate::scalars::date::DateFunction;
 use crate::scalars::expression::ExpressionFunction;
+use crate::scalars::json::JsonFunction;
 use crate::scalars::matches::MatchesFunction;
 use crate::scalars::math::MathFunction;
 use crate::scalars::numpy::NumpyFunction;
@@ -32,6 +33,7 @@ use crate::table::TableFunction;
 #[derive(Default)]
 pub struct FunctionRegistry {
     functions: RwLock<HashMap<String, FunctionRef>>,
+    async_functions: RwLock<HashMap<String, AsyncFunctionRef>>,
     aggregate_functions: RwLock<HashMap<String, AggregateFunctionMetaRef>>,
 }
 
@@ -44,6 +46,27 @@ impl FunctionRegistry {
             .insert(func.name().to_string(), func);
     }
 
+    pub fn register_async(&self, func: AsyncFunctionRef) {
+        let _ = self
+            .async_functions
+            .write()
+            .unwrap()
+            .insert(func.name().to_string(), func);
+    }
+
+    pub fn get_async_function(&self, name: &str) -> Option<AsyncFunctionRef> {
+        self.async_functions.read().unwrap().get(name).cloned()
+    }
+
+    pub fn async_functions(&self) -> Vec<AsyncFunctionRef> {
+        self.async_functions
+            .read()
+            .unwrap()
+            .values()
+            .cloned()
+            .collect()
+    }
+
     pub fn register_aggregate_function(&self, func: AggregateFunctionMetaRef) {
         let _ = self
             .aggregate_functions
@@ -94,6 +117,13 @@ pub static FUNCTION_REGISTRY: Lazy<Arc<FunctionRegistry>> = Lazy::new(|| {
     SystemFunction::register(&function_registry);
     TableFunction::register(&function_registry);
 
+    // Json related functions
+    JsonFunction::register(&function_registry);
+
+    // Geo functions
+    #[cfg(feature = "geo")]
+    crate::scalars::geo::GeoFunctions::register(&function_registry);
+
     Arc::new(function_registry)
 });
 

src/common/function/src/scalars.rs

@@ -15,9 +15,13 @@
 pub mod aggregate;
 pub(crate) mod date;
 pub mod expression;
+#[cfg(feature = "geo")]
+pub mod geo;
+pub mod json;
 pub mod matches;
 pub mod math;
 pub mod numpy;
+
 #[cfg(test)]
 pub(crate) mod test;
 pub(crate) mod timestamp;

src/common/function/src/scalars/aggregate/percentile.rs

@@ -75,7 +75,7 @@ where
         // to keep the not_greater length == floor+1
         // so to ensure the peek of the not_greater is array[floor]
         // and the peek of the greater is array[floor+1]
-        let p = if let Some(p) = self.p { p } else { 0.0_f64 };
+        let p = self.p.unwrap_or(0.0_f64);
         let floor = (((self.n - 1) as f64) * p / (100_f64)).floor();
         if value <= *self.not_greater.peek().unwrap() {
             self.not_greater.push(value);

src/common/function/src/scalars/geo.rs

@@ -0,0 +1,31 @@
+// 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;
+mod geohash;
+mod h3;
+
+use geohash::GeohashFunction;
+use h3::H3Function;
+
+use crate::function_registry::FunctionRegistry;
+
+pub(crate) struct GeoFunctions;
+
+impl GeoFunctions {
+    pub fn register(registry: &FunctionRegistry) {
+        registry.register(Arc::new(GeohashFunction));
+        registry.register(Arc::new(H3Function));
+    }
+}

src/common/function/src/scalars/geo/geohash.rs

@@ -0,0 +1,135 @@
+// 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::fmt;
+
+use common_error::ext::{BoxedError, PlainError};
+use common_error::status_code::StatusCode;
+use common_query::error::{self, InvalidFuncArgsSnafu, Result};
+use common_query::prelude::{Signature, TypeSignature};
+use datafusion::logical_expr::Volatility;
+use datatypes::prelude::ConcreteDataType;
+use datatypes::scalars::ScalarVectorBuilder;
+use datatypes::value::Value;
+use datatypes::vectors::{MutableVector, StringVectorBuilder, VectorRef};
+use geohash::Coord;
+use snafu::{ensure, ResultExt};
+
+use crate::function::{Function, FunctionContext};
+
+/// Function that return geohash string for a given geospatial coordinate.
+#[derive(Clone, Debug, Default)]
+pub struct GeohashFunction;
+
+const NAME: &str = "geohash";
+
+impl Function for GeohashFunction {
+    fn name(&self) -> &str {
+        NAME
+    }
+
+    fn return_type(&self, _input_types: &[ConcreteDataType]) -> Result<ConcreteDataType> {
+        Ok(ConcreteDataType::string_datatype())
+    }
+
+    fn signature(&self) -> Signature {
+        let mut signatures = Vec::new();
+        for coord_type in &[
+            ConcreteDataType::float32_datatype(),
+            ConcreteDataType::float64_datatype(),
+        ] {
+            for resolution_type in &[
+                ConcreteDataType::int8_datatype(),
+                ConcreteDataType::int16_datatype(),
+                ConcreteDataType::int32_datatype(),
+                ConcreteDataType::int64_datatype(),
+                ConcreteDataType::uint8_datatype(),
+                ConcreteDataType::uint16_datatype(),
+                ConcreteDataType::uint32_datatype(),
+                ConcreteDataType::uint64_datatype(),
+            ] {
+                signatures.push(TypeSignature::Exact(vec![
+                    // latitude
+                    coord_type.clone(),
+                    // longitude
+                    coord_type.clone(),
+                    // resolution
+                    resolution_type.clone(),
+                ]));
+            }
+        }
+        Signature::one_of(signatures, Volatility::Stable)
+    }
+
+    fn eval(&self, _func_ctx: FunctionContext, columns: &[VectorRef]) -> Result<VectorRef> {
+        ensure!(
+            columns.len() == 3,
+            InvalidFuncArgsSnafu {
+                err_msg: format!(
+                    "The length of the args is not correct, expect 3, provided : {}",
+                    columns.len()
+                ),
+            }
+        );
+
+        let lat_vec = &columns[0];
+        let lon_vec = &columns[1];
+        let resolution_vec = &columns[2];
+
+        let size = lat_vec.len();
+        let mut results = StringVectorBuilder::with_capacity(size);
+
+        for i in 0..size {
+            let lat = lat_vec.get(i).as_f64_lossy();
+            let lon = lon_vec.get(i).as_f64_lossy();
+            let r = match resolution_vec.get(i) {
+                Value::Int8(v) => v as usize,
+                Value::Int16(v) => v as usize,
+                Value::Int32(v) => v as usize,
+                Value::Int64(v) => v as usize,
+                Value::UInt8(v) => v as usize,
+                Value::UInt16(v) => v as usize,
+                Value::UInt32(v) => v as usize,
+                Value::UInt64(v) => v as usize,
+                _ => unreachable!(),
+            };
+
+            let result = match (lat, lon) {
+                (Some(lat), Some(lon)) => {
+                    let coord = Coord { x: lon, y: lat };
+                    let encoded = geohash::encode(coord, r)
+                        .map_err(|e| {
+                            BoxedError::new(PlainError::new(
+                                format!("Geohash error: {}", e),
+                                StatusCode::EngineExecuteQuery,
+                            ))
+                        })
+                        .context(error::ExecuteSnafu)?;
+                    Some(encoded)
+                }
+                _ => None,
+            };
+
+            results.push(result.as_deref());
+        }
+
+        Ok(results.to_vector())
+    }
+}
+
+impl fmt::Display for GeohashFunction {
+    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
+        write!(f, "{}", NAME)
+    }
+}

src/common/function/src/scalars/geo/h3.rs

@@ -0,0 +1,145 @@
+// 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::fmt;
+
+use common_error::ext::{BoxedError, PlainError};
+use common_error::status_code::StatusCode;
+use common_query::error::{self, InvalidFuncArgsSnafu, Result};
+use common_query::prelude::{Signature, TypeSignature};
+use datafusion::logical_expr::Volatility;
+use datatypes::prelude::ConcreteDataType;
+use datatypes::scalars::ScalarVectorBuilder;
+use datatypes::value::Value;
+use datatypes::vectors::{MutableVector, StringVectorBuilder, VectorRef};
+use h3o::{LatLng, Resolution};
+use snafu::{ensure, ResultExt};
+
+use crate::function::{Function, FunctionContext};
+
+/// Function that returns [h3] encoding string for a given geospatial coordinate.
+///
+/// [h3]: https://h3geo.org/
+#[derive(Clone, Debug, Default)]
+pub struct H3Function;
+
+const NAME: &str = "h3";
+
+impl Function for H3Function {
+    fn name(&self) -> &str {
+        NAME
+    }
+
+    fn return_type(&self, _input_types: &[ConcreteDataType]) -> Result<ConcreteDataType> {
+        Ok(ConcreteDataType::string_datatype())
+    }
+
+    fn signature(&self) -> Signature {
+        let mut signatures = Vec::new();
+        for coord_type in &[
+            ConcreteDataType::float32_datatype(),
+            ConcreteDataType::float64_datatype(),
+        ] {
+            for resolution_type in &[
+                ConcreteDataType::int8_datatype(),
+                ConcreteDataType::int16_datatype(),
+                ConcreteDataType::int32_datatype(),
+                ConcreteDataType::int64_datatype(),
+                ConcreteDataType::uint8_datatype(),
+                ConcreteDataType::uint16_datatype(),
+                ConcreteDataType::uint32_datatype(),
+                ConcreteDataType::uint64_datatype(),
+            ] {
+                signatures.push(TypeSignature::Exact(vec![
+                    // latitude
+                    coord_type.clone(),
+                    // longitude
+                    coord_type.clone(),
+                    // resolution
+                    resolution_type.clone(),
+                ]));
+            }
+        }
+        Signature::one_of(signatures, Volatility::Stable)
+    }
+
+    fn eval(&self, _func_ctx: FunctionContext, columns: &[VectorRef]) -> Result<VectorRef> {
+        ensure!(
+            columns.len() == 3,
+            InvalidFuncArgsSnafu {
+                err_msg: format!(
+                    "The length of the args is not correct, expect 3, provided : {}",
+                    columns.len()
+                ),
+            }
+        );
+
+        let lat_vec = &columns[0];
+        let lon_vec = &columns[1];
+        let resolution_vec = &columns[2];
+
+        let size = lat_vec.len();
+        let mut results = StringVectorBuilder::with_capacity(size);
+
+        for i in 0..size {
+            let lat = lat_vec.get(i).as_f64_lossy();
+            let lon = lon_vec.get(i).as_f64_lossy();
+            let r = match resolution_vec.get(i) {
+                Value::Int8(v) => v as u8,
+                Value::Int16(v) => v as u8,
+                Value::Int32(v) => v as u8,
+                Value::Int64(v) => v as u8,
+                Value::UInt8(v) => v,
+                Value::UInt16(v) => v as u8,
+                Value::UInt32(v) => v as u8,
+                Value::UInt64(v) => v as u8,
+                _ => unreachable!(),
+            };
+
+            let result = match (lat, lon) {
+                (Some(lat), Some(lon)) => {
+                    let coord = LatLng::new(lat, lon)
+                        .map_err(|e| {
+                            BoxedError::new(PlainError::new(
+                                format!("H3 error: {}", e),
+                                StatusCode::EngineExecuteQuery,
+                            ))
+                        })
+                        .context(error::ExecuteSnafu)?;
+                    let r = Resolution::try_from(r)
+                        .map_err(|e| {
+                            BoxedError::new(PlainError::new(
+                                format!("H3 error: {}", e),
+                                StatusCode::EngineExecuteQuery,
+                            ))
+                        })
+                        .context(error::ExecuteSnafu)?;
+                    let encoded = coord.to_cell(r).to_string();
+                    Some(encoded)
+                }
+                _ => None,
+            };
+
+            results.push(result.as_deref());
+        }
+
+        Ok(results.to_vector())
+    }
+}
+
+impl fmt::Display for H3Function {
+    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
+        write!(f, "{}", NAME)
+    }
+}

src/common/function/src/scalars/json.rs

@@ -0,0 +1,38 @@
+// 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;
+mod json_get;
+mod json_to_string;
+mod to_json;
+
+use json_get::{JsonGetBool, JsonGetFloat, JsonGetInt, JsonGetString};
+use json_to_string::JsonToStringFunction;
+use to_json::ToJsonFunction;
+
+use crate::function_registry::FunctionRegistry;
+
+pub(crate) struct JsonFunction;
+
+impl JsonFunction {
+    pub fn register(registry: &FunctionRegistry) {
+        registry.register(Arc::new(JsonToStringFunction));
+        registry.register(Arc::new(ToJsonFunction));
+
+        registry.register(Arc::new(JsonGetInt));
+        registry.register(Arc::new(JsonGetFloat));
+        registry.register(Arc::new(JsonGetString));
+        registry.register(Arc::new(JsonGetBool));
+    }
+}

src/common/function/src/scalars/json/json_get.rs

@@ -0,0 +1,454 @@
+// 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::fmt::{self, Display};
+
+use common_query::error::{InvalidFuncArgsSnafu, Result, UnsupportedInputDataTypeSnafu};
+use common_query::prelude::Signature;
+use datafusion::logical_expr::Volatility;
+use datatypes::data_type::ConcreteDataType;
+use datatypes::prelude::VectorRef;
+use datatypes::scalars::ScalarVectorBuilder;
+use datatypes::vectors::{
+    BooleanVectorBuilder, Float64VectorBuilder, Int64VectorBuilder, MutableVector,
+    StringVectorBuilder,
+};
+use snafu::ensure;
+
+use crate::function::{Function, FunctionContext};
+
+fn get_json_by_path(json: &[u8], path: &str) -> Option<Vec<u8>> {
+    let json_path = jsonb::jsonpath::parse_json_path(path.as_bytes());
+    match json_path {
+        Ok(json_path) => {
+            let mut sub_jsonb = Vec::new();
+            let mut sub_offsets = Vec::new();
+            match jsonb::get_by_path(json, json_path, &mut sub_jsonb, &mut sub_offsets) {
+                Ok(_) => Some(sub_jsonb),
+                Err(_) => None,
+            }
+        }
+        _ => None,
+    }
+}
+
+/// Get the value from the JSONB by the given path and return it as specified type.
+/// If the path does not exist or the value is not the type specified, return `NULL`.
+macro_rules! json_get {
+    // e.g. name = JsonGetInt, type = Int64, rust_type = i64, doc = "Get the value from the JSONB by the given path and return it as an integer."
+    ($name: ident, $type: ident, $rust_type: ident, $doc:expr) => {
+        paste::paste! {
+            #[doc = $doc]
+            #[derive(Clone, Debug, Default)]
+            pub struct $name;
+
+            impl Function for $name {
+                fn name(&self) -> &str {
+                    stringify!([<$name:snake>])
+                }
+
+                fn return_type(&self, _input_types: &[ConcreteDataType]) -> Result<ConcreteDataType> {
+                    Ok(ConcreteDataType::[<$type:snake _datatype>]())
+                }
+
+                fn signature(&self) -> Signature {
+                    Signature::exact(
+                        vec![
+                            ConcreteDataType::json_datatype(),
+                            ConcreteDataType::string_datatype(),
+                        ],
+                        Volatility::Immutable,
+                    )
+                }
+
+                fn eval(&self, _func_ctx: FunctionContext, columns: &[VectorRef]) -> Result<VectorRef> {
+                    ensure!(
+                        columns.len() == 2,
+                        InvalidFuncArgsSnafu {
+                            err_msg: format!(
+                                "The length of the args is not correct, expect exactly two, have: {}",
+                                columns.len()
+                            ),
+                        }
+                    );
+                    let jsons = &columns[0];
+                    let paths = &columns[1];
+
+                    let size = jsons.len();
+                    let datatype = jsons.data_type();
+                    let mut results = [<$type VectorBuilder>]::with_capacity(size);
+
+                    match datatype {
+                        // JSON data type uses binary vector
+                        ConcreteDataType::Binary(_) => {
+                            for i in 0..size {
+                                let json = jsons.get_ref(i);
+                                let path = paths.get_ref(i);
+
+                                let json = json.as_binary();
+                                let path = path.as_string();
+                                let result = match (json, path) {
+                                    (Ok(Some(json)), Ok(Some(path))) => {
+                                        get_json_by_path(json, path)
+                                            .and_then(|json| { jsonb::[<to_ $rust_type>](&json).ok() })
+                                    }
+                                    _ => None,
+                                };
+
+                                results.push(result);
+                            }
+                        }
+                        _ => {
+                            return UnsupportedInputDataTypeSnafu {
+                                function: stringify!([<$name:snake>]),
+                                datatypes: columns.iter().map(|c| c.data_type()).collect::<Vec<_>>(),
+                            }
+                            .fail();
+                        }
+                    }
+
+                    Ok(results.to_vector())
+                }
+            }
+
+            impl Display for $name {
+                fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
+                    write!(f, "{}", stringify!([<$name:snake>]).to_ascii_uppercase())
+                }
+            }
+        }
+    };
+}
+
+json_get!(
+    JsonGetInt,
+    Int64,
+    i64,
+    "Get the value from the JSONB by the given path and return it as an integer."
+);
+
+json_get!(
+    JsonGetFloat,
+    Float64,
+    f64,
+    "Get the value from the JSONB by the given path and return it as a float."
+);
+
+json_get!(
+    JsonGetBool,
+    Boolean,
+    bool,
+    "Get the value from the JSONB by the given path and return it as a boolean."
+);
+
+/// Get the value from the JSONB by the given path and return it as a string.
+#[derive(Clone, Debug, Default)]
+pub struct JsonGetString;
+
+impl Function for JsonGetString {
+    fn name(&self) -> &str {
+        "json_get_string"
+    }
+
+    fn return_type(&self, _input_types: &[ConcreteDataType]) -> Result<ConcreteDataType> {
+        Ok(ConcreteDataType::string_datatype())
+    }
+
+    fn signature(&self) -> Signature {
+        Signature::exact(
+            vec![
+                ConcreteDataType::json_datatype(),
+                ConcreteDataType::string_datatype(),
+            ],
+            Volatility::Immutable,
+        )
+    }
+
+    fn eval(&self, _func_ctx: FunctionContext, columns: &[VectorRef]) -> Result<VectorRef> {
+        ensure!(
+            columns.len() == 2,
+            InvalidFuncArgsSnafu {
+                err_msg: format!(
+                    "The length of the args is not correct, expect exactly two, have: {}",
+                    columns.len()
+                ),
+            }
+        );
+        let jsons = &columns[0];
+        let paths = &columns[1];
+
+        let size = jsons.len();
+        let datatype = jsons.data_type();
+        let mut results = StringVectorBuilder::with_capacity(size);
+
+        match datatype {
+            // JSON data type uses binary vector
+            ConcreteDataType::Binary(_) => {
+                for i in 0..size {
+                    let json = jsons.get_ref(i);
+                    let path = paths.get_ref(i);
+
+                    let json = json.as_binary();
+                    let path = path.as_string();
+                    let result = match (json, path) {
+                        (Ok(Some(json)), Ok(Some(path))) => {
+                            get_json_by_path(json, path).and_then(|json| jsonb::to_str(&json).ok())
+                        }
+                        _ => None,
+                    };
+
+                    results.push(result.as_deref());
+                }
+            }
+            _ => {
+                return UnsupportedInputDataTypeSnafu {
+                    function: "json_get_string",
+                    datatypes: columns.iter().map(|c| c.data_type()).collect::<Vec<_>>(),
+                }
+                .fail();
+            }
+        }
+
+        Ok(results.to_vector())
+    }
+}
+
+impl Display for JsonGetString {
+    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
+        write!(f, "{}", "json_get_string".to_ascii_uppercase())
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use std::sync::Arc;
+
+    use common_query::prelude::TypeSignature;
+    use datatypes::scalars::ScalarVector;
+    use datatypes::vectors::{BinaryVector, StringVector};
+
+    use super::*;
+
+    #[test]
+    fn test_json_get_int() {
+        let json_get_int = JsonGetInt;
+
+        assert_eq!("json_get_int", json_get_int.name());
+        assert_eq!(
+            ConcreteDataType::int64_datatype(),
+            json_get_int
+                .return_type(&[
+                    ConcreteDataType::json_datatype(),
+                    ConcreteDataType::string_datatype()
+                ])
+                .unwrap()
+        );
+
+        assert!(matches!(json_get_int.signature(),
+                         Signature {
+                             type_signature: TypeSignature::Exact(valid_types),
+                             volatility: Volatility::Immutable
+                         } if  valid_types == vec![ConcreteDataType::json_datatype(), ConcreteDataType::string_datatype()]
+        ));
+
+        let json_strings = [
+            r#"{"a": {"b": 2}, "b": 2, "c": 3}"#,
+            r#"{"a": 4, "b": {"c": 6}, "c": 6}"#,
+            r#"{"a": 7, "b": 8, "c": {"a": 7}}"#,
+        ];
+        let paths = vec!["$.a.b", "$.a", "$.c"];
+        let results = [Some(2), Some(4), None];
+
+        let jsonbs = json_strings
+            .iter()
+            .map(|s| {
+                let value = jsonb::parse_value(s.as_bytes()).unwrap();
+                value.to_vec()
+            })
+            .collect::<Vec<_>>();
+
+        let json_vector = BinaryVector::from_vec(jsonbs);
+        let path_vector = StringVector::from_vec(paths);
+        let args: Vec<VectorRef> = vec![Arc::new(json_vector), Arc::new(path_vector)];
+        let vector = json_get_int
+            .eval(FunctionContext::default(), &args)
+            .unwrap();
+
+        assert_eq!(3, vector.len());
+        for (i, gt) in results.iter().enumerate() {
+            let result = vector.get_ref(i);
+            let result = result.as_i64().unwrap();
+            assert_eq!(*gt, result);
+        }
+    }
+
+    #[test]
+    fn test_json_get_float() {
+        let json_get_float = JsonGetFloat;
+
+        assert_eq!("json_get_float", json_get_float.name());
+        assert_eq!(
+            ConcreteDataType::float64_datatype(),
+            json_get_float
+                .return_type(&[
+                    ConcreteDataType::json_datatype(),
+                    ConcreteDataType::string_datatype()
+                ])
+                .unwrap()
+        );
+
+        assert!(matches!(json_get_float.signature(),
+                         Signature {
+                             type_signature: TypeSignature::Exact(valid_types),
+                             volatility: Volatility::Immutable
+                         } if  valid_types == vec![ConcreteDataType::json_datatype(), ConcreteDataType::string_datatype()]
+        ));
+
+        let json_strings = [
+            r#"{"a": {"b": 2.1}, "b": 2.2, "c": 3.3}"#,
+            r#"{"a": 4.4, "b": {"c": 6.6}, "c": 6.6}"#,
+            r#"{"a": 7.7, "b": 8.8, "c": {"a": 7.7}}"#,
+        ];
+        let paths = vec!["$.a.b", "$.a", "$.c"];
+        let results = [Some(2.1), Some(4.4), None];
+
+        let jsonbs = json_strings
+            .iter()
+            .map(|s| {
+                let value = jsonb::parse_value(s.as_bytes()).unwrap();
+                value.to_vec()
+            })
+            .collect::<Vec<_>>();
+
+        let json_vector = BinaryVector::from_vec(jsonbs);
+        let path_vector = StringVector::from_vec(paths);
+        let args: Vec<VectorRef> = vec![Arc::new(json_vector), Arc::new(path_vector)];
+        let vector = json_get_float
+            .eval(FunctionContext::default(), &args)
+            .unwrap();
+
+        assert_eq!(3, vector.len());
+        for (i, gt) in results.iter().enumerate() {
+            let result = vector.get_ref(i);
+            let result = result.as_f64().unwrap();
+            assert_eq!(*gt, result);
+        }
+    }
+
+    #[test]
+    fn test_json_get_bool() {
+        let json_get_bool = JsonGetBool;
+
+        assert_eq!("json_get_bool", json_get_bool.name());
+        assert_eq!(
+            ConcreteDataType::boolean_datatype(),
+            json_get_bool
+                .return_type(&[
+                    ConcreteDataType::json_datatype(),
+                    ConcreteDataType::string_datatype()
+                ])
+                .unwrap()
+        );
+
+        assert!(matches!(json_get_bool.signature(),
+                         Signature {
+                             type_signature: TypeSignature::Exact(valid_types),
+                             volatility: Volatility::Immutable
+                         } if  valid_types == vec![ConcreteDataType::json_datatype(), ConcreteDataType::string_datatype()]
+        ));
+
+        let json_strings = [
+            r#"{"a": {"b": true}, "b": false, "c": true}"#,
+            r#"{"a": false, "b": {"c": true}, "c": false}"#,
+            r#"{"a": true, "b": false, "c": {"a": true}}"#,
+        ];
+        let paths = vec!["$.a.b", "$.a", "$.c"];
+        let results = [Some(true), Some(false), None];
+
+        let jsonbs = json_strings
+            .iter()
+            .map(|s| {
+                let value = jsonb::parse_value(s.as_bytes()).unwrap();
+                value.to_vec()
+            })
+            .collect::<Vec<_>>();
+
+        let json_vector = BinaryVector::from_vec(jsonbs);
+        let path_vector = StringVector::from_vec(paths);
+        let args: Vec<VectorRef> = vec![Arc::new(json_vector), Arc::new(path_vector)];
+        let vector = json_get_bool
+            .eval(FunctionContext::default(), &args)
+            .unwrap();
+
+        assert_eq!(3, vector.len());
+        for (i, gt) in results.iter().enumerate() {
+            let result = vector.get_ref(i);
+            let result = result.as_boolean().unwrap();
+            assert_eq!(*gt, result);
+        }
+    }
+
+    #[test]
+    fn test_json_get_string() {
+        let json_get_string = JsonGetString;
+
+        assert_eq!("json_get_string", json_get_string.name());
+        assert_eq!(
+            ConcreteDataType::string_datatype(),
+            json_get_string
+                .return_type(&[
+                    ConcreteDataType::json_datatype(),
+                    ConcreteDataType::string_datatype()
+                ])
+                .unwrap()
+        );
+
+        assert!(matches!(json_get_string.signature(),
+                         Signature {
+                             type_signature: TypeSignature::Exact(valid_types),
+                             volatility: Volatility::Immutable
+                         } if  valid_types == vec![ConcreteDataType::json_datatype(), ConcreteDataType::string_datatype()]
+        ));
+
+        let json_strings = [
+            r#"{"a": {"b": "a"}, "b": "b", "c": "c"}"#,
+            r#"{"a": "d", "b": {"c": "e"}, "c": "f"}"#,
+            r#"{"a": "g", "b": "h", "c": {"a": "g"}}"#,
+        ];
+        let paths = vec!["$.a.b", "$.a", ""];
+        let results = [Some("a"), Some("d"), None];
+
+        let jsonbs = json_strings
+            .iter()
+            .map(|s| {
+                let value = jsonb::parse_value(s.as_bytes()).unwrap();
+                value.to_vec()
+            })
+            .collect::<Vec<_>>();
+
+        let json_vector = BinaryVector::from_vec(jsonbs);
+        let path_vector = StringVector::from_vec(paths);
+        let args: Vec<VectorRef> = vec![Arc::new(json_vector), Arc::new(path_vector)];
+        let vector = json_get_string
+            .eval(FunctionContext::default(), &args)
+            .unwrap();
+
+        assert_eq!(3, vector.len());
+        for (i, gt) in results.iter().enumerate() {
+            let result = vector.get_ref(i);
+            let result = result.as_string().unwrap();
+            assert_eq!(*gt, result);
+        }
+    }
+}

src/common/function/src/scalars/json/json_to_string.rs

@@ -0,0 +1,174 @@
+// 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::fmt::{self, Display};
+
+use common_query::error::{InvalidFuncArgsSnafu, Result, UnsupportedInputDataTypeSnafu};
+use common_query::prelude::Signature;
+use datafusion::logical_expr::Volatility;
+use datatypes::data_type::ConcreteDataType;
+use datatypes::prelude::VectorRef;
+use datatypes::scalars::ScalarVectorBuilder;
+use datatypes::vectors::{MutableVector, StringVectorBuilder};
+use snafu::ensure;
+
+use crate::function::{Function, FunctionContext};
+
+/// Converts the `JSONB` into `String`. It's useful for displaying JSONB content.
+#[derive(Clone, Debug, Default)]
+pub struct JsonToStringFunction;
+
+const NAME: &str = "json_to_string";
+
+impl Function for JsonToStringFunction {
+    fn name(&self) -> &str {
+        NAME
+    }
+
+    fn return_type(&self, _input_types: &[ConcreteDataType]) -> Result<ConcreteDataType> {
+        Ok(ConcreteDataType::string_datatype())
+    }
+
+    fn signature(&self) -> Signature {
+        Signature::exact(
+            vec![ConcreteDataType::json_datatype()],
+            Volatility::Immutable,
+        )
+    }
+
+    fn eval(&self, _func_ctx: FunctionContext, columns: &[VectorRef]) -> Result<VectorRef> {
+        ensure!(
+            columns.len() == 1,
+            InvalidFuncArgsSnafu {
+                err_msg: format!(
+                    "The length of the args is not correct, expect exactly one, have: {}",
+                    columns.len()
+                ),
+            }
+        );
+        let jsons = &columns[0];
+
+        let size = jsons.len();
+        let datatype = jsons.data_type();
+        let mut results = StringVectorBuilder::with_capacity(size);
+
+        match datatype {
+            // JSON data type uses binary vector
+            ConcreteDataType::Binary(_) => {
+                for i in 0..size {
+                    let json = jsons.get_ref(i);
+
+                    let json = json.as_binary();
+                    let result = match json {
+                        Ok(Some(json)) => match jsonb::from_slice(json) {
+                            Ok(json) => {
+                                let json = json.to_string();
+                                Some(json)
+                            }
+                            Err(_) => {
+                                return InvalidFuncArgsSnafu {
+                                    err_msg: format!("Illegal json binary: {:?}", json),
+                                }
+                                .fail()
+                            }
+                        },
+                        _ => None,
+                    };
+
+                    results.push(result.as_deref());
+                }
+            }
+            _ => {
+                return UnsupportedInputDataTypeSnafu {
+                    function: NAME,
+                    datatypes: columns.iter().map(|c| c.data_type()).collect::<Vec<_>>(),
+                }
+                .fail();
+            }
+        }
+
+        Ok(results.to_vector())
+    }
+}
+
+impl Display for JsonToStringFunction {
+    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
+        write!(f, "JSON_TO_STRING")
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use std::sync::Arc;
+
+    use common_query::prelude::TypeSignature;
+    use datatypes::scalars::ScalarVector;
+    use datatypes::vectors::BinaryVector;
+
+    use super::*;
+
+    #[test]
+    fn test_get_by_path_function() {
+        let json_to_string = JsonToStringFunction;
+
+        assert_eq!("json_to_string", json_to_string.name());
+        assert_eq!(
+            ConcreteDataType::string_datatype(),
+            json_to_string
+                .return_type(&[ConcreteDataType::json_datatype()])
+                .unwrap()
+        );
+
+        assert!(matches!(json_to_string.signature(),
+                         Signature {
+                             type_signature: TypeSignature::Exact(valid_types),
+                             volatility: Volatility::Immutable
+                         } if  valid_types == vec![ConcreteDataType::json_datatype()]
+        ));
+
+        let json_strings = [
+            r#"{"a": {"b": 2}, "b": 2, "c": 3}"#,
+            r#"{"a": 4, "b": {"c": 6}, "c": 6}"#,
+            r#"{"a": 7, "b": 8, "c": {"a": 7}}"#,
+        ];
+
+        let jsonbs = json_strings
+            .iter()
+            .map(|s| {
+                let value = jsonb::parse_value(s.as_bytes()).unwrap();
+                value.to_vec()
+            })
+            .collect::<Vec<_>>();
+
+        let json_vector = BinaryVector::from_vec(jsonbs);
+        let args: Vec<VectorRef> = vec![Arc::new(json_vector)];
+        let vector = json_to_string
+            .eval(FunctionContext::default(), &args)
+            .unwrap();
+
+        assert_eq!(3, vector.len());
+        for (i, gt) in json_strings.iter().enumerate() {
+            let result = vector.get_ref(i);
+            let result = result.as_string().unwrap().unwrap();
+            // remove whitespaces
+            assert_eq!(gt.replace(" ", ""), result);
+        }
+
+        let invalid_jsonb = vec![b"invalid json"];
+        let invalid_json_vector = BinaryVector::from_vec(invalid_jsonb);
+        let args: Vec<VectorRef> = vec![Arc::new(invalid_json_vector)];
+        let vector = json_to_string.eval(FunctionContext::default(), &args);
+        assert!(vector.is_err());
+    }
+}

src/common/function/src/scalars/json/to_json.rs

@@ -0,0 +1,165 @@
+// 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::fmt::{self, Display};
+
+use common_query::error::{InvalidFuncArgsSnafu, Result, UnsupportedInputDataTypeSnafu};
+use common_query::prelude::Signature;
+use datafusion::logical_expr::Volatility;
+use datatypes::data_type::ConcreteDataType;
+use datatypes::prelude::VectorRef;
+use datatypes::scalars::ScalarVectorBuilder;
+use datatypes::vectors::{BinaryVectorBuilder, MutableVector};
+use snafu::ensure;
+
+use crate::function::{Function, FunctionContext};
+
+/// Parses the `String` into `JSONB`.
+#[derive(Clone, Debug, Default)]
+pub struct ToJsonFunction;
+
+const NAME: &str = "to_json";
+
+impl Function for ToJsonFunction {
+    fn name(&self) -> &str {
+        NAME
+    }
+
+    fn return_type(&self, _input_types: &[ConcreteDataType]) -> Result<ConcreteDataType> {
+        Ok(ConcreteDataType::json_datatype())
+    }
+
+    fn signature(&self) -> Signature {
+        Signature::exact(
+            vec![ConcreteDataType::string_datatype()],
+            Volatility::Immutable,
+        )
+    }
+
+    fn eval(&self, _func_ctx: FunctionContext, columns: &[VectorRef]) -> Result<VectorRef> {
+        ensure!(
+            columns.len() == 1,
+            InvalidFuncArgsSnafu {
+                err_msg: format!(
+                    "The length of the args is not correct, expect exactly one, have: {}",
+                    columns.len()
+                ),
+            }
+        );
+        let json_strings = &columns[0];
+
+        let size = json_strings.len();
+        let datatype = json_strings.data_type();
+        let mut results = BinaryVectorBuilder::with_capacity(size);
+
+        match datatype {
+            ConcreteDataType::String(_) => {
+                for i in 0..size {
+                    let json_string = json_strings.get_ref(i);
+
+                    let json_string = json_string.as_string();
+                    let result = match json_string {
+                        Ok(Some(json_string)) => match jsonb::parse_value(json_string.as_bytes()) {
+                            Ok(json) => Some(json.to_vec()),
+                            Err(_) => {
+                                return InvalidFuncArgsSnafu {
+                                    err_msg: format!(
+                                        "Cannot convert the string to json, have: {}",
+                                        json_string
+                                    ),
+                                }
+                                .fail()
+                            }
+                        },
+                        _ => None,
+                    };
+
+                    results.push(result.as_deref());
+                }
+            }
+            _ => {
+                return UnsupportedInputDataTypeSnafu {
+                    function: NAME,
+                    datatypes: columns.iter().map(|c| c.data_type()).collect::<Vec<_>>(),
+                }
+                .fail();
+            }
+        }
+
+        Ok(results.to_vector())
+    }
+}
+
+impl Display for ToJsonFunction {
+    fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
+        write!(f, "TO_JSON")
+    }
+}
+
+#[cfg(test)]
+mod tests {
+    use std::sync::Arc;
+
+    use common_query::prelude::TypeSignature;
+    use datatypes::scalars::ScalarVector;
+    use datatypes::vectors::StringVector;
+
+    use super::*;
+
+    #[test]
+    fn test_get_by_path_function() {
+        let to_json = ToJsonFunction;
+
+        assert_eq!("to_json", to_json.name());
+        assert_eq!(
+            ConcreteDataType::json_datatype(),
+            to_json
+                .return_type(&[ConcreteDataType::json_datatype()])
+                .unwrap()
+        );
+
+        assert!(matches!(to_json.signature(),
+                         Signature {
+                             type_signature: TypeSignature::Exact(valid_types),
+                             volatility: Volatility::Immutable
+                         } if  valid_types == vec![ConcreteDataType::string_datatype()]
+        ));
+
+        let json_strings = [
+            r#"{"a": {"b": 2}, "b": 2, "c": 3}"#,
+            r#"{"a": 4, "b": {"c": 6}, "c": 6}"#,
+            r#"{"a": 7, "b": 8, "c": {"a": 7}}"#,
+        ];
+
+        let jsonbs = json_strings
+            .iter()
+            .map(|s| {
+                let value = jsonb::parse_value(s.as_bytes()).unwrap();
+                value.to_vec()
+            })
+            .collect::<Vec<_>>();
+
+        let json_string_vector = StringVector::from_vec(json_strings.to_vec());
+        let args: Vec<VectorRef> = vec![Arc::new(json_string_vector)];
+        let vector = to_json.eval(FunctionContext::default(), &args).unwrap();
+
+        assert_eq!(3, vector.len());
+        for (i, gt) in jsonbs.iter().enumerate() {
+            let result = vector.get_ref(i);
+            let result = result.as_binary().unwrap().unwrap();
+            // remove whitespaces
+            assert_eq!(gt, result);
+        }
+    }
+}

src/common/function/src/state.rs

@@ -28,7 +28,7 @@ pub struct FunctionState {
 
 impl FunctionState {
     /// Create a mock [`FunctionState`] for test.
-    #[cfg(any(test, feature = "testing"))]
+    #[cfg(test)]
     pub fn mock() -> Self {
         use std::sync::Arc;
 

src/common/function/src/system.rs

@@ -22,7 +22,7 @@ mod version;
 use std::sync::Arc;
 
 use build::BuildFunction;
-use database::DatabaseFunction;
+use database::{CurrentSchemaFunction, DatabaseFunction};
 use pg_catalog::PGCatalogFunction;
 use procedure_state::ProcedureStateFunction;
 use timezone::TimezoneFunction;
@@ -37,8 +37,9 @@ impl SystemFunction {
         registry.register(Arc::new(BuildFunction));
         registry.register(Arc::new(VersionFunction));
         registry.register(Arc::new(DatabaseFunction));
+        registry.register(Arc::new(CurrentSchemaFunction));
         registry.register(Arc::new(TimezoneFunction));
-        registry.register(Arc::new(ProcedureStateFunction));
+        registry.register_async(Arc::new(ProcedureStateFunction));
         PGCatalogFunction::register(registry);
     }
 }

src/common/function/src/system/database.rs

@@ -26,11 +26,35 @@ use crate::function::{Function, FunctionContext};
 #[derive(Clone, Debug, Default)]
 pub struct DatabaseFunction;
 
-const NAME: &str = "database";
+#[derive(Clone, Debug, Default)]
+pub struct CurrentSchemaFunction;
+
+const DATABASE_FUNCTION_NAME: &str = "database";
+const CURRENT_SCHEMA_FUNCTION_NAME: &str = "current_schema";
 
 impl Function for DatabaseFunction {
     fn name(&self) -> &str {
-        NAME
+        DATABASE_FUNCTION_NAME
+    }
+
+    fn return_type(&self, _input_types: &[ConcreteDataType]) -> Result<ConcreteDataType> {
+        Ok(ConcreteDataType::string_datatype())
+    }
+
+    fn signature(&self) -> Signature {
+        Signature::uniform(0, vec![], Volatility::Immutable)
+    }
+
+    fn eval(&self, func_ctx: FunctionContext, _columns: &[VectorRef]) -> Result<VectorRef> {
+        let db = func_ctx.query_ctx.current_schema();
+
+        Ok(Arc::new(StringVector::from_slice(&[&db])) as _)
+    }
+}
+
+impl Function for CurrentSchemaFunction {
+    fn name(&self) -> &str {
+        CURRENT_SCHEMA_FUNCTION_NAME
     }
 
     fn return_type(&self, _input_types: &[ConcreteDataType]) -> Result<ConcreteDataType> {
@@ -54,6 +78,12 @@ impl fmt::Display for DatabaseFunction {
     }
 }
 
+impl fmt::Display for CurrentSchemaFunction {
+    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        write!(f, "CURRENT_SCHEMA")
+    }
+}
+
 #[cfg(test)]
 mod tests {
     use std::sync::Arc;