mirror of
https://github.com/GreptimeTeam/greptimedb.git
synced 2025-12-25 15:40:02 +00:00
Compare commits
20 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
|
ed676d97c7 | ||
|
|
14b2badded | ||
|
|
3626a50395 | ||
|
|
0d0dad4ba2 | ||
|
|
ae00e28b2a | ||
|
|
92d2fafb33 | ||
|
|
30b3600597 | ||
|
|
87f1a8c622 | ||
|
|
8e815fc385 | ||
|
|
ca46bd04ee | ||
|
|
d32ade7399 | ||
|
|
b4aa0c8b8b | ||
|
|
e647559d27 | ||
|
|
d2c4767d41 | ||
|
|
82cee11eea | ||
|
|
6d0470c3fb | ||
|
|
47a267e29c | ||
|
|
fa13d06fc6 | ||
|
|
26d9517c3e | ||
|
|
a7da9af5de |
15
.coderabbit.yaml
Normal file
15
.coderabbit.yaml
Normal file
@@ -0,0 +1,15 @@
|
||||
# yaml-language-server: $schema=https://coderabbit.ai/integrations/schema.v2.json
|
||||
language: "en-US"
|
||||
early_access: false
|
||||
reviews:
|
||||
profile: "chill"
|
||||
request_changes_workflow: false
|
||||
high_level_summary: true
|
||||
poem: true
|
||||
review_status: true
|
||||
collapse_walkthrough: false
|
||||
auto_review:
|
||||
enabled: false
|
||||
drafts: false
|
||||
chat:
|
||||
auto_reply: true
|
||||
14
.github/scripts/create-version.sh
vendored
14
.github/scripts/create-version.sh
vendored
@@ -10,17 +10,17 @@ set -e
|
||||
function create_version() {
|
||||
# Read from envrionment variables.
|
||||
if [ -z "$GITHUB_EVENT_NAME" ]; then
|
||||
echo "GITHUB_EVENT_NAME is empty" >&2
|
||||
echo "GITHUB_EVENT_NAME is empty"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
if [ -z "$NEXT_RELEASE_VERSION" ]; then
|
||||
echo "NEXT_RELEASE_VERSION is empty, use version from Cargo.toml" >&2
|
||||
export NEXT_RELEASE_VERSION=$(grep '^version = ' Cargo.toml | cut -d '"' -f 2 | head -n 1)
|
||||
echo "NEXT_RELEASE_VERSION is empty"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
if [ -z "$NIGHTLY_RELEASE_PREFIX" ]; then
|
||||
echo "NIGHTLY_RELEASE_PREFIX is empty" >&2
|
||||
echo "NIGHTLY_RELEASE_PREFIX is empty"
|
||||
exit 1
|
||||
fi
|
||||
|
||||
@@ -35,7 +35,7 @@ function create_version() {
|
||||
# It will be like 'dev-2023080819-f0e7216c'.
|
||||
if [ "$NEXT_RELEASE_VERSION" = dev ]; then
|
||||
if [ -z "$COMMIT_SHA" ]; then
|
||||
echo "COMMIT_SHA is empty in dev build" >&2
|
||||
echo "COMMIT_SHA is empty in dev build"
|
||||
exit 1
|
||||
fi
|
||||
echo "dev-$(date "+%Y%m%d-%s")-$(echo "$COMMIT_SHA" | cut -c1-8)"
|
||||
@@ -45,7 +45,7 @@ function create_version() {
|
||||
# Note: Only output 'version=xxx' to stdout when everything is ok, so that it can be used in GitHub Actions Outputs.
|
||||
if [ "$GITHUB_EVENT_NAME" = push ]; then
|
||||
if [ -z "$GITHUB_REF_NAME" ]; then
|
||||
echo "GITHUB_REF_NAME is empty in push event" >&2
|
||||
echo "GITHUB_REF_NAME is empty in push event"
|
||||
exit 1
|
||||
fi
|
||||
echo "$GITHUB_REF_NAME"
|
||||
@@ -54,7 +54,7 @@ function create_version() {
|
||||
elif [ "$GITHUB_EVENT_NAME" = schedule ]; then
|
||||
echo "$NEXT_RELEASE_VERSION-$NIGHTLY_RELEASE_PREFIX-$(date "+%Y%m%d")"
|
||||
else
|
||||
echo "Unsupported GITHUB_EVENT_NAME: $GITHUB_EVENT_NAME" >&2
|
||||
echo "Unsupported GITHUB_EVENT_NAME: $GITHUB_EVENT_NAME"
|
||||
exit 1
|
||||
fi
|
||||
}
|
||||
|
||||
3
.github/workflows/release.yml
vendored
3
.github/workflows/release.yml
vendored
@@ -90,6 +90,8 @@ env:
|
||||
|
||||
# The scheduled version is '${{ env.NEXT_RELEASE_VERSION }}-nightly-YYYYMMDD', like v0.2.0-nigthly-20230313;
|
||||
NIGHTLY_RELEASE_PREFIX: nightly
|
||||
# Note: The NEXT_RELEASE_VERSION should be modified manually by every formal release.
|
||||
NEXT_RELEASE_VERSION: v0.14.0
|
||||
|
||||
jobs:
|
||||
allocate-runners:
|
||||
@@ -133,6 +135,7 @@ jobs:
|
||||
env:
|
||||
GITHUB_EVENT_NAME: ${{ github.event_name }}
|
||||
GITHUB_REF_NAME: ${{ github.ref_name }}
|
||||
NEXT_RELEASE_VERSION: ${{ env.NEXT_RELEASE_VERSION }}
|
||||
NIGHTLY_RELEASE_PREFIX: ${{ env.NIGHTLY_RELEASE_PREFIX }}
|
||||
|
||||
- name: Allocate linux-amd64 runner
|
||||
|
||||
263
Cargo.lock
generated
263
Cargo.lock
generated
@@ -173,9 +173,9 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "anyhow"
|
||||
version = "1.0.98"
|
||||
version = "1.0.89"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "e16d2d3311acee920a9eb8d33b8cbc1787ce4a264e85f964c2404b969bdcd487"
|
||||
checksum = "86fdf8605db99b54d3cd748a44c6d04df638eb5dafb219b135d0149bd0db01f6"
|
||||
|
||||
[[package]]
|
||||
name = "anymap2"
|
||||
@@ -185,7 +185,7 @@ checksum = "d301b3b94cb4b2f23d7917810addbbaff90738e0ca2be692bd027e70d7e0330c"
|
||||
|
||||
[[package]]
|
||||
name = "api"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"common-base",
|
||||
"common-decimal",
|
||||
@@ -915,7 +915,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "auth"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"api",
|
||||
"async-trait",
|
||||
@@ -1537,7 +1537,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "cache"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"catalog",
|
||||
"common-error",
|
||||
@@ -1561,7 +1561,7 @@ checksum = "37b2a672a2cb129a2e41c10b1224bb368f9f37a2b16b612598138befd7b37eb5"
|
||||
|
||||
[[package]]
|
||||
name = "catalog"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"api",
|
||||
"arrow 54.2.1",
|
||||
@@ -1597,7 +1597,7 @@ dependencies = [
|
||||
"partition",
|
||||
"paste",
|
||||
"prometheus",
|
||||
"rustc-hash 2.1.1",
|
||||
"rustc-hash 2.0.0",
|
||||
"serde_json",
|
||||
"session",
|
||||
"snafu 0.8.5",
|
||||
@@ -1619,9 +1619,9 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "cc"
|
||||
version = "1.2.20"
|
||||
version = "1.1.24"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "04da6a0d40b948dfc4fa8f5bbf402b0fc1a64a28dbf7d12ffd683550f2c1b63a"
|
||||
checksum = "812acba72f0a070b003d3697490d2b55b837230ae7c6c6497f05cc2ddbb8d938"
|
||||
dependencies = [
|
||||
"jobserver",
|
||||
"libc",
|
||||
@@ -1874,7 +1874,7 @@ checksum = "1462739cb27611015575c0c11df5df7601141071f07518d56fcc1be504cbec97"
|
||||
|
||||
[[package]]
|
||||
name = "cli"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"async-trait",
|
||||
"auth",
|
||||
@@ -1917,7 +1917,7 @@ dependencies = [
|
||||
"session",
|
||||
"snafu 0.8.5",
|
||||
"store-api",
|
||||
"substrait 0.15.0",
|
||||
"substrait 0.14.0",
|
||||
"table",
|
||||
"tempfile",
|
||||
"tokio",
|
||||
@@ -1926,7 +1926,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "client"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"api",
|
||||
"arc-swap",
|
||||
@@ -1955,7 +1955,7 @@ dependencies = [
|
||||
"rand 0.9.0",
|
||||
"serde_json",
|
||||
"snafu 0.8.5",
|
||||
"substrait 0.15.0",
|
||||
"substrait 0.14.0",
|
||||
"substrait 0.37.3",
|
||||
"tokio",
|
||||
"tokio-stream",
|
||||
@@ -1996,7 +1996,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "cmd"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"async-trait",
|
||||
"auth",
|
||||
@@ -2056,7 +2056,7 @@ dependencies = [
|
||||
"similar-asserts",
|
||||
"snafu 0.8.5",
|
||||
"store-api",
|
||||
"substrait 0.15.0",
|
||||
"substrait 0.14.0",
|
||||
"table",
|
||||
"temp-env",
|
||||
"tempfile",
|
||||
@@ -2102,7 +2102,7 @@ checksum = "55b672471b4e9f9e95499ea597ff64941a309b2cdbffcc46f2cc5e2d971fd335"
|
||||
|
||||
[[package]]
|
||||
name = "common-base"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"anymap2",
|
||||
"async-trait",
|
||||
@@ -2124,11 +2124,11 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "common-catalog"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
|
||||
[[package]]
|
||||
name = "common-config"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"common-base",
|
||||
"common-error",
|
||||
@@ -2153,7 +2153,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "common-datasource"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"arrow 54.2.1",
|
||||
"arrow-schema 54.3.1",
|
||||
@@ -2190,7 +2190,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "common-decimal"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"bigdecimal 0.4.8",
|
||||
"common-error",
|
||||
@@ -2203,7 +2203,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "common-error"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"common-macro",
|
||||
"http 1.1.0",
|
||||
@@ -2214,7 +2214,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "common-frontend"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"async-trait",
|
||||
"common-error",
|
||||
@@ -2224,7 +2224,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "common-function"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"ahash 0.8.11",
|
||||
"api",
|
||||
@@ -2277,7 +2277,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "common-greptimedb-telemetry"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"async-trait",
|
||||
"common-runtime",
|
||||
@@ -2294,7 +2294,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "common-grpc"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"api",
|
||||
"arrow-flight",
|
||||
@@ -2325,7 +2325,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "common-grpc-expr"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"api",
|
||||
"common-base",
|
||||
@@ -2344,7 +2344,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "common-macro"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"arc-swap",
|
||||
"common-query",
|
||||
@@ -2358,7 +2358,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "common-mem-prof"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"common-error",
|
||||
"common-macro",
|
||||
@@ -2371,7 +2371,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "common-meta"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"anymap2",
|
||||
"api",
|
||||
@@ -2432,7 +2432,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "common-options"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"common-grpc",
|
||||
"humantime-serde",
|
||||
@@ -2441,11 +2441,11 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "common-plugins"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
|
||||
[[package]]
|
||||
name = "common-pprof"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"common-error",
|
||||
"common-macro",
|
||||
@@ -2457,7 +2457,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "common-procedure"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"async-stream",
|
||||
"async-trait",
|
||||
@@ -2484,7 +2484,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "common-procedure-test"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"async-trait",
|
||||
"common-procedure",
|
||||
@@ -2493,7 +2493,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "common-query"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"api",
|
||||
"async-trait",
|
||||
@@ -2510,7 +2510,7 @@ dependencies = [
|
||||
"futures-util",
|
||||
"serde",
|
||||
"snafu 0.8.5",
|
||||
"sqlparser 0.54.0 (git+https://github.com/GreptimeTeam/sqlparser-rs.git?rev=0cf6c04490d59435ee965edd2078e8855bd8471e)",
|
||||
"sqlparser 0.54.0 (git+https://github.com/GreptimeTeam/sqlparser-rs.git?rev=e98e6b322426a9d397a71efef17075966223c089)",
|
||||
"sqlparser_derive 0.1.1",
|
||||
"statrs",
|
||||
"store-api",
|
||||
@@ -2519,7 +2519,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "common-recordbatch"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"arc-swap",
|
||||
"common-error",
|
||||
@@ -2539,7 +2539,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "common-runtime"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"async-trait",
|
||||
"clap 4.5.19",
|
||||
@@ -2569,14 +2569,14 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "common-session"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"strum 0.27.1",
|
||||
]
|
||||
|
||||
[[package]]
|
||||
name = "common-telemetry"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"atty",
|
||||
"backtrace",
|
||||
@@ -2604,7 +2604,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "common-test-util"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"client",
|
||||
"common-query",
|
||||
@@ -2616,7 +2616,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "common-time"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"arrow 54.2.1",
|
||||
"chrono",
|
||||
@@ -2634,7 +2634,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "common-version"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"build-data",
|
||||
"const_format",
|
||||
@@ -2644,7 +2644,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "common-wal"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"common-base",
|
||||
"common-error",
|
||||
@@ -2946,9 +2946,9 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "crossbeam-channel"
|
||||
version = "0.5.15"
|
||||
version = "0.5.13"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "82b8f8f868b36967f9606790d1903570de9ceaf870a7bf9fbbd3016d636a2cb2"
|
||||
checksum = "33480d6946193aa8033910124896ca395333cae7e2d1113d1fef6c3272217df2"
|
||||
dependencies = [
|
||||
"crossbeam-utils",
|
||||
]
|
||||
@@ -3110,14 +3110,14 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "data-encoding"
|
||||
version = "2.9.0"
|
||||
version = "2.6.0"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "2a2330da5de22e8a3cb63252ce2abb30116bf5265e89c0e01bc17015ce30a476"
|
||||
checksum = "e8566979429cf69b49a5c740c60791108e86440e8be149bbea4fe54d2c32d6e2"
|
||||
|
||||
[[package]]
|
||||
name = "datafusion"
|
||||
version = "45.0.0"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=5bbedc6704162afb03478f56ffb629405a4e1220#5bbedc6704162afb03478f56ffb629405a4e1220"
|
||||
dependencies = [
|
||||
"arrow 54.2.1",
|
||||
"arrow-array 54.2.1",
|
||||
@@ -3168,7 +3168,7 @@ dependencies = [
|
||||
[[package]]
|
||||
name = "datafusion-catalog"
|
||||
version = "45.0.0"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=5bbedc6704162afb03478f56ffb629405a4e1220#5bbedc6704162afb03478f56ffb629405a4e1220"
|
||||
dependencies = [
|
||||
"arrow 54.2.1",
|
||||
"async-trait",
|
||||
@@ -3188,7 +3188,7 @@ dependencies = [
|
||||
[[package]]
|
||||
name = "datafusion-catalog-listing"
|
||||
version = "45.0.0"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=5bbedc6704162afb03478f56ffb629405a4e1220#5bbedc6704162afb03478f56ffb629405a4e1220"
|
||||
dependencies = [
|
||||
"arrow 54.2.1",
|
||||
"arrow-schema 54.3.1",
|
||||
@@ -3211,7 +3211,7 @@ dependencies = [
|
||||
[[package]]
|
||||
name = "datafusion-common"
|
||||
version = "45.0.0"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=5bbedc6704162afb03478f56ffb629405a4e1220#5bbedc6704162afb03478f56ffb629405a4e1220"
|
||||
dependencies = [
|
||||
"ahash 0.8.11",
|
||||
"arrow 54.2.1",
|
||||
@@ -3236,7 +3236,7 @@ dependencies = [
|
||||
[[package]]
|
||||
name = "datafusion-common-runtime"
|
||||
version = "45.0.0"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=5bbedc6704162afb03478f56ffb629405a4e1220#5bbedc6704162afb03478f56ffb629405a4e1220"
|
||||
dependencies = [
|
||||
"log",
|
||||
"tokio",
|
||||
@@ -3245,12 +3245,12 @@ dependencies = [
|
||||
[[package]]
|
||||
name = "datafusion-doc"
|
||||
version = "45.0.0"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=5bbedc6704162afb03478f56ffb629405a4e1220#5bbedc6704162afb03478f56ffb629405a4e1220"
|
||||
|
||||
[[package]]
|
||||
name = "datafusion-execution"
|
||||
version = "45.0.0"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=5bbedc6704162afb03478f56ffb629405a4e1220#5bbedc6704162afb03478f56ffb629405a4e1220"
|
||||
dependencies = [
|
||||
"arrow 54.2.1",
|
||||
"dashmap",
|
||||
@@ -3268,7 +3268,7 @@ dependencies = [
|
||||
[[package]]
|
||||
name = "datafusion-expr"
|
||||
version = "45.0.0"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=5bbedc6704162afb03478f56ffb629405a4e1220#5bbedc6704162afb03478f56ffb629405a4e1220"
|
||||
dependencies = [
|
||||
"arrow 54.2.1",
|
||||
"chrono",
|
||||
@@ -3288,7 +3288,7 @@ dependencies = [
|
||||
[[package]]
|
||||
name = "datafusion-expr-common"
|
||||
version = "45.0.0"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=5bbedc6704162afb03478f56ffb629405a4e1220#5bbedc6704162afb03478f56ffb629405a4e1220"
|
||||
dependencies = [
|
||||
"arrow 54.2.1",
|
||||
"datafusion-common",
|
||||
@@ -3299,7 +3299,7 @@ dependencies = [
|
||||
[[package]]
|
||||
name = "datafusion-functions"
|
||||
version = "45.0.0"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=5bbedc6704162afb03478f56ffb629405a4e1220#5bbedc6704162afb03478f56ffb629405a4e1220"
|
||||
dependencies = [
|
||||
"arrow 54.2.1",
|
||||
"arrow-buffer 54.3.1",
|
||||
@@ -3328,7 +3328,7 @@ dependencies = [
|
||||
[[package]]
|
||||
name = "datafusion-functions-aggregate"
|
||||
version = "45.0.0"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=5bbedc6704162afb03478f56ffb629405a4e1220#5bbedc6704162afb03478f56ffb629405a4e1220"
|
||||
dependencies = [
|
||||
"ahash 0.8.11",
|
||||
"arrow 54.2.1",
|
||||
@@ -3349,7 +3349,7 @@ dependencies = [
|
||||
[[package]]
|
||||
name = "datafusion-functions-aggregate-common"
|
||||
version = "45.0.0"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=5bbedc6704162afb03478f56ffb629405a4e1220#5bbedc6704162afb03478f56ffb629405a4e1220"
|
||||
dependencies = [
|
||||
"ahash 0.8.11",
|
||||
"arrow 54.2.1",
|
||||
@@ -3361,7 +3361,7 @@ dependencies = [
|
||||
[[package]]
|
||||
name = "datafusion-functions-nested"
|
||||
version = "45.0.0"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=5bbedc6704162afb03478f56ffb629405a4e1220#5bbedc6704162afb03478f56ffb629405a4e1220"
|
||||
dependencies = [
|
||||
"arrow 54.2.1",
|
||||
"arrow-array 54.2.1",
|
||||
@@ -3383,7 +3383,7 @@ dependencies = [
|
||||
[[package]]
|
||||
name = "datafusion-functions-table"
|
||||
version = "45.0.0"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=5bbedc6704162afb03478f56ffb629405a4e1220#5bbedc6704162afb03478f56ffb629405a4e1220"
|
||||
dependencies = [
|
||||
"arrow 54.2.1",
|
||||
"async-trait",
|
||||
@@ -3398,7 +3398,7 @@ dependencies = [
|
||||
[[package]]
|
||||
name = "datafusion-functions-window"
|
||||
version = "45.0.0"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=5bbedc6704162afb03478f56ffb629405a4e1220#5bbedc6704162afb03478f56ffb629405a4e1220"
|
||||
dependencies = [
|
||||
"datafusion-common",
|
||||
"datafusion-doc",
|
||||
@@ -3414,7 +3414,7 @@ dependencies = [
|
||||
[[package]]
|
||||
name = "datafusion-functions-window-common"
|
||||
version = "45.0.0"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=5bbedc6704162afb03478f56ffb629405a4e1220#5bbedc6704162afb03478f56ffb629405a4e1220"
|
||||
dependencies = [
|
||||
"datafusion-common",
|
||||
"datafusion-physical-expr-common",
|
||||
@@ -3423,7 +3423,7 @@ dependencies = [
|
||||
[[package]]
|
||||
name = "datafusion-macros"
|
||||
version = "45.0.0"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=5bbedc6704162afb03478f56ffb629405a4e1220#5bbedc6704162afb03478f56ffb629405a4e1220"
|
||||
dependencies = [
|
||||
"datafusion-expr",
|
||||
"quote",
|
||||
@@ -3433,7 +3433,7 @@ dependencies = [
|
||||
[[package]]
|
||||
name = "datafusion-optimizer"
|
||||
version = "45.0.0"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=5bbedc6704162afb03478f56ffb629405a4e1220#5bbedc6704162afb03478f56ffb629405a4e1220"
|
||||
dependencies = [
|
||||
"arrow 54.2.1",
|
||||
"chrono",
|
||||
@@ -3451,7 +3451,7 @@ dependencies = [
|
||||
[[package]]
|
||||
name = "datafusion-physical-expr"
|
||||
version = "45.0.0"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=5bbedc6704162afb03478f56ffb629405a4e1220#5bbedc6704162afb03478f56ffb629405a4e1220"
|
||||
dependencies = [
|
||||
"ahash 0.8.11",
|
||||
"arrow 54.2.1",
|
||||
@@ -3474,7 +3474,7 @@ dependencies = [
|
||||
[[package]]
|
||||
name = "datafusion-physical-expr-common"
|
||||
version = "45.0.0"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=5bbedc6704162afb03478f56ffb629405a4e1220#5bbedc6704162afb03478f56ffb629405a4e1220"
|
||||
dependencies = [
|
||||
"ahash 0.8.11",
|
||||
"arrow 54.2.1",
|
||||
@@ -3487,7 +3487,7 @@ dependencies = [
|
||||
[[package]]
|
||||
name = "datafusion-physical-optimizer"
|
||||
version = "45.0.0"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=5bbedc6704162afb03478f56ffb629405a4e1220#5bbedc6704162afb03478f56ffb629405a4e1220"
|
||||
dependencies = [
|
||||
"arrow 54.2.1",
|
||||
"arrow-schema 54.3.1",
|
||||
@@ -3508,7 +3508,7 @@ dependencies = [
|
||||
[[package]]
|
||||
name = "datafusion-physical-plan"
|
||||
version = "45.0.0"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=5bbedc6704162afb03478f56ffb629405a4e1220#5bbedc6704162afb03478f56ffb629405a4e1220"
|
||||
dependencies = [
|
||||
"ahash 0.8.11",
|
||||
"arrow 54.2.1",
|
||||
@@ -3538,7 +3538,7 @@ dependencies = [
|
||||
[[package]]
|
||||
name = "datafusion-sql"
|
||||
version = "45.0.0"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=5bbedc6704162afb03478f56ffb629405a4e1220#5bbedc6704162afb03478f56ffb629405a4e1220"
|
||||
dependencies = [
|
||||
"arrow 54.2.1",
|
||||
"arrow-array 54.2.1",
|
||||
@@ -3556,7 +3556,7 @@ dependencies = [
|
||||
[[package]]
|
||||
name = "datafusion-substrait"
|
||||
version = "45.0.0"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=e104c7cf62b11dd5fe41461b82514978234326b4#e104c7cf62b11dd5fe41461b82514978234326b4"
|
||||
source = "git+https://github.com/waynexia/arrow-datafusion.git?rev=5bbedc6704162afb03478f56ffb629405a4e1220#5bbedc6704162afb03478f56ffb629405a4e1220"
|
||||
dependencies = [
|
||||
"async-recursion",
|
||||
"async-trait",
|
||||
@@ -3572,7 +3572,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "datanode"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"api",
|
||||
"arrow-flight",
|
||||
@@ -3624,7 +3624,7 @@ dependencies = [
|
||||
"session",
|
||||
"snafu 0.8.5",
|
||||
"store-api",
|
||||
"substrait 0.15.0",
|
||||
"substrait 0.14.0",
|
||||
"table",
|
||||
"tokio",
|
||||
"toml 0.8.19",
|
||||
@@ -3633,7 +3633,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "datatypes"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"arrow 54.2.1",
|
||||
"arrow-array 54.2.1",
|
||||
@@ -3656,7 +3656,7 @@ dependencies = [
|
||||
"serde",
|
||||
"serde_json",
|
||||
"snafu 0.8.5",
|
||||
"sqlparser 0.54.0 (git+https://github.com/GreptimeTeam/sqlparser-rs.git?rev=0cf6c04490d59435ee965edd2078e8855bd8471e)",
|
||||
"sqlparser 0.54.0 (git+https://github.com/GreptimeTeam/sqlparser-rs.git?rev=e98e6b322426a9d397a71efef17075966223c089)",
|
||||
"sqlparser_derive 0.1.1",
|
||||
]
|
||||
|
||||
@@ -4259,7 +4259,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "file-engine"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"api",
|
||||
"async-trait",
|
||||
@@ -4382,7 +4382,7 @@ checksum = "8bf7cc16383c4b8d58b9905a8509f02926ce3058053c056376248d958c9df1e8"
|
||||
|
||||
[[package]]
|
||||
name = "flow"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"api",
|
||||
"arrow 54.2.1",
|
||||
@@ -4444,7 +4444,7 @@ dependencies = [
|
||||
"snafu 0.8.5",
|
||||
"store-api",
|
||||
"strum 0.27.1",
|
||||
"substrait 0.15.0",
|
||||
"substrait 0.14.0",
|
||||
"table",
|
||||
"tokio",
|
||||
"tonic 0.12.3",
|
||||
@@ -4499,7 +4499,7 @@ checksum = "6c2141d6d6c8512188a7891b4b01590a45f6dac67afb4f255c4124dbb86d4eaa"
|
||||
|
||||
[[package]]
|
||||
name = "frontend"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"api",
|
||||
"arc-swap",
|
||||
@@ -4553,10 +4553,10 @@ dependencies = [
|
||||
"session",
|
||||
"snafu 0.8.5",
|
||||
"sql",
|
||||
"sqlparser 0.54.0 (git+https://github.com/GreptimeTeam/sqlparser-rs.git?rev=0cf6c04490d59435ee965edd2078e8855bd8471e)",
|
||||
"sqlparser 0.54.0 (git+https://github.com/GreptimeTeam/sqlparser-rs.git?rev=e98e6b322426a9d397a71efef17075966223c089)",
|
||||
"store-api",
|
||||
"strfmt",
|
||||
"substrait 0.15.0",
|
||||
"substrait 0.14.0",
|
||||
"table",
|
||||
"tokio",
|
||||
"toml 0.8.19",
|
||||
@@ -5795,7 +5795,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "index"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"async-trait",
|
||||
"asynchronous-codec",
|
||||
@@ -6599,13 +6599,13 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "log"
|
||||
version = "0.4.27"
|
||||
version = "0.4.22"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "13dc2df351e3202783a1fe0d44375f7295ffb4049267b0f3018346dc122a1d94"
|
||||
checksum = "a7a70ba024b9dc04c27ea2f0c0548feb474ec5c54bba33a7f72f873a39d07b24"
|
||||
|
||||
[[package]]
|
||||
name = "log-query"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"chrono",
|
||||
"common-error",
|
||||
@@ -6617,7 +6617,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "log-store"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"async-stream",
|
||||
"async-trait",
|
||||
@@ -6911,7 +6911,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "meta-client"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"api",
|
||||
"async-trait",
|
||||
@@ -6939,7 +6939,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "meta-srv"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"api",
|
||||
"async-trait",
|
||||
@@ -7029,7 +7029,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "metric-engine"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"api",
|
||||
"aquamarine",
|
||||
@@ -7041,14 +7041,12 @@ dependencies = [
|
||||
"common-macro",
|
||||
"common-query",
|
||||
"common-recordbatch",
|
||||
"common-runtime",
|
||||
"common-telemetry",
|
||||
"common-test-util",
|
||||
"common-time",
|
||||
"datafusion",
|
||||
"datatypes",
|
||||
"futures-util",
|
||||
"humantime-serde",
|
||||
"itertools 0.14.0",
|
||||
"lazy_static",
|
||||
"mito2",
|
||||
@@ -7118,7 +7116,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "mito2"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"api",
|
||||
"aquamarine",
|
||||
@@ -7780,7 +7778,7 @@ version = "0.7.3"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "af1844ef2428cc3e1cb900be36181049ef3d3193c63e43026cfe202983b27a56"
|
||||
dependencies = [
|
||||
"proc-macro-crate 3.2.0",
|
||||
"proc-macro-crate 1.3.1",
|
||||
"proc-macro2",
|
||||
"quote",
|
||||
"syn 2.0.100",
|
||||
@@ -7824,7 +7822,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "object-store"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"anyhow",
|
||||
"bytes",
|
||||
@@ -8119,7 +8117,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "operator"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"ahash 0.8.11",
|
||||
"api",
|
||||
@@ -8166,9 +8164,9 @@ dependencies = [
|
||||
"session",
|
||||
"snafu 0.8.5",
|
||||
"sql",
|
||||
"sqlparser 0.54.0 (git+https://github.com/GreptimeTeam/sqlparser-rs.git?rev=0cf6c04490d59435ee965edd2078e8855bd8471e)",
|
||||
"sqlparser 0.54.0 (git+https://github.com/GreptimeTeam/sqlparser-rs.git?rev=e98e6b322426a9d397a71efef17075966223c089)",
|
||||
"store-api",
|
||||
"substrait 0.15.0",
|
||||
"substrait 0.14.0",
|
||||
"table",
|
||||
"tokio",
|
||||
"tokio-util",
|
||||
@@ -8423,7 +8421,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "partition"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"api",
|
||||
"async-trait",
|
||||
@@ -8443,7 +8441,7 @@ dependencies = [
|
||||
"session",
|
||||
"snafu 0.8.5",
|
||||
"sql",
|
||||
"sqlparser 0.54.0 (git+https://github.com/GreptimeTeam/sqlparser-rs.git?rev=0cf6c04490d59435ee965edd2078e8855bd8471e)",
|
||||
"sqlparser 0.54.0 (git+https://github.com/GreptimeTeam/sqlparser-rs.git?rev=e98e6b322426a9d397a71efef17075966223c089)",
|
||||
"store-api",
|
||||
"table",
|
||||
]
|
||||
@@ -8705,7 +8703,7 @@ checksum = "8b870d8c151b6f2fb93e84a13146138f05d02ed11c7e7c54f8826aaaf7c9f184"
|
||||
|
||||
[[package]]
|
||||
name = "pipeline"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"ahash 0.8.11",
|
||||
"api",
|
||||
@@ -8847,7 +8845,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "plugins"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"auth",
|
||||
"clap 4.5.19",
|
||||
@@ -9127,7 +9125,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "promql"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"ahash 0.8.11",
|
||||
"async-trait",
|
||||
@@ -9373,7 +9371,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "puffin"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"async-compression 0.4.13",
|
||||
"async-trait",
|
||||
@@ -9414,7 +9412,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "query"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"ahash 0.8.11",
|
||||
"api",
|
||||
@@ -9477,10 +9475,10 @@ dependencies = [
|
||||
"session",
|
||||
"snafu 0.8.5",
|
||||
"sql",
|
||||
"sqlparser 0.54.0 (git+https://github.com/GreptimeTeam/sqlparser-rs.git?rev=0cf6c04490d59435ee965edd2078e8855bd8471e)",
|
||||
"sqlparser 0.54.0 (git+https://github.com/GreptimeTeam/sqlparser-rs.git?rev=e98e6b322426a9d397a71efef17075966223c089)",
|
||||
"statrs",
|
||||
"store-api",
|
||||
"substrait 0.15.0",
|
||||
"substrait 0.14.0",
|
||||
"table",
|
||||
"tokio",
|
||||
"tokio-stream",
|
||||
@@ -9527,7 +9525,7 @@ dependencies = [
|
||||
"pin-project-lite",
|
||||
"quinn-proto",
|
||||
"quinn-udp",
|
||||
"rustc-hash 2.1.1",
|
||||
"rustc-hash 2.0.0",
|
||||
"rustls",
|
||||
"socket2",
|
||||
"thiserror 1.0.64",
|
||||
@@ -9544,7 +9542,7 @@ dependencies = [
|
||||
"bytes",
|
||||
"rand 0.8.5",
|
||||
"ring",
|
||||
"rustc-hash 2.1.1",
|
||||
"rustc-hash 2.0.0",
|
||||
"rustls",
|
||||
"slab",
|
||||
"thiserror 1.0.64",
|
||||
@@ -9821,9 +9819,9 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "regex"
|
||||
version = "1.11.1"
|
||||
version = "1.11.0"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "b544ef1b4eac5dc2db33ea63606ae9ffcfac26c1416a2806ae0bf5f56b201191"
|
||||
checksum = "38200e5ee88914975b69f657f0801b6f6dccafd44fd9326302a4aaeecfacb1d8"
|
||||
dependencies = [
|
||||
"aho-corasick",
|
||||
"memchr",
|
||||
@@ -10005,14 +10003,15 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "ring"
|
||||
version = "0.17.14"
|
||||
version = "0.17.8"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "a4689e6c2294d81e88dc6261c768b63bc4fcdb852be6d1352498b114f61383b7"
|
||||
checksum = "c17fa4cb658e3583423e915b9f3acc01cceaee1860e33d59ebae66adc3a2dc0d"
|
||||
dependencies = [
|
||||
"cc",
|
||||
"cfg-if",
|
||||
"getrandom 0.2.15",
|
||||
"libc",
|
||||
"spin",
|
||||
"untrusted",
|
||||
"windows-sys 0.52.0",
|
||||
]
|
||||
@@ -10333,9 +10332,9 @@ checksum = "08d43f7aa6b08d49f382cde6a7982047c3426db949b1424bc4b7ec9ae12c6ce2"
|
||||
|
||||
[[package]]
|
||||
name = "rustc-hash"
|
||||
version = "2.1.1"
|
||||
version = "2.0.0"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "357703d41365b4b27c590e3ed91eabb1b663f07c4c084095e60cbed4362dff0d"
|
||||
checksum = "583034fd73374156e66797ed8e5b0d5690409c9226b22d87cb7f19821c05d152"
|
||||
|
||||
[[package]]
|
||||
name = "rustc_version"
|
||||
@@ -10830,7 +10829,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "servers"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"ahash 0.8.11",
|
||||
"api",
|
||||
@@ -10950,7 +10949,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "session"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"api",
|
||||
"arc-swap",
|
||||
@@ -11158,9 +11157,9 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "smallbitvec"
|
||||
version = "2.6.0"
|
||||
version = "2.5.3"
|
||||
source = "registry+https://github.com/rust-lang/crates.io-index"
|
||||
checksum = "d31d263dd118560e1a492922182ab6ca6dc1d03a3bf54e7699993f31a4150e3f"
|
||||
checksum = "fcc3fc564a4b53fd1e8589628efafe57602d91bde78be18186b5f61e8faea470"
|
||||
|
||||
[[package]]
|
||||
name = "smallvec"
|
||||
@@ -11275,7 +11274,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "sql"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"api",
|
||||
"chrono",
|
||||
@@ -11303,7 +11302,7 @@ dependencies = [
|
||||
"serde",
|
||||
"serde_json",
|
||||
"snafu 0.8.5",
|
||||
"sqlparser 0.54.0 (git+https://github.com/GreptimeTeam/sqlparser-rs.git?rev=0cf6c04490d59435ee965edd2078e8855bd8471e)",
|
||||
"sqlparser 0.54.0 (git+https://github.com/GreptimeTeam/sqlparser-rs.git?rev=e98e6b322426a9d397a71efef17075966223c089)",
|
||||
"sqlparser_derive 0.1.1",
|
||||
"store-api",
|
||||
"table",
|
||||
@@ -11330,7 +11329,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "sqlness-runner"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"async-trait",
|
||||
"clap 4.5.19",
|
||||
@@ -11372,7 +11371,7 @@ dependencies = [
|
||||
[[package]]
|
||||
name = "sqlparser"
|
||||
version = "0.54.0"
|
||||
source = "git+https://github.com/GreptimeTeam/sqlparser-rs.git?rev=0cf6c04490d59435ee965edd2078e8855bd8471e#0cf6c04490d59435ee965edd2078e8855bd8471e"
|
||||
source = "git+https://github.com/GreptimeTeam/sqlparser-rs.git?rev=e98e6b322426a9d397a71efef17075966223c089#e98e6b322426a9d397a71efef17075966223c089"
|
||||
dependencies = [
|
||||
"lazy_static",
|
||||
"log",
|
||||
@@ -11380,7 +11379,7 @@ dependencies = [
|
||||
"regex",
|
||||
"serde",
|
||||
"sqlparser 0.54.0 (registry+https://github.com/rust-lang/crates.io-index)",
|
||||
"sqlparser_derive 0.3.0 (git+https://github.com/GreptimeTeam/sqlparser-rs.git?rev=0cf6c04490d59435ee965edd2078e8855bd8471e)",
|
||||
"sqlparser_derive 0.3.0 (git+https://github.com/GreptimeTeam/sqlparser-rs.git?rev=e98e6b322426a9d397a71efef17075966223c089)",
|
||||
]
|
||||
|
||||
[[package]]
|
||||
@@ -11408,7 +11407,7 @@ dependencies = [
|
||||
[[package]]
|
||||
name = "sqlparser_derive"
|
||||
version = "0.3.0"
|
||||
source = "git+https://github.com/GreptimeTeam/sqlparser-rs.git?rev=0cf6c04490d59435ee965edd2078e8855bd8471e#0cf6c04490d59435ee965edd2078e8855bd8471e"
|
||||
source = "git+https://github.com/GreptimeTeam/sqlparser-rs.git?rev=e98e6b322426a9d397a71efef17075966223c089#e98e6b322426a9d397a71efef17075966223c089"
|
||||
dependencies = [
|
||||
"proc-macro2",
|
||||
"quote",
|
||||
@@ -11649,7 +11648,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "store-api"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"api",
|
||||
"aquamarine",
|
||||
@@ -11798,7 +11797,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "substrait"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"async-trait",
|
||||
"bytes",
|
||||
@@ -11978,7 +11977,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "table"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"api",
|
||||
"async-trait",
|
||||
@@ -12229,7 +12228,7 @@ checksum = "3369f5ac52d5eb6ab48c6b4ffdc8efbcad6b89c765749064ba298f2c68a16a76"
|
||||
|
||||
[[package]]
|
||||
name = "tests-fuzz"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"arbitrary",
|
||||
"async-trait",
|
||||
@@ -12263,7 +12262,7 @@ dependencies = [
|
||||
"serde_yaml",
|
||||
"snafu 0.8.5",
|
||||
"sql",
|
||||
"sqlparser 0.54.0 (git+https://github.com/GreptimeTeam/sqlparser-rs.git?rev=0cf6c04490d59435ee965edd2078e8855bd8471e)",
|
||||
"sqlparser 0.54.0 (git+https://github.com/GreptimeTeam/sqlparser-rs.git?rev=e98e6b322426a9d397a71efef17075966223c089)",
|
||||
"sqlx",
|
||||
"store-api",
|
||||
"strum 0.27.1",
|
||||
@@ -12273,7 +12272,7 @@ dependencies = [
|
||||
|
||||
[[package]]
|
||||
name = "tests-integration"
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
dependencies = [
|
||||
"api",
|
||||
"arrow-flight",
|
||||
@@ -12340,7 +12339,7 @@ dependencies = [
|
||||
"sql",
|
||||
"sqlx",
|
||||
"store-api",
|
||||
"substrait 0.15.0",
|
||||
"substrait 0.14.0",
|
||||
"table",
|
||||
"tempfile",
|
||||
"time",
|
||||
|
||||
29
Cargo.toml
29
Cargo.toml
@@ -68,16 +68,15 @@ members = [
|
||||
resolver = "2"
|
||||
|
||||
[workspace.package]
|
||||
version = "0.15.0"
|
||||
version = "0.14.0"
|
||||
edition = "2021"
|
||||
license = "Apache-2.0"
|
||||
|
||||
[workspace.lints]
|
||||
clippy.print_stdout = "warn"
|
||||
clippy.print_stderr = "warn"
|
||||
clippy.dbg_macro = "warn"
|
||||
clippy.implicit_clone = "warn"
|
||||
clippy.result_large_err = "allow"
|
||||
clippy.large_enum_variant = "allow"
|
||||
clippy.doc_overindented_list_items = "allow"
|
||||
rust.unknown_lints = "deny"
|
||||
rust.unexpected_cfgs = { level = "warn", check-cfg = ['cfg(tokio_unstable)'] }
|
||||
|
||||
@@ -113,15 +112,15 @@ clap = { version = "4.4", features = ["derive"] }
|
||||
config = "0.13.0"
|
||||
crossbeam-utils = "0.8"
|
||||
dashmap = "6.1"
|
||||
datafusion = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "e104c7cf62b11dd5fe41461b82514978234326b4" }
|
||||
datafusion-common = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "e104c7cf62b11dd5fe41461b82514978234326b4" }
|
||||
datafusion-expr = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "e104c7cf62b11dd5fe41461b82514978234326b4" }
|
||||
datafusion-functions = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "e104c7cf62b11dd5fe41461b82514978234326b4" }
|
||||
datafusion-optimizer = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "e104c7cf62b11dd5fe41461b82514978234326b4" }
|
||||
datafusion-physical-expr = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "e104c7cf62b11dd5fe41461b82514978234326b4" }
|
||||
datafusion-physical-plan = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "e104c7cf62b11dd5fe41461b82514978234326b4" }
|
||||
datafusion-sql = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "e104c7cf62b11dd5fe41461b82514978234326b4" }
|
||||
datafusion-substrait = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "e104c7cf62b11dd5fe41461b82514978234326b4" }
|
||||
datafusion = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "5bbedc6704162afb03478f56ffb629405a4e1220" }
|
||||
datafusion-common = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "5bbedc6704162afb03478f56ffb629405a4e1220" }
|
||||
datafusion-expr = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "5bbedc6704162afb03478f56ffb629405a4e1220" }
|
||||
datafusion-functions = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "5bbedc6704162afb03478f56ffb629405a4e1220" }
|
||||
datafusion-optimizer = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "5bbedc6704162afb03478f56ffb629405a4e1220" }
|
||||
datafusion-physical-expr = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "5bbedc6704162afb03478f56ffb629405a4e1220" }
|
||||
datafusion-physical-plan = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "5bbedc6704162afb03478f56ffb629405a4e1220" }
|
||||
datafusion-sql = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "5bbedc6704162afb03478f56ffb629405a4e1220" }
|
||||
datafusion-substrait = { git = "https://github.com/waynexia/arrow-datafusion.git", rev = "5bbedc6704162afb03478f56ffb629405a4e1220" }
|
||||
deadpool = "0.12"
|
||||
deadpool-postgres = "0.14"
|
||||
derive_builder = "0.20"
|
||||
@@ -163,7 +162,7 @@ paste = "1.0"
|
||||
pin-project = "1.0"
|
||||
prometheus = { version = "0.13.3", features = ["process"] }
|
||||
promql-parser = { version = "0.5.1", features = ["ser"] }
|
||||
prost = { version = "0.13", features = ["no-recursion-limit"] }
|
||||
prost = "0.13"
|
||||
raft-engine = { version = "0.4.1", default-features = false }
|
||||
rand = "0.9"
|
||||
ratelimit = "0.10"
|
||||
@@ -192,7 +191,7 @@ simd-json = "0.15"
|
||||
similar-asserts = "1.6.0"
|
||||
smallvec = { version = "1", features = ["serde"] }
|
||||
snafu = "0.8"
|
||||
sqlparser = { git = "https://github.com/GreptimeTeam/sqlparser-rs.git", rev = "0cf6c04490d59435ee965edd2078e8855bd8471e", features = [
|
||||
sqlparser = { git = "https://github.com/GreptimeTeam/sqlparser-rs.git", rev = "e98e6b322426a9d397a71efef17075966223c089", features = [
|
||||
"visitor",
|
||||
"serde",
|
||||
] } # branch = "v0.54.x"
|
||||
|
||||
@@ -319,7 +319,6 @@
|
||||
| `selector` | String | `round_robin` | Datanode selector type.<br/>- `round_robin` (default value)<br/>- `lease_based`<br/>- `load_based`<br/>For details, please see "https://docs.greptime.com/developer-guide/metasrv/selector". |
|
||||
| `use_memory_store` | Bool | `false` | Store data in memory. |
|
||||
| `enable_region_failover` | Bool | `false` | Whether to enable region failover.<br/>This feature is only available on GreptimeDB running on cluster mode and<br/>- Using Remote WAL<br/>- Using shared storage (e.g., s3). |
|
||||
| `allow_region_failover_on_local_wal` | Bool | `false` | Whether to allow region failover on local WAL.<br/>**This option is not recommended to be set to true, because it may lead to data loss during failover.** |
|
||||
| `node_max_idle_time` | String | `24hours` | Max allowed idle time before removing node info from metasrv memory. |
|
||||
| `enable_telemetry` | Bool | `true` | Whether to enable greptimedb telemetry. Enabled by default. |
|
||||
| `runtime` | -- | -- | The runtime options. |
|
||||
|
||||
@@ -50,10 +50,6 @@ use_memory_store = false
|
||||
## - Using shared storage (e.g., s3).
|
||||
enable_region_failover = false
|
||||
|
||||
## Whether to allow region failover on local WAL.
|
||||
## **This option is not recommended to be set to true, because it may lead to data loss during failover.**
|
||||
allow_region_failover_on_local_wal = false
|
||||
|
||||
## Max allowed idle time before removing node info from metasrv memory.
|
||||
node_max_idle_time = "24hours"
|
||||
|
||||
|
||||
18
flake.lock
generated
18
flake.lock
generated
@@ -8,11 +8,11 @@
|
||||
"rust-analyzer-src": "rust-analyzer-src"
|
||||
},
|
||||
"locked": {
|
||||
"lastModified": 1742452566,
|
||||
"narHash": "sha256-sVuLDQ2UIWfXUBbctzrZrXM2X05YjX08K7XHMztt36E=",
|
||||
"lastModified": 1737613896,
|
||||
"narHash": "sha256-ldqXIglq74C7yKMFUzrS9xMT/EVs26vZpOD68Sh7OcU=",
|
||||
"owner": "nix-community",
|
||||
"repo": "fenix",
|
||||
"rev": "7d9ba794daf5e8cc7ee728859bc688d8e26d5f06",
|
||||
"rev": "303a062fdd8e89f233db05868468975d17855d80",
|
||||
"type": "github"
|
||||
},
|
||||
"original": {
|
||||
@@ -41,11 +41,11 @@
|
||||
},
|
||||
"nixpkgs": {
|
||||
"locked": {
|
||||
"lastModified": 1743576891,
|
||||
"narHash": "sha256-vXiKURtntURybE6FMNFAVpRPr8+e8KoLPrYs9TGuAKc=",
|
||||
"lastModified": 1737569578,
|
||||
"narHash": "sha256-6qY0pk2QmUtBT9Mywdvif0i/CLVgpCjMUn6g9vB+f3M=",
|
||||
"owner": "NixOS",
|
||||
"repo": "nixpkgs",
|
||||
"rev": "44a69ed688786e98a101f02b712c313f1ade37ab",
|
||||
"rev": "47addd76727f42d351590c905d9d1905ca895b82",
|
||||
"type": "github"
|
||||
},
|
||||
"original": {
|
||||
@@ -65,11 +65,11 @@
|
||||
"rust-analyzer-src": {
|
||||
"flake": false,
|
||||
"locked": {
|
||||
"lastModified": 1742296961,
|
||||
"narHash": "sha256-gCpvEQOrugHWLimD1wTFOJHagnSEP6VYBDspq96Idu0=",
|
||||
"lastModified": 1737581772,
|
||||
"narHash": "sha256-t1P2Pe3FAX9TlJsCZbmJ3wn+C4qr6aSMypAOu8WNsN0=",
|
||||
"owner": "rust-lang",
|
||||
"repo": "rust-analyzer",
|
||||
"rev": "15d87419f1a123d8f888d608129c3ce3ff8f13d4",
|
||||
"rev": "582af7ee9c8d84f5d534272fc7de9f292bd849be",
|
||||
"type": "github"
|
||||
},
|
||||
"original": {
|
||||
|
||||
@@ -21,7 +21,7 @@
|
||||
lib = nixpkgs.lib;
|
||||
rustToolchain = fenix.packages.${system}.fromToolchainName {
|
||||
name = (lib.importTOML ./rust-toolchain.toml).toolchain.channel;
|
||||
sha256 = "sha256-i0Sh/ZFFsHlZ3oFZFc24qdk6Cd8Do8OPU4HJQsrKOeM=";
|
||||
sha256 = "sha256-f/CVA1EC61EWbh0SjaRNhLL0Ypx2ObupbzigZp8NmL4=";
|
||||
};
|
||||
in
|
||||
{
|
||||
|
||||
@@ -4,21 +4,15 @@
|
||||
|
||||
This repository maintains the Grafana dashboards for GreptimeDB. It has two types of dashboards:
|
||||
|
||||
- `cluster/dashboard.json`: The Grafana dashboard for the GreptimeDB cluster. Read the [dashboard.md](./dashboards/cluster/dashboard.md) for more details.
|
||||
- `standalone/dashboard.json`: The Grafana dashboard for the standalone GreptimeDB instance. **It's generated from the `cluster/dashboard.json` by removing the instance filter through the `make dashboards` command**. Read the [dashboard.md](./dashboards/standalone/dashboard.md) for more details.
|
||||
- `cluster/`: The dashboard for the GreptimeDB cluster. Read the [dashboard.md](./dashboards/cluster/dashboard.md) for more details.
|
||||
- `standalone/`: The dashboard for the standalone GreptimeDB instance. Read the [dashboard.md](./dashboards/standalone/dashboard.md) for more details.
|
||||
|
||||
As the rapid development of GreptimeDB, the metrics may be changed, and please feel free to submit your feedback and/or contribution to this dashboard 🤗
|
||||
|
||||
**NOTE**:
|
||||
|
||||
- The Grafana version should be greater than 9.0.
|
||||
|
||||
- If you want to modify the dashboards, you only need to modify the `cluster/dashboard.json` and run the `make dashboards` command to generate the `standalone/dashboard.json` and other related files.
|
||||
|
||||
To maintain the dashboards easily, we use the [`dac`](https://github.com/zyy17/dac) tool to generate the intermediate dashboards and markdown documents:
|
||||
To maintain the dashboards, we use the [`dac`](https://github.com/zyy17/dac) tool to generate the intermediate dashboards and markdown documents:
|
||||
|
||||
- `cluster/dashboard.yaml`: The intermediate dashboard for the GreptimeDB cluster.
|
||||
- `standalone/dashboard.yaml`: The intermediate dashboard for the standalone GreptimeDB instance.
|
||||
- `standalone/dashboard.yaml`: The intermediatedashboard for the standalone GreptimeDB instance.
|
||||
|
||||
## Data Sources
|
||||
|
||||
|
||||
File diff suppressed because it is too large
Load Diff
@@ -1,97 +1,96 @@
|
||||
# Overview
|
||||
| Title | Query | Type | Description | Datasource | Unit | Legend Format |
|
||||
| --- | --- | --- | --- | --- | --- | --- |
|
||||
| Uptime | `time() - process_start_time_seconds` | `stat` | The start time of GreptimeDB. | `prometheus` | `s` | `__auto` |
|
||||
| Version | `SELECT pkg_version FROM information_schema.build_info` | `stat` | GreptimeDB version. | `mysql` | -- | -- |
|
||||
| Total Ingestion Rate | `sum(rate(greptime_table_operator_ingest_rows[$__rate_interval]))` | `stat` | Total ingestion rate. | `prometheus` | `rowsps` | `__auto` |
|
||||
| Total Storage Size | `select SUM(disk_size) from information_schema.region_statistics;` | `stat` | Total number of data file size. | `mysql` | `decbytes` | -- |
|
||||
| Total Rows | `select SUM(region_rows) from information_schema.region_statistics;` | `stat` | Total number of data rows in the cluster. Calculated by sum of rows from each region. | `mysql` | `sishort` | -- |
|
||||
| Deployment | `SELECT count(*) as datanode FROM information_schema.cluster_info WHERE peer_type = 'DATANODE';`<br/>`SELECT count(*) as frontend FROM information_schema.cluster_info WHERE peer_type = 'FRONTEND';`<br/>`SELECT count(*) as metasrv FROM information_schema.cluster_info WHERE peer_type = 'METASRV';`<br/>`SELECT count(*) as flownode FROM information_schema.cluster_info WHERE peer_type = 'FLOWNODE';` | `stat` | The deployment topology of GreptimeDB. | `mysql` | -- | -- |
|
||||
| Database Resources | `SELECT COUNT(*) as databases FROM information_schema.schemata WHERE schema_name NOT IN ('greptime_private', 'information_schema')`<br/>`SELECT COUNT(*) as tables FROM information_schema.tables WHERE table_schema != 'information_schema'`<br/>`SELECT COUNT(region_id) as regions FROM information_schema.region_peers`<br/>`SELECT COUNT(*) as flows FROM information_schema.flows` | `stat` | The number of the key resources in GreptimeDB. | `mysql` | -- | -- |
|
||||
| Data Size | `SELECT SUM(memtable_size) * 0.42825 as WAL FROM information_schema.region_statistics;`<br/>`SELECT SUM(index_size) as index FROM information_schema.region_statistics;`<br/>`SELECT SUM(manifest_size) as manifest FROM information_schema.region_statistics;` | `stat` | The data size of wal/index/manifest in the GreptimeDB. | `mysql` | `decbytes` | -- |
|
||||
| Uptime | `time() - process_start_time_seconds` | `stat` | The start time of GreptimeDB. | `s` | `prometheus` | `__auto` |
|
||||
| Version | `SELECT pkg_version FROM information_schema.build_info` | `stat` | GreptimeDB version. | -- | `mysql` | -- |
|
||||
| Total Ingestion Rate | `sum(rate(greptime_table_operator_ingest_rows[$__rate_interval]))` | `stat` | Total ingestion rate. | `rowsps` | `prometheus` | `__auto` |
|
||||
| Total Storage Size | `select SUM(disk_size) from information_schema.region_statistics;` | `stat` | Total number of data file size. | `decbytes` | `mysql` | -- |
|
||||
| Total Rows | `select SUM(region_rows) from information_schema.region_statistics;` | `stat` | Total number of data rows in the cluster. Calculated by sum of rows from each region. | `sishort` | `mysql` | -- |
|
||||
| Deployment | `SELECT count(*) as datanode FROM information_schema.cluster_info WHERE peer_type = 'DATANODE';`<br/>`SELECT count(*) as frontend FROM information_schema.cluster_info WHERE peer_type = 'FRONTEND';`<br/>`SELECT count(*) as metasrv FROM information_schema.cluster_info WHERE peer_type = 'METASRV';`<br/>`SELECT count(*) as flownode FROM information_schema.cluster_info WHERE peer_type = 'FLOWNODE';` | `stat` | The deployment topology of GreptimeDB. | -- | `mysql` | -- |
|
||||
| Database Resources | `SELECT COUNT(*) as databases FROM information_schema.schemata WHERE schema_name NOT IN ('greptime_private', 'information_schema')`<br/>`SELECT COUNT(*) as tables FROM information_schema.tables WHERE table_schema != 'information_schema'`<br/>`SELECT COUNT(region_id) as regions FROM information_schema.region_peers`<br/>`SELECT COUNT(*) as flows FROM information_schema.flows` | `stat` | The number of the key resources in GreptimeDB. | -- | `mysql` | -- |
|
||||
| Data Size | `SELECT SUM(memtable_size) * 0.42825 as WAL FROM information_schema.region_statistics;`<br/>`SELECT SUM(index_size) as index FROM information_schema.region_statistics;`<br/>`SELECT SUM(manifest_size) as manifest FROM information_schema.region_statistics;` | `stat` | The data size of wal/index/manifest in the GreptimeDB. | `decbytes` | `mysql` | -- |
|
||||
# Ingestion
|
||||
| Title | Query | Type | Description | Datasource | Unit | Legend Format |
|
||||
| --- | --- | --- | --- | --- | --- | --- |
|
||||
| Total Ingestion Rate | `sum(rate(greptime_table_operator_ingest_rows{instance=~"$frontend"}[$__rate_interval]))` | `timeseries` | Total ingestion rate.<br/><br/>Here we listed 3 primary protocols:<br/><br/>- Prometheus remote write<br/>- Greptime's gRPC API (when using our ingest SDK)<br/>- Log ingestion http API<br/> | `prometheus` | `rowsps` | `ingestion` |
|
||||
| Ingestion Rate by Type | `sum(rate(greptime_servers_http_logs_ingestion_counter[$__rate_interval]))`<br/>`sum(rate(greptime_servers_prometheus_remote_write_samples[$__rate_interval]))` | `timeseries` | Total ingestion rate.<br/><br/>Here we listed 3 primary protocols:<br/><br/>- Prometheus remote write<br/>- Greptime's gRPC API (when using our ingest SDK)<br/>- Log ingestion http API<br/> | `prometheus` | `rowsps` | `http-logs` |
|
||||
| Total Ingestion Rate | `sum(rate(greptime_table_operator_ingest_rows{instance=~"$frontend"}[$__rate_interval]))` | `timeseries` | Total ingestion rate.<br/><br/>Here we listed 3 primary protocols:<br/><br/>- Prometheus remote write<br/>- Greptime's gRPC API (when using our ingest SDK)<br/>- Log ingestion http API<br/> | `rowsps` | `prometheus` | `ingestion` |
|
||||
| Ingestion Rate by Type | `sum(rate(greptime_servers_http_logs_ingestion_counter[$__rate_interval]))`<br/>`sum(rate(greptime_servers_prometheus_remote_write_samples[$__rate_interval]))` | `timeseries` | Total ingestion rate.<br/><br/>Here we listed 3 primary protocols:<br/><br/>- Prometheus remote write<br/>- Greptime's gRPC API (when using our ingest SDK)<br/>- Log ingestion http API<br/> | `rowsps` | `prometheus` | `http-logs` |
|
||||
# Queries
|
||||
| Title | Query | Type | Description | Datasource | Unit | Legend Format |
|
||||
| --- | --- | --- | --- | --- | --- | --- |
|
||||
| Total Query Rate | `sum (rate(greptime_servers_mysql_query_elapsed_count{instance=~"$frontend"}[$__rate_interval]))`<br/>`sum (rate(greptime_servers_postgres_query_elapsed_count{instance=~"$frontend"}[$__rate_interval]))`<br/>`sum (rate(greptime_servers_http_promql_elapsed_counte{instance=~"$frontend"}[$__rate_interval]))` | `timeseries` | Total rate of query API calls by protocol. This metric is collected from frontends.<br/><br/>Here we listed 3 main protocols:<br/>- MySQL<br/>- Postgres<br/>- Prometheus API<br/><br/>Note that there are some other minor query APIs like /sql are not included | `prometheus` | `reqps` | `mysql` |
|
||||
| Total Query Rate | `sum (rate(greptime_servers_mysql_query_elapsed_count{instance=~"$frontend"}[$__rate_interval]))`<br/>`sum (rate(greptime_servers_postgres_query_elapsed_count{instance=~"$frontend"}[$__rate_interval]))`<br/>`sum (rate(greptime_servers_http_promql_elapsed_counte{instance=~"$frontend"}[$__rate_interval]))` | `timeseries` | Total rate of query API calls by protocol. This metric is collected from frontends.<br/><br/>Here we listed 3 main protocols:<br/>- MySQL<br/>- Postgres<br/>- Prometheus API<br/><br/>Note that there are some other minor query APIs like /sql are not included | `reqps` | `prometheus` | `mysql` |
|
||||
# Resources
|
||||
| Title | Query | Type | Description | Datasource | Unit | Legend Format |
|
||||
| --- | --- | --- | --- | --- | --- | --- |
|
||||
| Datanode Memory per Instance | `sum(process_resident_memory_bytes{instance=~"$datanode"}) by (instance, pod)` | `timeseries` | Current memory usage by instance | `prometheus` | `decbytes` | `[{{instance}}]-[{{ pod }}]` |
|
||||
| Datanode CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{instance=~"$datanode"}[$__rate_interval]) * 1000) by (instance, pod)` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]` |
|
||||
| Frontend Memory per Instance | `sum(process_resident_memory_bytes{instance=~"$frontend"}) by (instance, pod)` | `timeseries` | Current memory usage by instance | `prometheus` | `decbytes` | `[{{ instance }}]-[{{ pod }}]` |
|
||||
| Frontend CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{instance=~"$frontend"}[$__rate_interval]) * 1000) by (instance, pod)` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]-cpu` |
|
||||
| Metasrv Memory per Instance | `sum(process_resident_memory_bytes{instance=~"$metasrv"}) by (instance, pod)` | `timeseries` | Current memory usage by instance | `prometheus` | `decbytes` | `[{{ instance }}]-[{{ pod }}]-resident` |
|
||||
| Metasrv CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{instance=~"$metasrv"}[$__rate_interval]) * 1000) by (instance, pod)` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]` |
|
||||
| Flownode Memory per Instance | `sum(process_resident_memory_bytes{instance=~"$flownode"}) by (instance, pod)` | `timeseries` | Current memory usage by instance | `prometheus` | `decbytes` | `[{{ instance }}]-[{{ pod }}]` |
|
||||
| Flownode CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{instance=~"$flownode"}[$__rate_interval]) * 1000) by (instance, pod)` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]` |
|
||||
| Datanode Memory per Instance | `sum(process_resident_memory_bytes{instance=~"$datanode"}) by (instance, pod)` | `timeseries` | Current memory usage by instance | `decbytes` | `prometheus` | `[{{instance}}]-[{{ pod }}]` |
|
||||
| Datanode CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{instance=~"$datanode"}[$__rate_interval]) * 1000) by (instance, pod)` | `timeseries` | Current cpu usage by instance | `none` | `prometheus` | `[{{ instance }}]-[{{ pod }}]` |
|
||||
| Frontend Memory per Instance | `sum(process_resident_memory_bytes{instance=~"$frontend"}) by (instance, pod)` | `timeseries` | Current memory usage by instance | `decbytes` | `prometheus` | `[{{ instance }}]-[{{ pod }}]` |
|
||||
| Frontend CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{instance=~"$frontend"}[$__rate_interval]) * 1000) by (instance, pod)` | `timeseries` | Current cpu usage by instance | `none` | `prometheus` | `[{{ instance }}]-[{{ pod }}]-cpu` |
|
||||
| Metasrv Memory per Instance | `sum(process_resident_memory_bytes{instance=~"$metasrv"}) by (instance, pod)` | `timeseries` | Current memory usage by instance | `decbytes` | `prometheus` | `[{{ instance }}]-[{{ pod }}]-resident` |
|
||||
| Metasrv CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{instance=~"$metasrv"}[$__rate_interval]) * 1000) by (instance, pod)` | `timeseries` | Current cpu usage by instance | `none` | `prometheus` | `[{{ instance }}]-[{{ pod }}]` |
|
||||
| Flownode Memory per Instance | `sum(process_resident_memory_bytes{instance=~"$flownode"}) by (instance, pod)` | `timeseries` | Current memory usage by instance | `decbytes` | `prometheus` | `[{{ instance }}]-[{{ pod }}]` |
|
||||
| Flownode CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{instance=~"$flownode"}[$__rate_interval]) * 1000) by (instance, pod)` | `timeseries` | Current cpu usage by instance | `none` | `prometheus` | `[{{ instance }}]-[{{ pod }}]` |
|
||||
# Frontend Requests
|
||||
| Title | Query | Type | Description | Datasource | Unit | Legend Format |
|
||||
| --- | --- | --- | --- | --- | --- | --- |
|
||||
| HTTP QPS per Instance | `sum by(instance, pod, path, method, code) (rate(greptime_servers_http_requests_elapsed_count{instance=~"$frontend",path!~"/health\|/metrics"}[$__rate_interval]))` | `timeseries` | HTTP QPS per Instance. | `prometheus` | `reqps` | `[{{instance}}]-[{{pod}}]-[{{path}}]-[{{method}}]-[{{code}}]` |
|
||||
| HTTP P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, path, method, code) (rate(greptime_servers_http_requests_elapsed_bucket{instance=~"$frontend",path!~"/health\|/metrics"}[$__rate_interval])))` | `timeseries` | HTTP P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{path}}]-[{{method}}]-[{{code}}]-p99` |
|
||||
| gRPC QPS per Instance | `sum by(instance, pod, path, code) (rate(greptime_servers_grpc_requests_elapsed_count{instance=~"$frontend"}[$__rate_interval]))` | `timeseries` | gRPC QPS per Instance. | `prometheus` | `reqps` | `[{{instance}}]-[{{pod}}]-[{{path}}]-[{{code}}]` |
|
||||
| gRPC P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, path, code) (rate(greptime_servers_grpc_requests_elapsed_bucket{instance=~"$frontend"}[$__rate_interval])))` | `timeseries` | gRPC P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{path}}]-[{{method}}]-[{{code}}]-p99` |
|
||||
| MySQL QPS per Instance | `sum by(pod, instance)(rate(greptime_servers_mysql_query_elapsed_count{instance=~"$frontend"}[$__rate_interval]))` | `timeseries` | MySQL QPS per Instance. | `prometheus` | `reqps` | `[{{instance}}]-[{{pod}}]` |
|
||||
| MySQL P99 per Instance | `histogram_quantile(0.99, sum by(pod, instance, le) (rate(greptime_servers_mysql_query_elapsed_bucket{instance=~"$frontend"}[$__rate_interval])))` | `timeseries` | MySQL P99 per Instance. | `prometheus` | `s` | `[{{ instance }}]-[{{ pod }}]-p99` |
|
||||
| PostgreSQL QPS per Instance | `sum by(pod, instance)(rate(greptime_servers_postgres_query_elapsed_count{instance=~"$frontend"}[$__rate_interval]))` | `timeseries` | PostgreSQL QPS per Instance. | `prometheus` | `reqps` | `[{{instance}}]-[{{pod}}]` |
|
||||
| PostgreSQL P99 per Instance | `histogram_quantile(0.99, sum by(pod,instance,le) (rate(greptime_servers_postgres_query_elapsed_bucket{instance=~"$frontend"}[$__rate_interval])))` | `timeseries` | PostgreSQL P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-p99` |
|
||||
| HTTP QPS per Instance | `sum by(instance, pod, path, method, code) (rate(greptime_servers_http_requests_elapsed_count{instance=~"$frontend",path!~"/health\|/metrics"}[$__rate_interval]))` | `timeseries` | HTTP QPS per Instance. | `reqps` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{path}}]-[{{method}}]-[{{code}}]` |
|
||||
| HTTP P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, path, method, code) (rate(greptime_servers_http_requests_elapsed_bucket{instance=~"$frontend",path!~"/health\|/metrics"}[$__rate_interval])))` | `timeseries` | HTTP P99 per Instance. | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{path}}]-[{{method}}]-[{{code}}]-p99` |
|
||||
| gRPC QPS per Instance | `sum by(instance, pod, path, code) (rate(greptime_servers_grpc_requests_elapsed_count{instance=~"$frontend"}[$__rate_interval]))` | `timeseries` | gRPC QPS per Instance. | `reqps` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{path}}]-[{{code}}]` |
|
||||
| gRPC P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, path, code) (rate(greptime_servers_grpc_requests_elapsed_bucket{instance=~"$frontend"}[$__rate_interval])))` | `timeseries` | gRPC P99 per Instance. | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{path}}]-[{{method}}]-[{{code}}]-p99` |
|
||||
| MySQL QPS per Instance | `sum by(pod, instance)(rate(greptime_servers_mysql_query_elapsed_count{instance=~"$frontend"}[$__rate_interval]))` | `timeseries` | MySQL QPS per Instance. | `reqps` | `prometheus` | `[{{instance}}]-[{{pod}}]` |
|
||||
| MySQL P99 per Instance | `histogram_quantile(0.99, sum by(pod, instance, le) (rate(greptime_servers_mysql_query_elapsed_bucket{instance=~"$frontend"}[$__rate_interval])))` | `timeseries` | MySQL P99 per Instance. | `s` | `prometheus` | `[{{ instance }}]-[{{ pod }}]-p99` |
|
||||
| PostgreSQL QPS per Instance | `sum by(pod, instance)(rate(greptime_servers_postgres_query_elapsed_count{instance=~"$frontend"}[$__rate_interval]))` | `timeseries` | PostgreSQL QPS per Instance. | `reqps` | `prometheus` | `[{{instance}}]-[{{pod}}]` |
|
||||
| PostgreSQL P99 per Instance | `histogram_quantile(0.99, sum by(pod,instance,le) (rate(greptime_servers_postgres_query_elapsed_bucket{instance=~"$frontend"}[$__rate_interval])))` | `timeseries` | PostgreSQL P99 per Instance. | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-p99` |
|
||||
# Frontend to Datanode
|
||||
| Title | Query | Type | Description | Datasource | Unit | Legend Format |
|
||||
| --- | --- | --- | --- | --- | --- | --- |
|
||||
| Ingest Rows per Instance | `sum by(instance, pod)(rate(greptime_table_operator_ingest_rows{instance=~"$frontend"}[$__rate_interval]))` | `timeseries` | Ingestion rate by row as in each frontend | `prometheus` | `rowsps` | `[{{instance}}]-[{{pod}}]` |
|
||||
| Region Call QPS per Instance | `sum by(instance, pod, request_type) (rate(greptime_grpc_region_request_count{instance=~"$frontend"}[$__rate_interval]))` | `timeseries` | Region Call QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{request_type}}]` |
|
||||
| Region Call P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, request_type) (rate(greptime_grpc_region_request_bucket{instance=~"$frontend"}[$__rate_interval])))` | `timeseries` | Region Call P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{request_type}}]` |
|
||||
| Ingest Rows per Instance | `sum by(instance, pod)(rate(greptime_table_operator_ingest_rows{instance=~"$frontend"}[$__rate_interval]))` | `timeseries` | Ingestion rate by row as in each frontend | `rowsps` | `prometheus` | `[{{instance}}]-[{{pod}}]` |
|
||||
| Region Call QPS per Instance | `sum by(instance, pod, request_type) (rate(greptime_grpc_region_request_count{instance=~"$frontend"}[$__rate_interval]))` | `timeseries` | Region Call QPS per Instance. | `ops` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{request_type}}]` |
|
||||
| Region Call P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, request_type) (rate(greptime_grpc_region_request_bucket{instance=~"$frontend"}[$__rate_interval])))` | `timeseries` | Region Call P99 per Instance. | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{request_type}}]` |
|
||||
# Mito Engine
|
||||
| Title | Query | Type | Description | Datasource | Unit | Legend Format |
|
||||
| --- | --- | --- | --- | --- | --- | --- |
|
||||
| Request OPS per Instance | `sum by(instance, pod, type) (rate(greptime_mito_handle_request_elapsed_count{instance=~"$datanode"}[$__rate_interval]))` | `timeseries` | Request QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{type}}]` |
|
||||
| Request P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, type) (rate(greptime_mito_handle_request_elapsed_bucket{instance=~"$datanode"}[$__rate_interval])))` | `timeseries` | Request P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{type}}]` |
|
||||
| Write Buffer per Instance | `greptime_mito_write_buffer_bytes{instance=~"$datanode"}` | `timeseries` | Write Buffer per Instance. | `prometheus` | `decbytes` | `[{{instance}}]-[{{pod}}]` |
|
||||
| Write Rows per Instance | `sum by (instance, pod) (rate(greptime_mito_write_rows_total{instance=~"$datanode"}[$__rate_interval]))` | `timeseries` | Ingestion size by row counts. | `prometheus` | `rowsps` | `[{{instance}}]-[{{pod}}]` |
|
||||
| Flush OPS per Instance | `sum by(instance, pod, reason) (rate(greptime_mito_flush_requests_total{instance=~"$datanode"}[$__rate_interval]))` | `timeseries` | Flush QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{reason}}]` |
|
||||
| Write Stall per Instance | `sum by(instance, pod) (greptime_mito_write_stall_total{instance=~"$datanode"})` | `timeseries` | Write Stall per Instance. | `prometheus` | -- | `[{{instance}}]-[{{pod}}]` |
|
||||
| Read Stage OPS per Instance | `sum by(instance, pod) (rate(greptime_mito_read_stage_elapsed_count{instance=~"$datanode", stage="total"}[$__rate_interval]))` | `timeseries` | Read Stage OPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]` |
|
||||
| Read Stage P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, stage) (rate(greptime_mito_read_stage_elapsed_bucket{instance=~"$datanode"}[$__rate_interval])))` | `timeseries` | Read Stage P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{stage}}]` |
|
||||
| Write Stage P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, stage) (rate(greptime_mito_write_stage_elapsed_bucket{instance=~"$datanode"}[$__rate_interval])))` | `timeseries` | Write Stage P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{stage}}]` |
|
||||
| Compaction OPS per Instance | `sum by(instance, pod) (rate(greptime_mito_compaction_total_elapsed_count{instance=~"$datanode"}[$__rate_interval]))` | `timeseries` | Compaction OPS per Instance. | `prometheus` | `ops` | `[{{ instance }}]-[{{pod}}]` |
|
||||
| Compaction P99 per Instance by Stage | `histogram_quantile(0.99, sum by(instance, pod, le, stage) (rate(greptime_mito_compaction_stage_elapsed_bucket{instance=~"$datanode"}[$__rate_interval])))` | `timeseries` | Compaction latency by stage | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{stage}}]-p99` |
|
||||
| Compaction P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le,stage) (rate(greptime_mito_compaction_total_elapsed_bucket{instance=~"$datanode"}[$__rate_interval])))` | `timeseries` | Compaction P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{stage}}]-compaction` |
|
||||
| WAL write size | `histogram_quantile(0.95, sum by(le,instance, pod) (rate(raft_engine_write_size_bucket[$__rate_interval])))`<br/>`histogram_quantile(0.99, sum by(le,instance,pod) (rate(raft_engine_write_size_bucket[$__rate_interval])))`<br/>`sum by (instance, pod)(rate(raft_engine_write_size_sum[$__rate_interval]))` | `timeseries` | Write-ahead logs write size as bytes. This chart includes stats of p95 and p99 size by instance, total WAL write rate. | `prometheus` | `bytes` | `[{{instance}}]-[{{pod}}]-req-size-p95` |
|
||||
| Cached Bytes per Instance | `greptime_mito_cache_bytes{instance=~"$datanode"}` | `timeseries` | Cached Bytes per Instance. | `prometheus` | `decbytes` | `[{{instance}}]-[{{pod}}]-[{{type}}]` |
|
||||
| Inflight Compaction | `greptime_mito_inflight_compaction_count` | `timeseries` | Ongoing compaction task count | `prometheus` | `none` | `[{{instance}}]-[{{pod}}]` |
|
||||
| WAL sync duration seconds | `histogram_quantile(0.99, sum by(le, type, node, instance, pod) (rate(raft_engine_sync_log_duration_seconds_bucket[$__rate_interval])))` | `timeseries` | Raft engine (local disk) log store sync latency, p99 | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-p99` |
|
||||
| Log Store op duration seconds | `histogram_quantile(0.99, sum by(le,logstore,optype,instance, pod) (rate(greptime_logstore_op_elapsed_bucket[$__rate_interval])))` | `timeseries` | Write-ahead log operations latency at p99 | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{logstore}}]-[{{optype}}]-p99` |
|
||||
| Inflight Flush | `greptime_mito_inflight_flush_count` | `timeseries` | Ongoing flush task count | `prometheus` | `none` | `[{{instance}}]-[{{pod}}]` |
|
||||
| Request OPS per Instance | `sum by(instance, pod, type) (rate(greptime_mito_handle_request_elapsed_count{instance=~"$datanode"}[$__rate_interval]))` | `timeseries` | Request QPS per Instance. | `ops` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{type}}]` |
|
||||
| Request P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, type) (rate(greptime_mito_handle_request_elapsed_bucket{instance=~"$datanode"}[$__rate_interval])))` | `timeseries` | Request P99 per Instance. | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{type}}]` |
|
||||
| Write Buffer per Instance | `greptime_mito_write_buffer_bytes{instance=~"$datanode"}` | `timeseries` | Write Buffer per Instance. | `decbytes` | `prometheus` | `[{{instance}}]-[{{pod}}]` |
|
||||
| Write Rows per Instance | `sum by (instance, pod) (rate(greptime_mito_write_rows_total{instance=~"$datanode"}[$__rate_interval]))` | `timeseries` | Ingestion size by row counts. | `rowsps` | `prometheus` | `[{{instance}}]-[{{pod}}]` |
|
||||
| Flush OPS per Instance | `sum by(instance, pod, reason) (rate(greptime_mito_flush_requests_total{instance=~"$datanode"}[$__rate_interval]))` | `timeseries` | Flush QPS per Instance. | `ops` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{reason}}]` |
|
||||
| Write Stall per Instance | `sum by(instance, pod) (greptime_mito_write_stall_total{instance=~"$datanode"})` | `timeseries` | Write Stall per Instance. | `decbytes` | `prometheus` | `[{{instance}}]-[{{pod}}]` |
|
||||
| Read Stage OPS per Instance | `sum by(instance, pod) (rate(greptime_mito_read_stage_elapsed_count{instance=~"$datanode", stage="total"}[$__rate_interval]))` | `timeseries` | Read Stage OPS per Instance. | `ops` | `prometheus` | `[{{instance}}]-[{{pod}}]` |
|
||||
| Read Stage P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, stage) (rate(greptime_mito_read_stage_elapsed_bucket{instance=~"$datanode"}[$__rate_interval])))` | `timeseries` | Read Stage P99 per Instance. | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{stage}}]` |
|
||||
| Write Stage P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, stage) (rate(greptime_mito_write_stage_elapsed_bucket{instance=~"$datanode"}[$__rate_interval])))` | `timeseries` | Write Stage P99 per Instance. | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{stage}}]` |
|
||||
| Compaction OPS per Instance | `sum by(instance, pod) (rate(greptime_mito_compaction_total_elapsed_count{instance=~"$datanode"}[$__rate_interval]))` | `timeseries` | Compaction OPS per Instance. | `ops` | `prometheus` | `[{{ instance }}]-[{{pod}}]` |
|
||||
| Compaction P99 per Instance by Stage | `histogram_quantile(0.99, sum by(instance, pod, le, stage) (rate(greptime_mito_compaction_stage_elapsed_bucket{instance=~"$datanode"}[$__rate_interval])))` | `timeseries` | Compaction latency by stage | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{stage}}]-p99` |
|
||||
| Compaction P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le,stage) (rate(greptime_mito_compaction_total_elapsed_bucket{instance=~"$datanode"}[$__rate_interval])))` | `timeseries` | Compaction P99 per Instance. | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{stage}}]-compaction` |
|
||||
| WAL write size | `histogram_quantile(0.95, sum by(le,instance, pod) (rate(raft_engine_write_size_bucket[$__rate_interval])))`<br/>`histogram_quantile(0.99, sum by(le,instance,pod) (rate(raft_engine_write_size_bucket[$__rate_interval])))`<br/>`sum by (instance, pod)(rate(raft_engine_write_size_sum[$__rate_interval]))` | `timeseries` | Write-ahead logs write size as bytes. This chart includes stats of p95 and p99 size by instance, total WAL write rate. | `bytes` | `prometheus` | `[{{instance}}]-[{{pod}}]-req-size-p95` |
|
||||
| Cached Bytes per Instance | `greptime_mito_cache_bytes{instance=~"$datanode"}` | `timeseries` | Cached Bytes per Instance. | `decbytes` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{type}}]` |
|
||||
| Inflight Compaction | `greptime_mito_inflight_compaction_count` | `timeseries` | Ongoing compaction task count | `none` | `prometheus` | `[{{instance}}]-[{{pod}}]` |
|
||||
| WAL sync duration seconds | `histogram_quantile(0.99, sum by(le, type, node, instance, pod) (rate(raft_engine_sync_log_duration_seconds_bucket[$__rate_interval])))` | `timeseries` | Raft engine (local disk) log store sync latency, p99 | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-p99` |
|
||||
| Log Store op duration seconds | `histogram_quantile(0.99, sum by(le,logstore,optype,instance, pod) (rate(greptime_logstore_op_elapsed_bucket[$__rate_interval])))` | `timeseries` | Write-ahead log operations latency at p99 | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{logstore}}]-[{{optype}}]-p99` |
|
||||
| Inflight Flush | `greptime_mito_inflight_flush_count` | `timeseries` | Ongoing flush task count | `none` | `prometheus` | `[{{instance}}]-[{{pod}}]` |
|
||||
# OpenDAL
|
||||
| Title | Query | Type | Description | Datasource | Unit | Legend Format |
|
||||
| --- | --- | --- | --- | --- | --- | --- |
|
||||
| QPS per Instance | `sum by(instance, pod, scheme, operation) (rate(opendal_operation_duration_seconds_count{instance=~"$datanode"}[$__rate_interval]))` | `timeseries` | QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]` |
|
||||
| Read QPS per Instance | `sum by(instance, pod, scheme) (rate(opendal_operation_duration_seconds_count{instance=~"$datanode", operation="read"}[$__rate_interval]))` | `timeseries` | Read QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]` |
|
||||
| Read P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, scheme) (rate(opendal_operation_duration_seconds_bucket{instance=~"$datanode",operation="read"}[$__rate_interval])))` | `timeseries` | Read P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-{{scheme}}` |
|
||||
| Write QPS per Instance | `sum by(instance, pod, scheme) (rate(opendal_operation_duration_seconds_count{instance=~"$datanode", operation="write"}[$__rate_interval]))` | `timeseries` | Write QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-{{scheme}}` |
|
||||
| Write P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, scheme) (rate(opendal_operation_duration_seconds_bucket{instance=~"$datanode", operation="write"}[$__rate_interval])))` | `timeseries` | Write P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]` |
|
||||
| List QPS per Instance | `sum by(instance, pod, scheme) (rate(opendal_operation_duration_seconds_count{instance=~"$datanode", operation="list"}[$__rate_interval]))` | `timeseries` | List QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]` |
|
||||
| List P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, scheme) (rate(opendal_operation_duration_seconds_bucket{instance=~"$datanode", operation="list"}[$__rate_interval])))` | `timeseries` | List P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]` |
|
||||
| Other Requests per Instance | `sum by(instance, pod, scheme, operation) (rate(opendal_operation_duration_seconds_count{instance=~"$datanode",operation!~"read\|write\|list\|stat"}[$__rate_interval]))` | `timeseries` | Other Requests per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]` |
|
||||
| Other Request P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, scheme, operation) (rate(opendal_operation_duration_seconds_bucket{instance=~"$datanode", operation!~"read\|write\|list"}[$__rate_interval])))` | `timeseries` | Other Request P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]` |
|
||||
| Opendal traffic | `sum by(instance, pod, scheme, operation) (rate(opendal_operation_bytes_sum{instance=~"$datanode"}[$__rate_interval]))` | `timeseries` | Total traffic as in bytes by instance and operation | `prometheus` | `decbytes` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]` |
|
||||
| OpenDAL errors per Instance | `sum by(instance, pod, scheme, operation, error) (rate(opendal_operation_errors_total{instance=~"$datanode", error!="NotFound"}[$__rate_interval]))` | `timeseries` | OpenDAL error counts per Instance. | `prometheus` | -- | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]-[{{error}}]` |
|
||||
| QPS per Instance | `sum by(instance, pod, scheme, operation) (rate(opendal_operation_duration_seconds_count{instance=~"$datanode"}[$__rate_interval]))` | `timeseries` | QPS per Instance. | `ops` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]` |
|
||||
| Read QPS per Instance | `sum by(instance, pod, scheme) (rate(opendal_operation_duration_seconds_count{instance=~"$datanode", operation="read"}[$__rate_interval]))` | `timeseries` | Read QPS per Instance. | `ops` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]` |
|
||||
| Read P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, scheme) (rate(opendal_operation_duration_seconds_bucket{instance=~"$datanode",operation="read"}[$__rate_interval])))` | `timeseries` | Read P99 per Instance. | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-{{scheme}}` |
|
||||
| Write QPS per Instance | `sum by(instance, pod, scheme) (rate(opendal_operation_duration_seconds_count{instance=~"$datanode", operation="write"}[$__rate_interval]))` | `timeseries` | Write QPS per Instance. | `ops` | `prometheus` | `[{{instance}}]-[{{pod}}]-{{scheme}}` |
|
||||
| Write P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, scheme) (rate(opendal_operation_duration_seconds_bucket{instance=~"$datanode", operation="write"}[$__rate_interval])))` | `timeseries` | Write P99 per Instance. | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]` |
|
||||
| List QPS per Instance | `sum by(instance, pod, scheme) (rate(opendal_operation_duration_seconds_count{instance=~"$datanode", operation="list"}[$__rate_interval]))` | `timeseries` | List QPS per Instance. | `ops` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]` |
|
||||
| List P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, scheme) (rate(opendal_operation_duration_seconds_bucket{instance=~"$datanode", operation="list"}[$__rate_interval])))` | `timeseries` | List P99 per Instance. | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]` |
|
||||
| Other Requests per Instance | `sum by(instance, pod, scheme, operation) (rate(opendal_operation_duration_seconds_count{instance=~"$datanode",operation!~"read\|write\|list\|stat"}[$__rate_interval]))` | `timeseries` | Other Requests per Instance. | `ops` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]` |
|
||||
| Other Request P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, scheme, operation) (rate(opendal_operation_duration_seconds_bucket{instance=~"$datanode", operation!~"read\|write\|list"}[$__rate_interval])))` | `timeseries` | Other Request P99 per Instance. | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]` |
|
||||
| Opendal traffic | `sum by(instance, pod, scheme, operation) (rate(opendal_operation_bytes_sum{instance=~"$datanode"}[$__rate_interval]))` | `timeseries` | Total traffic as in bytes by instance and operation | `ops` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]` |
|
||||
# Metasrv
|
||||
| Title | Query | Type | Description | Datasource | Unit | Legend Format |
|
||||
| --- | --- | --- | --- | --- | --- | --- |
|
||||
| Region migration datanode | `greptime_meta_region_migration_stat{datanode_type="src"}`<br/>`greptime_meta_region_migration_stat{datanode_type="desc"}` | `state-timeline` | Counter of region migration by source and destination | `prometheus` | `none` | `from-datanode-{{datanode_id}}` |
|
||||
| Region migration error | `greptime_meta_region_migration_error` | `timeseries` | Counter of region migration error | `prometheus` | `none` | `__auto` |
|
||||
| Datanode load | `greptime_datanode_load` | `timeseries` | Gauge of load information of each datanode, collected via heartbeat between datanode and metasrv. This information is for metasrv to schedule workloads. | `prometheus` | `none` | `__auto` |
|
||||
| Region migration datanode | `greptime_meta_region_migration_stat{datanode_type="src"}`<br/>`greptime_meta_region_migration_stat{datanode_type="desc"}` | `state-timeline` | Counter of region migration by source and destination | `none` | `prometheus` | `from-datanode-{{datanode_id}}` |
|
||||
| Region migration error | `greptime_meta_region_migration_error` | `timeseries` | Counter of region migration error | `none` | `prometheus` | `__auto` |
|
||||
| Datanode load | `greptime_datanode_load` | `timeseries` | Gauge of load information of each datanode, collected via heartbeat between datanode and metasrv. This information is for metasrv to schedule workloads. | `none` | `prometheus` | `__auto` |
|
||||
# Flownode
|
||||
| Title | Query | Type | Description | Datasource | Unit | Legend Format |
|
||||
| --- | --- | --- | --- | --- | --- | --- |
|
||||
| Flow Ingest / Output Rate | `sum by(instance, pod, direction) (rate(greptime_flow_processed_rows[$__rate_interval]))` | `timeseries` | Flow Ingest / Output Rate. | `prometheus` | -- | `[{{pod}}]-[{{instance}}]-[{{direction}}]` |
|
||||
| Flow Ingest Latency | `histogram_quantile(0.95, sum(rate(greptime_flow_insert_elapsed_bucket[$__rate_interval])) by (le, instance, pod))`<br/>`histogram_quantile(0.99, sum(rate(greptime_flow_insert_elapsed_bucket[$__rate_interval])) by (le, instance, pod))` | `timeseries` | Flow Ingest Latency. | `prometheus` | -- | `[{{instance}}]-[{{pod}}]-p95` |
|
||||
| Flow Operation Latency | `histogram_quantile(0.95, sum(rate(greptime_flow_processing_time_bucket[$__rate_interval])) by (le,instance,pod,type))`<br/>`histogram_quantile(0.99, sum(rate(greptime_flow_processing_time_bucket[$__rate_interval])) by (le,instance,pod,type))` | `timeseries` | Flow Operation Latency. | `prometheus` | -- | `[{{instance}}]-[{{pod}}]-[{{type}}]-p95` |
|
||||
| Flow Buffer Size per Instance | `greptime_flow_input_buf_size` | `timeseries` | Flow Buffer Size per Instance. | `prometheus` | -- | `[{{instance}}]-[{{pod}]` |
|
||||
| Flow Processing Error per Instance | `sum by(instance,pod,code) (rate(greptime_flow_errors[$__rate_interval]))` | `timeseries` | Flow Processing Error per Instance. | `prometheus` | -- | `[{{instance}}]-[{{pod}}]-[{{code}}]` |
|
||||
| Flow Ingest / Output Rate | `sum by(instance, pod, direction) (rate(greptime_flow_processed_rows[$__rate_interval]))` | `timeseries` | Flow Ingest / Output Rate. | -- | `prometheus` | `[{{pod}}]-[{{instance}}]-[{{direction}}]` |
|
||||
| Flow Ingest Latency | `histogram_quantile(0.95, sum(rate(greptime_flow_insert_elapsed_bucket[$__rate_interval])) by (le, instance, pod))`<br/>`histogram_quantile(0.99, sum(rate(greptime_flow_insert_elapsed_bucket[$__rate_interval])) by (le, instance, pod))` | `timeseries` | Flow Ingest Latency. | -- | `prometheus` | `[{{instance}}]-[{{pod}}]-p95` |
|
||||
| Flow Operation Latency | `histogram_quantile(0.95, sum(rate(greptime_flow_processing_time_bucket[$__rate_interval])) by (le,instance,pod,type))`<br/>`histogram_quantile(0.99, sum(rate(greptime_flow_processing_time_bucket[$__rate_interval])) by (le,instance,pod,type))` | `timeseries` | Flow Operation Latency. | -- | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{type}}]-p95` |
|
||||
| Flow Buffer Size per Instance | `greptime_flow_input_buf_size` | `timeseries` | Flow Buffer Size per Instance. | -- | `prometheus` | `[{{instance}}]-[{{pod}]` |
|
||||
| Flow Processing Error per Instance | `sum by(instance,pod,code) (rate(greptime_flow_errors[$__rate_interval]))` | `timeseries` | Flow Processing Error per Instance. | -- | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{code}}]` |
|
||||
|
||||
@@ -426,6 +426,7 @@ groups:
|
||||
- title: Write Stall per Instance
|
||||
type: timeseries
|
||||
description: Write Stall per Instance.
|
||||
unit: decbytes
|
||||
queries:
|
||||
- expr: sum by(instance, pod) (greptime_mito_write_stall_total{instance=~"$datanode"})
|
||||
datasource:
|
||||
@@ -657,22 +658,13 @@ groups:
|
||||
- title: Opendal traffic
|
||||
type: timeseries
|
||||
description: Total traffic as in bytes by instance and operation
|
||||
unit: decbytes
|
||||
unit: ops
|
||||
queries:
|
||||
- expr: sum by(instance, pod, scheme, operation) (rate(opendal_operation_bytes_sum{instance=~"$datanode"}[$__rate_interval]))
|
||||
datasource:
|
||||
type: prometheus
|
||||
uid: ${metrics}
|
||||
legendFormat: '[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]'
|
||||
- title: OpenDAL errors per Instance
|
||||
type: timeseries
|
||||
description: OpenDAL error counts per Instance.
|
||||
queries:
|
||||
- expr: sum by(instance, pod, scheme, operation, error) (rate(opendal_operation_errors_total{instance=~"$datanode", error!="NotFound"}[$__rate_interval]))
|
||||
datasource:
|
||||
type: prometheus
|
||||
uid: ${metrics}
|
||||
legendFormat: '[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]-[{{error}}]'
|
||||
- title: Metasrv
|
||||
panels:
|
||||
- title: Region migration datanode
|
||||
|
||||
File diff suppressed because it is too large
Load Diff
@@ -1,97 +1,96 @@
|
||||
# Overview
|
||||
| Title | Query | Type | Description | Datasource | Unit | Legend Format |
|
||||
| --- | --- | --- | --- | --- | --- | --- |
|
||||
| Uptime | `time() - process_start_time_seconds` | `stat` | The start time of GreptimeDB. | `prometheus` | `s` | `__auto` |
|
||||
| Version | `SELECT pkg_version FROM information_schema.build_info` | `stat` | GreptimeDB version. | `mysql` | -- | -- |
|
||||
| Total Ingestion Rate | `sum(rate(greptime_table_operator_ingest_rows[$__rate_interval]))` | `stat` | Total ingestion rate. | `prometheus` | `rowsps` | `__auto` |
|
||||
| Total Storage Size | `select SUM(disk_size) from information_schema.region_statistics;` | `stat` | Total number of data file size. | `mysql` | `decbytes` | -- |
|
||||
| Total Rows | `select SUM(region_rows) from information_schema.region_statistics;` | `stat` | Total number of data rows in the cluster. Calculated by sum of rows from each region. | `mysql` | `sishort` | -- |
|
||||
| Deployment | `SELECT count(*) as datanode FROM information_schema.cluster_info WHERE peer_type = 'DATANODE';`<br/>`SELECT count(*) as frontend FROM information_schema.cluster_info WHERE peer_type = 'FRONTEND';`<br/>`SELECT count(*) as metasrv FROM information_schema.cluster_info WHERE peer_type = 'METASRV';`<br/>`SELECT count(*) as flownode FROM information_schema.cluster_info WHERE peer_type = 'FLOWNODE';` | `stat` | The deployment topology of GreptimeDB. | `mysql` | -- | -- |
|
||||
| Database Resources | `SELECT COUNT(*) as databases FROM information_schema.schemata WHERE schema_name NOT IN ('greptime_private', 'information_schema')`<br/>`SELECT COUNT(*) as tables FROM information_schema.tables WHERE table_schema != 'information_schema'`<br/>`SELECT COUNT(region_id) as regions FROM information_schema.region_peers`<br/>`SELECT COUNT(*) as flows FROM information_schema.flows` | `stat` | The number of the key resources in GreptimeDB. | `mysql` | -- | -- |
|
||||
| Data Size | `SELECT SUM(memtable_size) * 0.42825 as WAL FROM information_schema.region_statistics;`<br/>`SELECT SUM(index_size) as index FROM information_schema.region_statistics;`<br/>`SELECT SUM(manifest_size) as manifest FROM information_schema.region_statistics;` | `stat` | The data size of wal/index/manifest in the GreptimeDB. | `mysql` | `decbytes` | -- |
|
||||
| Uptime | `time() - process_start_time_seconds` | `stat` | The start time of GreptimeDB. | `s` | `prometheus` | `__auto` |
|
||||
| Version | `SELECT pkg_version FROM information_schema.build_info` | `stat` | GreptimeDB version. | -- | `mysql` | -- |
|
||||
| Total Ingestion Rate | `sum(rate(greptime_table_operator_ingest_rows[$__rate_interval]))` | `stat` | Total ingestion rate. | `rowsps` | `prometheus` | `__auto` |
|
||||
| Total Storage Size | `select SUM(disk_size) from information_schema.region_statistics;` | `stat` | Total number of data file size. | `decbytes` | `mysql` | -- |
|
||||
| Total Rows | `select SUM(region_rows) from information_schema.region_statistics;` | `stat` | Total number of data rows in the cluster. Calculated by sum of rows from each region. | `sishort` | `mysql` | -- |
|
||||
| Deployment | `SELECT count(*) as datanode FROM information_schema.cluster_info WHERE peer_type = 'DATANODE';`<br/>`SELECT count(*) as frontend FROM information_schema.cluster_info WHERE peer_type = 'FRONTEND';`<br/>`SELECT count(*) as metasrv FROM information_schema.cluster_info WHERE peer_type = 'METASRV';`<br/>`SELECT count(*) as flownode FROM information_schema.cluster_info WHERE peer_type = 'FLOWNODE';` | `stat` | The deployment topology of GreptimeDB. | -- | `mysql` | -- |
|
||||
| Database Resources | `SELECT COUNT(*) as databases FROM information_schema.schemata WHERE schema_name NOT IN ('greptime_private', 'information_schema')`<br/>`SELECT COUNT(*) as tables FROM information_schema.tables WHERE table_schema != 'information_schema'`<br/>`SELECT COUNT(region_id) as regions FROM information_schema.region_peers`<br/>`SELECT COUNT(*) as flows FROM information_schema.flows` | `stat` | The number of the key resources in GreptimeDB. | -- | `mysql` | -- |
|
||||
| Data Size | `SELECT SUM(memtable_size) * 0.42825 as WAL FROM information_schema.region_statistics;`<br/>`SELECT SUM(index_size) as index FROM information_schema.region_statistics;`<br/>`SELECT SUM(manifest_size) as manifest FROM information_schema.region_statistics;` | `stat` | The data size of wal/index/manifest in the GreptimeDB. | `decbytes` | `mysql` | -- |
|
||||
# Ingestion
|
||||
| Title | Query | Type | Description | Datasource | Unit | Legend Format |
|
||||
| --- | --- | --- | --- | --- | --- | --- |
|
||||
| Total Ingestion Rate | `sum(rate(greptime_table_operator_ingest_rows{}[$__rate_interval]))` | `timeseries` | Total ingestion rate.<br/><br/>Here we listed 3 primary protocols:<br/><br/>- Prometheus remote write<br/>- Greptime's gRPC API (when using our ingest SDK)<br/>- Log ingestion http API<br/> | `prometheus` | `rowsps` | `ingestion` |
|
||||
| Ingestion Rate by Type | `sum(rate(greptime_servers_http_logs_ingestion_counter[$__rate_interval]))`<br/>`sum(rate(greptime_servers_prometheus_remote_write_samples[$__rate_interval]))` | `timeseries` | Total ingestion rate.<br/><br/>Here we listed 3 primary protocols:<br/><br/>- Prometheus remote write<br/>- Greptime's gRPC API (when using our ingest SDK)<br/>- Log ingestion http API<br/> | `prometheus` | `rowsps` | `http-logs` |
|
||||
| Total Ingestion Rate | `sum(rate(greptime_table_operator_ingest_rows{}[$__rate_interval]))` | `timeseries` | Total ingestion rate.<br/><br/>Here we listed 3 primary protocols:<br/><br/>- Prometheus remote write<br/>- Greptime's gRPC API (when using our ingest SDK)<br/>- Log ingestion http API<br/> | `rowsps` | `prometheus` | `ingestion` |
|
||||
| Ingestion Rate by Type | `sum(rate(greptime_servers_http_logs_ingestion_counter[$__rate_interval]))`<br/>`sum(rate(greptime_servers_prometheus_remote_write_samples[$__rate_interval]))` | `timeseries` | Total ingestion rate.<br/><br/>Here we listed 3 primary protocols:<br/><br/>- Prometheus remote write<br/>- Greptime's gRPC API (when using our ingest SDK)<br/>- Log ingestion http API<br/> | `rowsps` | `prometheus` | `http-logs` |
|
||||
# Queries
|
||||
| Title | Query | Type | Description | Datasource | Unit | Legend Format |
|
||||
| --- | --- | --- | --- | --- | --- | --- |
|
||||
| Total Query Rate | `sum (rate(greptime_servers_mysql_query_elapsed_count{}[$__rate_interval]))`<br/>`sum (rate(greptime_servers_postgres_query_elapsed_count{}[$__rate_interval]))`<br/>`sum (rate(greptime_servers_http_promql_elapsed_counte{}[$__rate_interval]))` | `timeseries` | Total rate of query API calls by protocol. This metric is collected from frontends.<br/><br/>Here we listed 3 main protocols:<br/>- MySQL<br/>- Postgres<br/>- Prometheus API<br/><br/>Note that there are some other minor query APIs like /sql are not included | `prometheus` | `reqps` | `mysql` |
|
||||
| Total Query Rate | `sum (rate(greptime_servers_mysql_query_elapsed_count{}[$__rate_interval]))`<br/>`sum (rate(greptime_servers_postgres_query_elapsed_count{}[$__rate_interval]))`<br/>`sum (rate(greptime_servers_http_promql_elapsed_counte{}[$__rate_interval]))` | `timeseries` | Total rate of query API calls by protocol. This metric is collected from frontends.<br/><br/>Here we listed 3 main protocols:<br/>- MySQL<br/>- Postgres<br/>- Prometheus API<br/><br/>Note that there are some other minor query APIs like /sql are not included | `reqps` | `prometheus` | `mysql` |
|
||||
# Resources
|
||||
| Title | Query | Type | Description | Datasource | Unit | Legend Format |
|
||||
| --- | --- | --- | --- | --- | --- | --- |
|
||||
| Datanode Memory per Instance | `sum(process_resident_memory_bytes{}) by (instance, pod)` | `timeseries` | Current memory usage by instance | `prometheus` | `decbytes` | `[{{instance}}]-[{{ pod }}]` |
|
||||
| Datanode CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{}[$__rate_interval]) * 1000) by (instance, pod)` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]` |
|
||||
| Frontend Memory per Instance | `sum(process_resident_memory_bytes{}) by (instance, pod)` | `timeseries` | Current memory usage by instance | `prometheus` | `decbytes` | `[{{ instance }}]-[{{ pod }}]` |
|
||||
| Frontend CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{}[$__rate_interval]) * 1000) by (instance, pod)` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]-cpu` |
|
||||
| Metasrv Memory per Instance | `sum(process_resident_memory_bytes{}) by (instance, pod)` | `timeseries` | Current memory usage by instance | `prometheus` | `decbytes` | `[{{ instance }}]-[{{ pod }}]-resident` |
|
||||
| Metasrv CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{}[$__rate_interval]) * 1000) by (instance, pod)` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]` |
|
||||
| Flownode Memory per Instance | `sum(process_resident_memory_bytes{}) by (instance, pod)` | `timeseries` | Current memory usage by instance | `prometheus` | `decbytes` | `[{{ instance }}]-[{{ pod }}]` |
|
||||
| Flownode CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{}[$__rate_interval]) * 1000) by (instance, pod)` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]` |
|
||||
| Datanode Memory per Instance | `sum(process_resident_memory_bytes{}) by (instance, pod)` | `timeseries` | Current memory usage by instance | `decbytes` | `prometheus` | `[{{instance}}]-[{{ pod }}]` |
|
||||
| Datanode CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{}[$__rate_interval]) * 1000) by (instance, pod)` | `timeseries` | Current cpu usage by instance | `none` | `prometheus` | `[{{ instance }}]-[{{ pod }}]` |
|
||||
| Frontend Memory per Instance | `sum(process_resident_memory_bytes{}) by (instance, pod)` | `timeseries` | Current memory usage by instance | `decbytes` | `prometheus` | `[{{ instance }}]-[{{ pod }}]` |
|
||||
| Frontend CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{}[$__rate_interval]) * 1000) by (instance, pod)` | `timeseries` | Current cpu usage by instance | `none` | `prometheus` | `[{{ instance }}]-[{{ pod }}]-cpu` |
|
||||
| Metasrv Memory per Instance | `sum(process_resident_memory_bytes{}) by (instance, pod)` | `timeseries` | Current memory usage by instance | `decbytes` | `prometheus` | `[{{ instance }}]-[{{ pod }}]-resident` |
|
||||
| Metasrv CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{}[$__rate_interval]) * 1000) by (instance, pod)` | `timeseries` | Current cpu usage by instance | `none` | `prometheus` | `[{{ instance }}]-[{{ pod }}]` |
|
||||
| Flownode Memory per Instance | `sum(process_resident_memory_bytes{}) by (instance, pod)` | `timeseries` | Current memory usage by instance | `decbytes` | `prometheus` | `[{{ instance }}]-[{{ pod }}]` |
|
||||
| Flownode CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{}[$__rate_interval]) * 1000) by (instance, pod)` | `timeseries` | Current cpu usage by instance | `none` | `prometheus` | `[{{ instance }}]-[{{ pod }}]` |
|
||||
# Frontend Requests
|
||||
| Title | Query | Type | Description | Datasource | Unit | Legend Format |
|
||||
| --- | --- | --- | --- | --- | --- | --- |
|
||||
| HTTP QPS per Instance | `sum by(instance, pod, path, method, code) (rate(greptime_servers_http_requests_elapsed_count{path!~"/health\|/metrics"}[$__rate_interval]))` | `timeseries` | HTTP QPS per Instance. | `prometheus` | `reqps` | `[{{instance}}]-[{{pod}}]-[{{path}}]-[{{method}}]-[{{code}}]` |
|
||||
| HTTP P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, path, method, code) (rate(greptime_servers_http_requests_elapsed_bucket{path!~"/health\|/metrics"}[$__rate_interval])))` | `timeseries` | HTTP P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{path}}]-[{{method}}]-[{{code}}]-p99` |
|
||||
| gRPC QPS per Instance | `sum by(instance, pod, path, code) (rate(greptime_servers_grpc_requests_elapsed_count{}[$__rate_interval]))` | `timeseries` | gRPC QPS per Instance. | `prometheus` | `reqps` | `[{{instance}}]-[{{pod}}]-[{{path}}]-[{{code}}]` |
|
||||
| gRPC P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, path, code) (rate(greptime_servers_grpc_requests_elapsed_bucket{}[$__rate_interval])))` | `timeseries` | gRPC P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{path}}]-[{{method}}]-[{{code}}]-p99` |
|
||||
| MySQL QPS per Instance | `sum by(pod, instance)(rate(greptime_servers_mysql_query_elapsed_count{}[$__rate_interval]))` | `timeseries` | MySQL QPS per Instance. | `prometheus` | `reqps` | `[{{instance}}]-[{{pod}}]` |
|
||||
| MySQL P99 per Instance | `histogram_quantile(0.99, sum by(pod, instance, le) (rate(greptime_servers_mysql_query_elapsed_bucket{}[$__rate_interval])))` | `timeseries` | MySQL P99 per Instance. | `prometheus` | `s` | `[{{ instance }}]-[{{ pod }}]-p99` |
|
||||
| PostgreSQL QPS per Instance | `sum by(pod, instance)(rate(greptime_servers_postgres_query_elapsed_count{}[$__rate_interval]))` | `timeseries` | PostgreSQL QPS per Instance. | `prometheus` | `reqps` | `[{{instance}}]-[{{pod}}]` |
|
||||
| PostgreSQL P99 per Instance | `histogram_quantile(0.99, sum by(pod,instance,le) (rate(greptime_servers_postgres_query_elapsed_bucket{}[$__rate_interval])))` | `timeseries` | PostgreSQL P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-p99` |
|
||||
| HTTP QPS per Instance | `sum by(instance, pod, path, method, code) (rate(greptime_servers_http_requests_elapsed_count{path!~"/health\|/metrics"}[$__rate_interval]))` | `timeseries` | HTTP QPS per Instance. | `reqps` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{path}}]-[{{method}}]-[{{code}}]` |
|
||||
| HTTP P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, path, method, code) (rate(greptime_servers_http_requests_elapsed_bucket{path!~"/health\|/metrics"}[$__rate_interval])))` | `timeseries` | HTTP P99 per Instance. | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{path}}]-[{{method}}]-[{{code}}]-p99` |
|
||||
| gRPC QPS per Instance | `sum by(instance, pod, path, code) (rate(greptime_servers_grpc_requests_elapsed_count{}[$__rate_interval]))` | `timeseries` | gRPC QPS per Instance. | `reqps` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{path}}]-[{{code}}]` |
|
||||
| gRPC P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, path, code) (rate(greptime_servers_grpc_requests_elapsed_bucket{}[$__rate_interval])))` | `timeseries` | gRPC P99 per Instance. | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{path}}]-[{{method}}]-[{{code}}]-p99` |
|
||||
| MySQL QPS per Instance | `sum by(pod, instance)(rate(greptime_servers_mysql_query_elapsed_count{}[$__rate_interval]))` | `timeseries` | MySQL QPS per Instance. | `reqps` | `prometheus` | `[{{instance}}]-[{{pod}}]` |
|
||||
| MySQL P99 per Instance | `histogram_quantile(0.99, sum by(pod, instance, le) (rate(greptime_servers_mysql_query_elapsed_bucket{}[$__rate_interval])))` | `timeseries` | MySQL P99 per Instance. | `s` | `prometheus` | `[{{ instance }}]-[{{ pod }}]-p99` |
|
||||
| PostgreSQL QPS per Instance | `sum by(pod, instance)(rate(greptime_servers_postgres_query_elapsed_count{}[$__rate_interval]))` | `timeseries` | PostgreSQL QPS per Instance. | `reqps` | `prometheus` | `[{{instance}}]-[{{pod}}]` |
|
||||
| PostgreSQL P99 per Instance | `histogram_quantile(0.99, sum by(pod,instance,le) (rate(greptime_servers_postgres_query_elapsed_bucket{}[$__rate_interval])))` | `timeseries` | PostgreSQL P99 per Instance. | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-p99` |
|
||||
# Frontend to Datanode
|
||||
| Title | Query | Type | Description | Datasource | Unit | Legend Format |
|
||||
| --- | --- | --- | --- | --- | --- | --- |
|
||||
| Ingest Rows per Instance | `sum by(instance, pod)(rate(greptime_table_operator_ingest_rows{}[$__rate_interval]))` | `timeseries` | Ingestion rate by row as in each frontend | `prometheus` | `rowsps` | `[{{instance}}]-[{{pod}}]` |
|
||||
| Region Call QPS per Instance | `sum by(instance, pod, request_type) (rate(greptime_grpc_region_request_count{}[$__rate_interval]))` | `timeseries` | Region Call QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{request_type}}]` |
|
||||
| Region Call P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, request_type) (rate(greptime_grpc_region_request_bucket{}[$__rate_interval])))` | `timeseries` | Region Call P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{request_type}}]` |
|
||||
| Ingest Rows per Instance | `sum by(instance, pod)(rate(greptime_table_operator_ingest_rows{}[$__rate_interval]))` | `timeseries` | Ingestion rate by row as in each frontend | `rowsps` | `prometheus` | `[{{instance}}]-[{{pod}}]` |
|
||||
| Region Call QPS per Instance | `sum by(instance, pod, request_type) (rate(greptime_grpc_region_request_count{}[$__rate_interval]))` | `timeseries` | Region Call QPS per Instance. | `ops` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{request_type}}]` |
|
||||
| Region Call P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, request_type) (rate(greptime_grpc_region_request_bucket{}[$__rate_interval])))` | `timeseries` | Region Call P99 per Instance. | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{request_type}}]` |
|
||||
# Mito Engine
|
||||
| Title | Query | Type | Description | Datasource | Unit | Legend Format |
|
||||
| --- | --- | --- | --- | --- | --- | --- |
|
||||
| Request OPS per Instance | `sum by(instance, pod, type) (rate(greptime_mito_handle_request_elapsed_count{}[$__rate_interval]))` | `timeseries` | Request QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{type}}]` |
|
||||
| Request P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, type) (rate(greptime_mito_handle_request_elapsed_bucket{}[$__rate_interval])))` | `timeseries` | Request P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{type}}]` |
|
||||
| Write Buffer per Instance | `greptime_mito_write_buffer_bytes{}` | `timeseries` | Write Buffer per Instance. | `prometheus` | `decbytes` | `[{{instance}}]-[{{pod}}]` |
|
||||
| Write Rows per Instance | `sum by (instance, pod) (rate(greptime_mito_write_rows_total{}[$__rate_interval]))` | `timeseries` | Ingestion size by row counts. | `prometheus` | `rowsps` | `[{{instance}}]-[{{pod}}]` |
|
||||
| Flush OPS per Instance | `sum by(instance, pod, reason) (rate(greptime_mito_flush_requests_total{}[$__rate_interval]))` | `timeseries` | Flush QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{reason}}]` |
|
||||
| Write Stall per Instance | `sum by(instance, pod) (greptime_mito_write_stall_total{})` | `timeseries` | Write Stall per Instance. | `prometheus` | -- | `[{{instance}}]-[{{pod}}]` |
|
||||
| Read Stage OPS per Instance | `sum by(instance, pod) (rate(greptime_mito_read_stage_elapsed_count{ stage="total"}[$__rate_interval]))` | `timeseries` | Read Stage OPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]` |
|
||||
| Read Stage P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, stage) (rate(greptime_mito_read_stage_elapsed_bucket{}[$__rate_interval])))` | `timeseries` | Read Stage P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{stage}}]` |
|
||||
| Write Stage P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, stage) (rate(greptime_mito_write_stage_elapsed_bucket{}[$__rate_interval])))` | `timeseries` | Write Stage P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{stage}}]` |
|
||||
| Compaction OPS per Instance | `sum by(instance, pod) (rate(greptime_mito_compaction_total_elapsed_count{}[$__rate_interval]))` | `timeseries` | Compaction OPS per Instance. | `prometheus` | `ops` | `[{{ instance }}]-[{{pod}}]` |
|
||||
| Compaction P99 per Instance by Stage | `histogram_quantile(0.99, sum by(instance, pod, le, stage) (rate(greptime_mito_compaction_stage_elapsed_bucket{}[$__rate_interval])))` | `timeseries` | Compaction latency by stage | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{stage}}]-p99` |
|
||||
| Compaction P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le,stage) (rate(greptime_mito_compaction_total_elapsed_bucket{}[$__rate_interval])))` | `timeseries` | Compaction P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{stage}}]-compaction` |
|
||||
| WAL write size | `histogram_quantile(0.95, sum by(le,instance, pod) (rate(raft_engine_write_size_bucket[$__rate_interval])))`<br/>`histogram_quantile(0.99, sum by(le,instance,pod) (rate(raft_engine_write_size_bucket[$__rate_interval])))`<br/>`sum by (instance, pod)(rate(raft_engine_write_size_sum[$__rate_interval]))` | `timeseries` | Write-ahead logs write size as bytes. This chart includes stats of p95 and p99 size by instance, total WAL write rate. | `prometheus` | `bytes` | `[{{instance}}]-[{{pod}}]-req-size-p95` |
|
||||
| Cached Bytes per Instance | `greptime_mito_cache_bytes{}` | `timeseries` | Cached Bytes per Instance. | `prometheus` | `decbytes` | `[{{instance}}]-[{{pod}}]-[{{type}}]` |
|
||||
| Inflight Compaction | `greptime_mito_inflight_compaction_count` | `timeseries` | Ongoing compaction task count | `prometheus` | `none` | `[{{instance}}]-[{{pod}}]` |
|
||||
| WAL sync duration seconds | `histogram_quantile(0.99, sum by(le, type, node, instance, pod) (rate(raft_engine_sync_log_duration_seconds_bucket[$__rate_interval])))` | `timeseries` | Raft engine (local disk) log store sync latency, p99 | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-p99` |
|
||||
| Log Store op duration seconds | `histogram_quantile(0.99, sum by(le,logstore,optype,instance, pod) (rate(greptime_logstore_op_elapsed_bucket[$__rate_interval])))` | `timeseries` | Write-ahead log operations latency at p99 | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{logstore}}]-[{{optype}}]-p99` |
|
||||
| Inflight Flush | `greptime_mito_inflight_flush_count` | `timeseries` | Ongoing flush task count | `prometheus` | `none` | `[{{instance}}]-[{{pod}}]` |
|
||||
| Request OPS per Instance | `sum by(instance, pod, type) (rate(greptime_mito_handle_request_elapsed_count{}[$__rate_interval]))` | `timeseries` | Request QPS per Instance. | `ops` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{type}}]` |
|
||||
| Request P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, type) (rate(greptime_mito_handle_request_elapsed_bucket{}[$__rate_interval])))` | `timeseries` | Request P99 per Instance. | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{type}}]` |
|
||||
| Write Buffer per Instance | `greptime_mito_write_buffer_bytes{}` | `timeseries` | Write Buffer per Instance. | `decbytes` | `prometheus` | `[{{instance}}]-[{{pod}}]` |
|
||||
| Write Rows per Instance | `sum by (instance, pod) (rate(greptime_mito_write_rows_total{}[$__rate_interval]))` | `timeseries` | Ingestion size by row counts. | `rowsps` | `prometheus` | `[{{instance}}]-[{{pod}}]` |
|
||||
| Flush OPS per Instance | `sum by(instance, pod, reason) (rate(greptime_mito_flush_requests_total{}[$__rate_interval]))` | `timeseries` | Flush QPS per Instance. | `ops` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{reason}}]` |
|
||||
| Write Stall per Instance | `sum by(instance, pod) (greptime_mito_write_stall_total{})` | `timeseries` | Write Stall per Instance. | `decbytes` | `prometheus` | `[{{instance}}]-[{{pod}}]` |
|
||||
| Read Stage OPS per Instance | `sum by(instance, pod) (rate(greptime_mito_read_stage_elapsed_count{ stage="total"}[$__rate_interval]))` | `timeseries` | Read Stage OPS per Instance. | `ops` | `prometheus` | `[{{instance}}]-[{{pod}}]` |
|
||||
| Read Stage P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, stage) (rate(greptime_mito_read_stage_elapsed_bucket{}[$__rate_interval])))` | `timeseries` | Read Stage P99 per Instance. | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{stage}}]` |
|
||||
| Write Stage P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, stage) (rate(greptime_mito_write_stage_elapsed_bucket{}[$__rate_interval])))` | `timeseries` | Write Stage P99 per Instance. | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{stage}}]` |
|
||||
| Compaction OPS per Instance | `sum by(instance, pod) (rate(greptime_mito_compaction_total_elapsed_count{}[$__rate_interval]))` | `timeseries` | Compaction OPS per Instance. | `ops` | `prometheus` | `[{{ instance }}]-[{{pod}}]` |
|
||||
| Compaction P99 per Instance by Stage | `histogram_quantile(0.99, sum by(instance, pod, le, stage) (rate(greptime_mito_compaction_stage_elapsed_bucket{}[$__rate_interval])))` | `timeseries` | Compaction latency by stage | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{stage}}]-p99` |
|
||||
| Compaction P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le,stage) (rate(greptime_mito_compaction_total_elapsed_bucket{}[$__rate_interval])))` | `timeseries` | Compaction P99 per Instance. | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{stage}}]-compaction` |
|
||||
| WAL write size | `histogram_quantile(0.95, sum by(le,instance, pod) (rate(raft_engine_write_size_bucket[$__rate_interval])))`<br/>`histogram_quantile(0.99, sum by(le,instance,pod) (rate(raft_engine_write_size_bucket[$__rate_interval])))`<br/>`sum by (instance, pod)(rate(raft_engine_write_size_sum[$__rate_interval]))` | `timeseries` | Write-ahead logs write size as bytes. This chart includes stats of p95 and p99 size by instance, total WAL write rate. | `bytes` | `prometheus` | `[{{instance}}]-[{{pod}}]-req-size-p95` |
|
||||
| Cached Bytes per Instance | `greptime_mito_cache_bytes{}` | `timeseries` | Cached Bytes per Instance. | `decbytes` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{type}}]` |
|
||||
| Inflight Compaction | `greptime_mito_inflight_compaction_count` | `timeseries` | Ongoing compaction task count | `none` | `prometheus` | `[{{instance}}]-[{{pod}}]` |
|
||||
| WAL sync duration seconds | `histogram_quantile(0.99, sum by(le, type, node, instance, pod) (rate(raft_engine_sync_log_duration_seconds_bucket[$__rate_interval])))` | `timeseries` | Raft engine (local disk) log store sync latency, p99 | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-p99` |
|
||||
| Log Store op duration seconds | `histogram_quantile(0.99, sum by(le,logstore,optype,instance, pod) (rate(greptime_logstore_op_elapsed_bucket[$__rate_interval])))` | `timeseries` | Write-ahead log operations latency at p99 | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{logstore}}]-[{{optype}}]-p99` |
|
||||
| Inflight Flush | `greptime_mito_inflight_flush_count` | `timeseries` | Ongoing flush task count | `none` | `prometheus` | `[{{instance}}]-[{{pod}}]` |
|
||||
# OpenDAL
|
||||
| Title | Query | Type | Description | Datasource | Unit | Legend Format |
|
||||
| --- | --- | --- | --- | --- | --- | --- |
|
||||
| QPS per Instance | `sum by(instance, pod, scheme, operation) (rate(opendal_operation_duration_seconds_count{}[$__rate_interval]))` | `timeseries` | QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]` |
|
||||
| Read QPS per Instance | `sum by(instance, pod, scheme) (rate(opendal_operation_duration_seconds_count{ operation="read"}[$__rate_interval]))` | `timeseries` | Read QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]` |
|
||||
| Read P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, scheme) (rate(opendal_operation_duration_seconds_bucket{operation="read"}[$__rate_interval])))` | `timeseries` | Read P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-{{scheme}}` |
|
||||
| Write QPS per Instance | `sum by(instance, pod, scheme) (rate(opendal_operation_duration_seconds_count{ operation="write"}[$__rate_interval]))` | `timeseries` | Write QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-{{scheme}}` |
|
||||
| Write P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, scheme) (rate(opendal_operation_duration_seconds_bucket{ operation="write"}[$__rate_interval])))` | `timeseries` | Write P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]` |
|
||||
| List QPS per Instance | `sum by(instance, pod, scheme) (rate(opendal_operation_duration_seconds_count{ operation="list"}[$__rate_interval]))` | `timeseries` | List QPS per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]` |
|
||||
| List P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, scheme) (rate(opendal_operation_duration_seconds_bucket{ operation="list"}[$__rate_interval])))` | `timeseries` | List P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]` |
|
||||
| Other Requests per Instance | `sum by(instance, pod, scheme, operation) (rate(opendal_operation_duration_seconds_count{operation!~"read\|write\|list\|stat"}[$__rate_interval]))` | `timeseries` | Other Requests per Instance. | `prometheus` | `ops` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]` |
|
||||
| Other Request P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, scheme, operation) (rate(opendal_operation_duration_seconds_bucket{ operation!~"read\|write\|list"}[$__rate_interval])))` | `timeseries` | Other Request P99 per Instance. | `prometheus` | `s` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]` |
|
||||
| Opendal traffic | `sum by(instance, pod, scheme, operation) (rate(opendal_operation_bytes_sum{}[$__rate_interval]))` | `timeseries` | Total traffic as in bytes by instance and operation | `prometheus` | `decbytes` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]` |
|
||||
| OpenDAL errors per Instance | `sum by(instance, pod, scheme, operation, error) (rate(opendal_operation_errors_total{ error!="NotFound"}[$__rate_interval]))` | `timeseries` | OpenDAL error counts per Instance. | `prometheus` | -- | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]-[{{error}}]` |
|
||||
| QPS per Instance | `sum by(instance, pod, scheme, operation) (rate(opendal_operation_duration_seconds_count{}[$__rate_interval]))` | `timeseries` | QPS per Instance. | `ops` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]` |
|
||||
| Read QPS per Instance | `sum by(instance, pod, scheme) (rate(opendal_operation_duration_seconds_count{ operation="read"}[$__rate_interval]))` | `timeseries` | Read QPS per Instance. | `ops` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]` |
|
||||
| Read P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, scheme) (rate(opendal_operation_duration_seconds_bucket{operation="read"}[$__rate_interval])))` | `timeseries` | Read P99 per Instance. | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-{{scheme}}` |
|
||||
| Write QPS per Instance | `sum by(instance, pod, scheme) (rate(opendal_operation_duration_seconds_count{ operation="write"}[$__rate_interval]))` | `timeseries` | Write QPS per Instance. | `ops` | `prometheus` | `[{{instance}}]-[{{pod}}]-{{scheme}}` |
|
||||
| Write P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, scheme) (rate(opendal_operation_duration_seconds_bucket{ operation="write"}[$__rate_interval])))` | `timeseries` | Write P99 per Instance. | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]` |
|
||||
| List QPS per Instance | `sum by(instance, pod, scheme) (rate(opendal_operation_duration_seconds_count{ operation="list"}[$__rate_interval]))` | `timeseries` | List QPS per Instance. | `ops` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]` |
|
||||
| List P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, scheme) (rate(opendal_operation_duration_seconds_bucket{ operation="list"}[$__rate_interval])))` | `timeseries` | List P99 per Instance. | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]` |
|
||||
| Other Requests per Instance | `sum by(instance, pod, scheme, operation) (rate(opendal_operation_duration_seconds_count{operation!~"read\|write\|list\|stat"}[$__rate_interval]))` | `timeseries` | Other Requests per Instance. | `ops` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]` |
|
||||
| Other Request P99 per Instance | `histogram_quantile(0.99, sum by(instance, pod, le, scheme, operation) (rate(opendal_operation_duration_seconds_bucket{ operation!~"read\|write\|list"}[$__rate_interval])))` | `timeseries` | Other Request P99 per Instance. | `s` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]` |
|
||||
| Opendal traffic | `sum by(instance, pod, scheme, operation) (rate(opendal_operation_bytes_sum{}[$__rate_interval]))` | `timeseries` | Total traffic as in bytes by instance and operation | `ops` | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]` |
|
||||
# Metasrv
|
||||
| Title | Query | Type | Description | Datasource | Unit | Legend Format |
|
||||
| --- | --- | --- | --- | --- | --- | --- |
|
||||
| Region migration datanode | `greptime_meta_region_migration_stat{datanode_type="src"}`<br/>`greptime_meta_region_migration_stat{datanode_type="desc"}` | `state-timeline` | Counter of region migration by source and destination | `prometheus` | `none` | `from-datanode-{{datanode_id}}` |
|
||||
| Region migration error | `greptime_meta_region_migration_error` | `timeseries` | Counter of region migration error | `prometheus` | `none` | `__auto` |
|
||||
| Datanode load | `greptime_datanode_load` | `timeseries` | Gauge of load information of each datanode, collected via heartbeat between datanode and metasrv. This information is for metasrv to schedule workloads. | `prometheus` | `none` | `__auto` |
|
||||
| Region migration datanode | `greptime_meta_region_migration_stat{datanode_type="src"}`<br/>`greptime_meta_region_migration_stat{datanode_type="desc"}` | `state-timeline` | Counter of region migration by source and destination | `none` | `prometheus` | `from-datanode-{{datanode_id}}` |
|
||||
| Region migration error | `greptime_meta_region_migration_error` | `timeseries` | Counter of region migration error | `none` | `prometheus` | `__auto` |
|
||||
| Datanode load | `greptime_datanode_load` | `timeseries` | Gauge of load information of each datanode, collected via heartbeat between datanode and metasrv. This information is for metasrv to schedule workloads. | `none` | `prometheus` | `__auto` |
|
||||
# Flownode
|
||||
| Title | Query | Type | Description | Datasource | Unit | Legend Format |
|
||||
| --- | --- | --- | --- | --- | --- | --- |
|
||||
| Flow Ingest / Output Rate | `sum by(instance, pod, direction) (rate(greptime_flow_processed_rows[$__rate_interval]))` | `timeseries` | Flow Ingest / Output Rate. | `prometheus` | -- | `[{{pod}}]-[{{instance}}]-[{{direction}}]` |
|
||||
| Flow Ingest Latency | `histogram_quantile(0.95, sum(rate(greptime_flow_insert_elapsed_bucket[$__rate_interval])) by (le, instance, pod))`<br/>`histogram_quantile(0.99, sum(rate(greptime_flow_insert_elapsed_bucket[$__rate_interval])) by (le, instance, pod))` | `timeseries` | Flow Ingest Latency. | `prometheus` | -- | `[{{instance}}]-[{{pod}}]-p95` |
|
||||
| Flow Operation Latency | `histogram_quantile(0.95, sum(rate(greptime_flow_processing_time_bucket[$__rate_interval])) by (le,instance,pod,type))`<br/>`histogram_quantile(0.99, sum(rate(greptime_flow_processing_time_bucket[$__rate_interval])) by (le,instance,pod,type))` | `timeseries` | Flow Operation Latency. | `prometheus` | -- | `[{{instance}}]-[{{pod}}]-[{{type}}]-p95` |
|
||||
| Flow Buffer Size per Instance | `greptime_flow_input_buf_size` | `timeseries` | Flow Buffer Size per Instance. | `prometheus` | -- | `[{{instance}}]-[{{pod}]` |
|
||||
| Flow Processing Error per Instance | `sum by(instance,pod,code) (rate(greptime_flow_errors[$__rate_interval]))` | `timeseries` | Flow Processing Error per Instance. | `prometheus` | -- | `[{{instance}}]-[{{pod}}]-[{{code}}]` |
|
||||
| Flow Ingest / Output Rate | `sum by(instance, pod, direction) (rate(greptime_flow_processed_rows[$__rate_interval]))` | `timeseries` | Flow Ingest / Output Rate. | -- | `prometheus` | `[{{pod}}]-[{{instance}}]-[{{direction}}]` |
|
||||
| Flow Ingest Latency | `histogram_quantile(0.95, sum(rate(greptime_flow_insert_elapsed_bucket[$__rate_interval])) by (le, instance, pod))`<br/>`histogram_quantile(0.99, sum(rate(greptime_flow_insert_elapsed_bucket[$__rate_interval])) by (le, instance, pod))` | `timeseries` | Flow Ingest Latency. | -- | `prometheus` | `[{{instance}}]-[{{pod}}]-p95` |
|
||||
| Flow Operation Latency | `histogram_quantile(0.95, sum(rate(greptime_flow_processing_time_bucket[$__rate_interval])) by (le,instance,pod,type))`<br/>`histogram_quantile(0.99, sum(rate(greptime_flow_processing_time_bucket[$__rate_interval])) by (le,instance,pod,type))` | `timeseries` | Flow Operation Latency. | -- | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{type}}]-p95` |
|
||||
| Flow Buffer Size per Instance | `greptime_flow_input_buf_size` | `timeseries` | Flow Buffer Size per Instance. | -- | `prometheus` | `[{{instance}}]-[{{pod}]` |
|
||||
| Flow Processing Error per Instance | `sum by(instance,pod,code) (rate(greptime_flow_errors[$__rate_interval]))` | `timeseries` | Flow Processing Error per Instance. | -- | `prometheus` | `[{{instance}}]-[{{pod}}]-[{{code}}]` |
|
||||
|
||||
@@ -426,6 +426,7 @@ groups:
|
||||
- title: Write Stall per Instance
|
||||
type: timeseries
|
||||
description: Write Stall per Instance.
|
||||
unit: decbytes
|
||||
queries:
|
||||
- expr: sum by(instance, pod) (greptime_mito_write_stall_total{})
|
||||
datasource:
|
||||
@@ -657,22 +658,13 @@ groups:
|
||||
- title: Opendal traffic
|
||||
type: timeseries
|
||||
description: Total traffic as in bytes by instance and operation
|
||||
unit: decbytes
|
||||
unit: ops
|
||||
queries:
|
||||
- expr: sum by(instance, pod, scheme, operation) (rate(opendal_operation_bytes_sum{}[$__rate_interval]))
|
||||
datasource:
|
||||
type: prometheus
|
||||
uid: ${metrics}
|
||||
legendFormat: '[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]'
|
||||
- title: OpenDAL errors per Instance
|
||||
type: timeseries
|
||||
description: OpenDAL error counts per Instance.
|
||||
queries:
|
||||
- expr: sum by(instance, pod, scheme, operation, error) (rate(opendal_operation_errors_total{ error!="NotFound"}[$__rate_interval]))
|
||||
datasource:
|
||||
type: prometheus
|
||||
uid: ${metrics}
|
||||
legendFormat: '[{{instance}}]-[{{pod}}]-[{{scheme}}]-[{{operation}}]-[{{error}}]'
|
||||
- title: Metasrv
|
||||
panels:
|
||||
- title: Region migration datanode
|
||||
|
||||
@@ -2,7 +2,7 @@
|
||||
|
||||
CLUSTER_DASHBOARD_DIR=${1:-grafana/dashboards/cluster}
|
||||
STANDALONE_DASHBOARD_DIR=${2:-grafana/dashboards/standalone}
|
||||
DAC_IMAGE=ghcr.io/zyy17/dac:20250423-522bd35
|
||||
DAC_IMAGE=ghcr.io/zyy17/dac:20250422-c9435ce
|
||||
|
||||
remove_instance_filters() {
|
||||
# Remove the instance filters for the standalone dashboards.
|
||||
@@ -10,15 +10,8 @@ remove_instance_filters() {
|
||||
}
|
||||
|
||||
generate_intermediate_dashboards_and_docs() {
|
||||
docker run -v ${PWD}:/greptimedb --rm ${DAC_IMAGE} \
|
||||
-i /greptimedb/$CLUSTER_DASHBOARD_DIR/dashboard.json \
|
||||
-o /greptimedb/$CLUSTER_DASHBOARD_DIR/dashboard.yaml \
|
||||
-m /greptimedb/$CLUSTER_DASHBOARD_DIR/dashboard.md
|
||||
|
||||
docker run -v ${PWD}:/greptimedb --rm ${DAC_IMAGE} \
|
||||
-i /greptimedb/$STANDALONE_DASHBOARD_DIR/dashboard.json \
|
||||
-o /greptimedb/$STANDALONE_DASHBOARD_DIR/dashboard.yaml \
|
||||
-m /greptimedb/$STANDALONE_DASHBOARD_DIR/dashboard.md
|
||||
docker run -v ${PWD}:/greptimedb --rm ${DAC_IMAGE} -i /greptimedb/$CLUSTER_DASHBOARD_DIR/dashboard.json -o /greptimedb/$CLUSTER_DASHBOARD_DIR/dashboard.yaml -m > $CLUSTER_DASHBOARD_DIR/dashboard.md
|
||||
docker run -v ${PWD}:/greptimedb --rm ${DAC_IMAGE} -i /greptimedb/$STANDALONE_DASHBOARD_DIR/dashboard.json -o /greptimedb/$STANDALONE_DASHBOARD_DIR/dashboard.yaml -m > $STANDALONE_DASHBOARD_DIR/dashboard.md
|
||||
}
|
||||
|
||||
remove_instance_filters
|
||||
|
||||
@@ -1,2 +1,2 @@
|
||||
[toolchain]
|
||||
channel = "nightly-2025-04-15"
|
||||
channel = "nightly-2024-12-25"
|
||||
|
||||
@@ -84,6 +84,12 @@ mod tests {
|
||||
let key1 = "3178510";
|
||||
let key2 = "4215648";
|
||||
|
||||
// have collision
|
||||
assert_eq!(
|
||||
oid_map.hasher.hash_one(key1) as u32,
|
||||
oid_map.hasher.hash_one(key2) as u32
|
||||
);
|
||||
|
||||
// insert them into oid_map
|
||||
let oid1 = oid_map.get_oid(key1);
|
||||
let oid2 = oid_map.get_oid(key2);
|
||||
|
||||
@@ -132,7 +132,7 @@ impl SubCommand {
|
||||
}
|
||||
|
||||
#[derive(Debug, Default, Parser)]
|
||||
pub struct StartCommand {
|
||||
struct StartCommand {
|
||||
/// The address to bind the gRPC server.
|
||||
#[clap(long, alias = "bind-addr")]
|
||||
rpc_bind_addr: Option<String>,
|
||||
@@ -172,7 +172,7 @@ pub struct StartCommand {
|
||||
}
|
||||
|
||||
impl StartCommand {
|
||||
pub fn load_options(&self, global_options: &GlobalOptions) -> Result<MetasrvOptions> {
|
||||
fn load_options(&self, global_options: &GlobalOptions) -> Result<MetasrvOptions> {
|
||||
let mut opts = MetasrvOptions::load_layered_options(
|
||||
self.config_file.as_deref(),
|
||||
self.env_prefix.as_ref(),
|
||||
@@ -261,7 +261,7 @@ impl StartCommand {
|
||||
Ok(())
|
||||
}
|
||||
|
||||
pub async fn build(&self, opts: MetasrvOptions) -> Result<Instance> {
|
||||
async fn build(&self, opts: MetasrvOptions) -> Result<Instance> {
|
||||
common_runtime::init_global_runtimes(&opts.runtime);
|
||||
|
||||
let guard = common_telemetry::init_global_logging(
|
||||
|
||||
@@ -56,8 +56,8 @@ use datanode::datanode::{Datanode, DatanodeBuilder};
|
||||
use datanode::region_server::RegionServer;
|
||||
use file_engine::config::EngineConfig as FileEngineConfig;
|
||||
use flow::{
|
||||
FlowConfig, FlownodeBuilder, FlownodeInstance, FlownodeOptions, FrontendClient,
|
||||
FrontendInvoker, GrpcQueryHandlerWithBoxedError, StreamingEngine,
|
||||
FlowConfig, FlowStreamingEngine, FlownodeBuilder, FlownodeInstance, FlownodeOptions,
|
||||
FrontendClient, FrontendInvoker, GrpcQueryHandlerWithBoxedError,
|
||||
};
|
||||
use frontend::frontend::{Frontend, FrontendOptions};
|
||||
use frontend::instance::builder::FrontendBuilder;
|
||||
@@ -544,9 +544,9 @@ impl StartCommand {
|
||||
|
||||
// set the ref to query for the local flow state
|
||||
{
|
||||
let flow_streaming_engine = flownode.flow_engine().streaming_engine();
|
||||
let flow_worker_manager = flownode.flow_engine().streaming_engine();
|
||||
information_extension
|
||||
.set_flow_streaming_engine(flow_streaming_engine)
|
||||
.set_flow_worker_manager(flow_worker_manager)
|
||||
.await;
|
||||
}
|
||||
|
||||
@@ -615,10 +615,10 @@ impl StartCommand {
|
||||
.replace(weak_grpc_handler);
|
||||
|
||||
// set the frontend invoker for flownode
|
||||
let flow_streaming_engine = flownode.flow_engine().streaming_engine();
|
||||
let flow_worker_manager = flownode.flow_engine().streaming_engine();
|
||||
// flow server need to be able to use frontend to write insert requests back
|
||||
let invoker = FrontendInvoker::build_from(
|
||||
flow_streaming_engine.clone(),
|
||||
flow_worker_manager.clone(),
|
||||
catalog_manager.clone(),
|
||||
kv_backend.clone(),
|
||||
layered_cache_registry.clone(),
|
||||
@@ -627,7 +627,7 @@ impl StartCommand {
|
||||
)
|
||||
.await
|
||||
.context(error::StartFlownodeSnafu)?;
|
||||
flow_streaming_engine.set_frontend_invoker(invoker).await;
|
||||
flow_worker_manager.set_frontend_invoker(invoker).await;
|
||||
|
||||
let export_metrics_task = ExportMetricsTask::try_new(&opts.export_metrics, Some(&plugins))
|
||||
.context(error::ServersSnafu)?;
|
||||
@@ -703,7 +703,7 @@ pub struct StandaloneInformationExtension {
|
||||
region_server: RegionServer,
|
||||
procedure_manager: ProcedureManagerRef,
|
||||
start_time_ms: u64,
|
||||
flow_streaming_engine: RwLock<Option<Arc<StreamingEngine>>>,
|
||||
flow_worker_manager: RwLock<Option<Arc<FlowStreamingEngine>>>,
|
||||
}
|
||||
|
||||
impl StandaloneInformationExtension {
|
||||
@@ -712,14 +712,14 @@ impl StandaloneInformationExtension {
|
||||
region_server,
|
||||
procedure_manager,
|
||||
start_time_ms: common_time::util::current_time_millis() as u64,
|
||||
flow_streaming_engine: RwLock::new(None),
|
||||
flow_worker_manager: RwLock::new(None),
|
||||
}
|
||||
}
|
||||
|
||||
/// Set the flow streaming engine for the standalone instance.
|
||||
pub async fn set_flow_streaming_engine(&self, flow_streaming_engine: Arc<StreamingEngine>) {
|
||||
let mut guard = self.flow_streaming_engine.write().await;
|
||||
*guard = Some(flow_streaming_engine);
|
||||
/// Set the flow worker manager for the standalone instance.
|
||||
pub async fn set_flow_worker_manager(&self, flow_worker_manager: Arc<FlowStreamingEngine>) {
|
||||
let mut guard = self.flow_worker_manager.write().await;
|
||||
*guard = Some(flow_worker_manager);
|
||||
}
|
||||
}
|
||||
|
||||
@@ -798,7 +798,7 @@ impl InformationExtension for StandaloneInformationExtension {
|
||||
|
||||
async fn flow_stats(&self) -> std::result::Result<Option<FlowStat>, Self::Error> {
|
||||
Ok(Some(
|
||||
self.flow_streaming_engine
|
||||
self.flow_worker_manager
|
||||
.read()
|
||||
.await
|
||||
.as_ref()
|
||||
|
||||
@@ -74,7 +74,6 @@ fn test_load_datanode_example_config() {
|
||||
RegionEngineConfig::File(FileEngineConfig {}),
|
||||
RegionEngineConfig::Metric(MetricEngineConfig {
|
||||
experimental_sparse_primary_key_encoding: false,
|
||||
flush_metadata_region_interval: Duration::from_secs(30),
|
||||
}),
|
||||
],
|
||||
logging: LoggingOptions {
|
||||
@@ -217,7 +216,6 @@ fn test_load_standalone_example_config() {
|
||||
RegionEngineConfig::File(FileEngineConfig {}),
|
||||
RegionEngineConfig::Metric(MetricEngineConfig {
|
||||
experimental_sparse_primary_key_encoding: false,
|
||||
flush_metadata_region_interval: Duration::from_secs(30),
|
||||
}),
|
||||
],
|
||||
storage: StorageConfig {
|
||||
|
||||
@@ -12,9 +12,8 @@
|
||||
// See the License for the specific language governing permissions and
|
||||
// limitations under the License.
|
||||
|
||||
use std::fmt;
|
||||
use std::iter::repeat_n;
|
||||
use std::sync::Arc;
|
||||
use std::{fmt, iter};
|
||||
|
||||
use common_query::error::{InvalidFuncArgsSnafu, Result};
|
||||
use common_query::prelude::Volatility;
|
||||
@@ -127,10 +126,9 @@ impl Function for MatchesTermFunction {
|
||||
let term = term_column.get_ref(0).as_string().unwrap();
|
||||
match term {
|
||||
None => {
|
||||
return Ok(Arc::new(BooleanVector::from_iter(repeat_n(
|
||||
None,
|
||||
text_column.len(),
|
||||
))));
|
||||
return Ok(Arc::new(BooleanVector::from_iter(
|
||||
iter::repeat(None).take(text_column.len()),
|
||||
)));
|
||||
}
|
||||
Some(term) => Some(MatchesTermFinder::new(term)),
|
||||
}
|
||||
@@ -219,7 +217,7 @@ impl MatchesTermFinder {
|
||||
}
|
||||
|
||||
let mut pos = 0;
|
||||
while let Some(found_pos) = self.finder.find(&text.as_bytes()[pos..]) {
|
||||
while let Some(found_pos) = self.finder.find(text[pos..].as_bytes()) {
|
||||
let actual_pos = pos + found_pos;
|
||||
|
||||
let prev_ok = self.starts_with_non_alnum
|
||||
|
||||
@@ -37,7 +37,7 @@ impl fmt::Display for RateFunction {
|
||||
|
||||
impl Function for RateFunction {
|
||||
fn name(&self) -> &str {
|
||||
"rate"
|
||||
"prom_rate"
|
||||
}
|
||||
|
||||
fn return_type(&self, _input_types: &[ConcreteDataType]) -> Result<ConcreteDataType> {
|
||||
@@ -82,7 +82,7 @@ mod tests {
|
||||
#[test]
|
||||
fn test_rate_function() {
|
||||
let rate = RateFunction;
|
||||
assert_eq!("rate", rate.name());
|
||||
assert_eq!("prom_rate", rate.name());
|
||||
assert_eq!(
|
||||
ConcreteDataType::float64_datatype(),
|
||||
rate.return_type(&[]).unwrap()
|
||||
|
||||
@@ -115,13 +115,6 @@ impl Function for UddSketchCalcFunction {
|
||||
}
|
||||
};
|
||||
|
||||
// Check if the sketch is empty, if so, return null
|
||||
// This is important to avoid panics when calling estimate_quantile on an empty sketch
|
||||
// In practice, this will happen if input is all null
|
||||
if sketch.bucket_iter().count() == 0 {
|
||||
builder.push_null();
|
||||
continue;
|
||||
}
|
||||
// Compute the estimated quantile from the sketch
|
||||
let result = sketch.estimate_quantile(perc);
|
||||
builder.push(Some(result));
|
||||
|
||||
@@ -187,7 +187,6 @@ mod tests {
|
||||
},
|
||||
flownode_ids: BTreeMap::from([(0, 1), (1, 2), (2, 3)]),
|
||||
catalog_name: DEFAULT_CATALOG_NAME.to_string(),
|
||||
query_context: None,
|
||||
flow_name: "my_flow".to_string(),
|
||||
raw_sql: "sql".to_string(),
|
||||
expire_after: Some(300),
|
||||
|
||||
@@ -449,7 +449,6 @@ impl From<&CreateFlowData> for (FlowInfoValue, Vec<(FlowPartitionId, FlowRouteVa
|
||||
sink_table_name,
|
||||
flownode_ids,
|
||||
catalog_name,
|
||||
query_context: Some(value.query_context.clone()),
|
||||
flow_name,
|
||||
raw_sql: sql,
|
||||
expire_after,
|
||||
|
||||
@@ -790,14 +790,6 @@ pub enum Error {
|
||||
#[snafu(source)]
|
||||
source: common_procedure::error::Error,
|
||||
},
|
||||
|
||||
#[snafu(display("Failed to parse timezone"))]
|
||||
InvalidTimeZone {
|
||||
#[snafu(implicit)]
|
||||
location: Location,
|
||||
#[snafu(source)]
|
||||
error: common_time::error::Error,
|
||||
},
|
||||
}
|
||||
|
||||
pub type Result<T> = std::result::Result<T, Error>;
|
||||
@@ -869,8 +861,7 @@ impl ErrorExt for Error {
|
||||
| InvalidSetDatabaseOption { .. }
|
||||
| InvalidUnsetDatabaseOption { .. }
|
||||
| InvalidTopicNamePrefix { .. }
|
||||
| InvalidTimeZone { .. } => StatusCode::InvalidArguments,
|
||||
InvalidFlowRequestBody { .. } => StatusCode::InvalidArguments,
|
||||
| InvalidFlowRequestBody { .. } => StatusCode::InvalidArguments,
|
||||
|
||||
FlowNotFound { .. } => StatusCode::FlowNotFound,
|
||||
FlowRouteNotFound { .. } => StatusCode::Unexpected,
|
||||
|
||||
@@ -452,7 +452,6 @@ mod tests {
|
||||
};
|
||||
FlowInfoValue {
|
||||
catalog_name: catalog_name.to_string(),
|
||||
query_context: None,
|
||||
flow_name: flow_name.to_string(),
|
||||
source_table_ids,
|
||||
sink_table_name,
|
||||
@@ -626,7 +625,6 @@ mod tests {
|
||||
};
|
||||
let flow_value = FlowInfoValue {
|
||||
catalog_name: "greptime".to_string(),
|
||||
query_context: None,
|
||||
flow_name: "flow".to_string(),
|
||||
source_table_ids: vec![1024, 1025, 1026],
|
||||
sink_table_name: another_sink_table_name,
|
||||
@@ -866,7 +864,6 @@ mod tests {
|
||||
};
|
||||
let flow_value = FlowInfoValue {
|
||||
catalog_name: "greptime".to_string(),
|
||||
query_context: None,
|
||||
flow_name: "flow".to_string(),
|
||||
source_table_ids: vec![1024, 1025, 1026],
|
||||
sink_table_name: another_sink_table_name,
|
||||
|
||||
@@ -121,13 +121,6 @@ pub struct FlowInfoValue {
|
||||
pub(crate) flownode_ids: BTreeMap<FlowPartitionId, FlownodeId>,
|
||||
/// The catalog name.
|
||||
pub(crate) catalog_name: String,
|
||||
/// The query context used when create flow.
|
||||
/// Although flow doesn't belong to any schema, this query_context is needed to remember
|
||||
/// the query context when `create_flow` is executed
|
||||
/// for recovering flow using the same sql&query_context after db restart.
|
||||
/// if none, should use default query context
|
||||
#[serde(default)]
|
||||
pub(crate) query_context: Option<crate::rpc::ddl::QueryContext>,
|
||||
/// The flow name.
|
||||
pub(crate) flow_name: String,
|
||||
/// The raw sql.
|
||||
@@ -162,10 +155,6 @@ impl FlowInfoValue {
|
||||
&self.catalog_name
|
||||
}
|
||||
|
||||
pub fn query_context(&self) -> &Option<crate::rpc::ddl::QueryContext> {
|
||||
&self.query_context
|
||||
}
|
||||
|
||||
pub fn flow_name(&self) -> &String {
|
||||
&self.flow_name
|
||||
}
|
||||
|
||||
@@ -15,6 +15,8 @@
|
||||
#![feature(assert_matches)]
|
||||
#![feature(btree_extract_if)]
|
||||
#![feature(let_chains)]
|
||||
#![feature(extract_if)]
|
||||
#![feature(hash_extract_if)]
|
||||
|
||||
pub mod cache;
|
||||
pub mod cache_invalidator;
|
||||
|
||||
@@ -113,10 +113,8 @@ impl LeaderRegionManifestInfo {
|
||||
pub fn prunable_entry_id(&self) -> u64 {
|
||||
match self {
|
||||
LeaderRegionManifestInfo::Mito {
|
||||
flushed_entry_id,
|
||||
topic_latest_entry_id,
|
||||
..
|
||||
} => (*flushed_entry_id).max(*topic_latest_entry_id),
|
||||
flushed_entry_id, ..
|
||||
} => *flushed_entry_id,
|
||||
LeaderRegionManifestInfo::Metric {
|
||||
data_flushed_entry_id,
|
||||
data_topic_latest_entry_id,
|
||||
|
||||
@@ -35,20 +35,17 @@ use api::v1::{
|
||||
};
|
||||
use base64::engine::general_purpose;
|
||||
use base64::Engine as _;
|
||||
use common_time::{DatabaseTimeToLive, Timezone};
|
||||
use common_time::DatabaseTimeToLive;
|
||||
use prost::Message;
|
||||
use serde::{Deserialize, Serialize};
|
||||
use serde_with::{serde_as, DefaultOnNull};
|
||||
use session::context::{QueryContextBuilder, QueryContextRef};
|
||||
use session::context::QueryContextRef;
|
||||
use snafu::{OptionExt, ResultExt};
|
||||
use table::metadata::{RawTableInfo, TableId};
|
||||
use table::table_name::TableName;
|
||||
use table::table_reference::TableReference;
|
||||
|
||||
use crate::error::{
|
||||
self, InvalidSetDatabaseOptionSnafu, InvalidTimeZoneSnafu, InvalidUnsetDatabaseOptionSnafu,
|
||||
Result,
|
||||
};
|
||||
use crate::error::{self, InvalidSetDatabaseOptionSnafu, InvalidUnsetDatabaseOptionSnafu, Result};
|
||||
use crate::key::FlowId;
|
||||
|
||||
/// DDL tasks
|
||||
@@ -1205,7 +1202,7 @@ impl From<DropFlowTask> for PbDropFlowTask {
|
||||
}
|
||||
}
|
||||
|
||||
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq)]
|
||||
#[derive(Debug, Clone, Serialize, Deserialize)]
|
||||
pub struct QueryContext {
|
||||
current_catalog: String,
|
||||
current_schema: String,
|
||||
@@ -1226,19 +1223,6 @@ impl From<QueryContextRef> for QueryContext {
|
||||
}
|
||||
}
|
||||
|
||||
impl TryFrom<QueryContext> for session::context::QueryContext {
|
||||
type Error = error::Error;
|
||||
fn try_from(value: QueryContext) -> std::result::Result<Self, Self::Error> {
|
||||
Ok(QueryContextBuilder::default()
|
||||
.current_catalog(value.current_catalog)
|
||||
.current_schema(value.current_schema)
|
||||
.timezone(Timezone::from_tz_string(&value.timezone).context(InvalidTimeZoneSnafu)?)
|
||||
.extensions(value.extensions)
|
||||
.channel((value.channel as u32).into())
|
||||
.build())
|
||||
}
|
||||
}
|
||||
|
||||
impl From<QueryContext> for PbQueryContext {
|
||||
fn from(
|
||||
QueryContext {
|
||||
|
||||
@@ -176,12 +176,15 @@ impl TableRoute {
|
||||
})?
|
||||
.into();
|
||||
|
||||
let leader_peer = peers.get(region_route.leader_peer_index as usize).cloned();
|
||||
let leader_peer = peers
|
||||
.get(region_route.leader_peer_index as usize)
|
||||
.cloned()
|
||||
.map(Into::into);
|
||||
|
||||
let follower_peers = region_route
|
||||
.follower_peer_indexes
|
||||
.into_iter()
|
||||
.filter_map(|x| peers.get(x as usize).cloned())
|
||||
.filter_map(|x| peers.get(x as usize).cloned().map(Into::into))
|
||||
.collect::<Vec<_>>();
|
||||
|
||||
region_routes.push(RegionRoute {
|
||||
|
||||
@@ -24,7 +24,7 @@ use datatypes::prelude::*;
|
||||
use datatypes::vectors::{Helper as VectorHelper, VectorRef};
|
||||
use snafu::ResultExt;
|
||||
|
||||
use crate::error::{self, FromScalarValueSnafu, IntoVectorSnafu, Result};
|
||||
use crate::error::{self, Error, FromScalarValueSnafu, IntoVectorSnafu, Result};
|
||||
use crate::prelude::*;
|
||||
|
||||
pub type AggregateFunctionCreatorRef = Arc<dyn AggregateFunctionCreator>;
|
||||
@@ -166,7 +166,8 @@ impl DfAccumulator for DfAccumulatorAdaptor {
|
||||
let output_type = self.creator.output_type()?;
|
||||
let scalar_value = value
|
||||
.try_to_scalar_value(&output_type)
|
||||
.context(error::ToScalarValueSnafu)?;
|
||||
.context(error::ToScalarValueSnafu)
|
||||
.map_err(Error::from)?;
|
||||
Ok(scalar_value)
|
||||
}
|
||||
|
||||
|
||||
@@ -57,9 +57,9 @@ use tokio::sync::Notify;
|
||||
|
||||
use crate::config::{DatanodeOptions, RegionEngineConfig, StorageConfig};
|
||||
use crate::error::{
|
||||
self, BuildMetricEngineSnafu, BuildMitoEngineSnafu, CreateDirSnafu, GetMetadataSnafu,
|
||||
MissingCacheSnafu, MissingKvBackendSnafu, MissingNodeIdSnafu, OpenLogStoreSnafu, Result,
|
||||
ShutdownInstanceSnafu, ShutdownServerSnafu, StartServerSnafu,
|
||||
self, BuildMitoEngineSnafu, CreateDirSnafu, GetMetadataSnafu, MissingCacheSnafu,
|
||||
MissingKvBackendSnafu, MissingNodeIdSnafu, OpenLogStoreSnafu, Result, ShutdownInstanceSnafu,
|
||||
ShutdownServerSnafu, StartServerSnafu,
|
||||
};
|
||||
use crate::event_listener::{
|
||||
new_region_server_event_channel, NoopRegionServerEventListener, RegionServerEventListenerRef,
|
||||
@@ -416,11 +416,10 @@ impl DatanodeBuilder {
|
||||
)
|
||||
.await?;
|
||||
|
||||
let metric_engine = MetricEngine::try_new(
|
||||
let metric_engine = MetricEngine::new(
|
||||
mito_engine.clone(),
|
||||
metric_engine_config.take().unwrap_or_default(),
|
||||
)
|
||||
.context(BuildMetricEngineSnafu)?;
|
||||
);
|
||||
engines.push(Arc::new(mito_engine) as _);
|
||||
engines.push(Arc::new(metric_engine) as _);
|
||||
}
|
||||
|
||||
@@ -336,13 +336,6 @@ pub enum Error {
|
||||
location: Location,
|
||||
},
|
||||
|
||||
#[snafu(display("Failed to build metric engine"))]
|
||||
BuildMetricEngine {
|
||||
source: metric_engine::error::Error,
|
||||
#[snafu(implicit)]
|
||||
location: Location,
|
||||
},
|
||||
|
||||
#[snafu(display("Failed to serialize options to TOML"))]
|
||||
TomlFormat {
|
||||
#[snafu(implicit)]
|
||||
@@ -459,7 +452,6 @@ impl ErrorExt for Error {
|
||||
|
||||
FindLogicalRegions { source, .. } => source.status_code(),
|
||||
BuildMitoEngine { source, .. } => source.status_code(),
|
||||
BuildMetricEngine { source, .. } => source.status_code(),
|
||||
ConcurrentQueryLimiterClosed { .. } | ConcurrentQueryLimiterTimeout { .. } => {
|
||||
StatusCode::RegionBusy
|
||||
}
|
||||
|
||||
@@ -253,10 +253,9 @@ fn create_current_timestamp_vector(
|
||||
data_type: &ConcreteDataType,
|
||||
num_rows: usize,
|
||||
) -> Result<VectorRef> {
|
||||
let current_timestamp_vector = TimestampMillisecondVector::from_values(std::iter::repeat_n(
|
||||
util::current_time_millis(),
|
||||
num_rows,
|
||||
));
|
||||
let current_timestamp_vector = TimestampMillisecondVector::from_values(
|
||||
std::iter::repeat(util::current_time_millis()).take(num_rows),
|
||||
);
|
||||
if data_type.is_timestamp() {
|
||||
current_timestamp_vector.cast(data_type)
|
||||
} else {
|
||||
|
||||
@@ -198,7 +198,8 @@ impl fmt::Debug for ConstantVector {
|
||||
|
||||
impl Serializable for ConstantVector {
|
||||
fn serialize_to_json(&self) -> Result<Vec<serde_json::Value>> {
|
||||
std::iter::repeat_n(self.get(0), self.len())
|
||||
std::iter::repeat(self.get(0))
|
||||
.take(self.len())
|
||||
.map(serde_json::Value::try_from)
|
||||
.collect::<serde_json::Result<_>>()
|
||||
.context(SerializeSnafu)
|
||||
|
||||
@@ -412,7 +412,7 @@ pub(crate) fn replicate_decimal128(
|
||||
// Safety: std::iter::Repeat and std::iter::Take implement TrustedLen.
|
||||
builder
|
||||
.mutable_array
|
||||
.append_trusted_len_iter(std::iter::repeat_n(data, repeat_times));
|
||||
.append_trusted_len_iter(std::iter::repeat(data).take(repeat_times));
|
||||
}
|
||||
}
|
||||
None => {
|
||||
|
||||
@@ -16,8 +16,8 @@ use std::any::Any;
|
||||
use std::sync::Arc;
|
||||
|
||||
use arrow::array::Array;
|
||||
use arrow::datatypes::Int64Type;
|
||||
use arrow_array::{ArrayRef, DictionaryArray, Int64Array};
|
||||
use arrow::datatypes::Int32Type;
|
||||
use arrow_array::{ArrayRef, DictionaryArray, Int32Array};
|
||||
use serde_json::Value as JsonValue;
|
||||
use snafu::ResultExt;
|
||||
|
||||
@@ -32,7 +32,7 @@ use crate::vectors::{self, Helper, Validity, Vector, VectorRef};
|
||||
/// Vector of dictionaries, basically backed by Arrow's `DictionaryArray`.
|
||||
#[derive(Debug, PartialEq)]
|
||||
pub struct DictionaryVector {
|
||||
array: DictionaryArray<Int64Type>,
|
||||
array: DictionaryArray<Int32Type>,
|
||||
/// The datatype of the items in the dictionary.
|
||||
item_type: ConcreteDataType,
|
||||
/// The vector of items in the dictionary.
|
||||
@@ -41,7 +41,7 @@ pub struct DictionaryVector {
|
||||
|
||||
impl DictionaryVector {
|
||||
/// Create a new instance of `DictionaryVector` from a dictionary array and item type
|
||||
pub fn new(array: DictionaryArray<Int64Type>, item_type: ConcreteDataType) -> Result<Self> {
|
||||
pub fn new(array: DictionaryArray<Int32Type>, item_type: ConcreteDataType) -> Result<Self> {
|
||||
let item_vector = Helper::try_into_vector(array.values())?;
|
||||
|
||||
Ok(Self {
|
||||
@@ -52,12 +52,12 @@ impl DictionaryVector {
|
||||
}
|
||||
|
||||
/// Returns the underlying Arrow dictionary array
|
||||
pub fn array(&self) -> &DictionaryArray<Int64Type> {
|
||||
pub fn array(&self) -> &DictionaryArray<Int32Type> {
|
||||
&self.array
|
||||
}
|
||||
|
||||
/// Returns the keys array of this dictionary
|
||||
pub fn keys(&self) -> &arrow_array::PrimitiveArray<Int64Type> {
|
||||
pub fn keys(&self) -> &arrow_array::PrimitiveArray<Int32Type> {
|
||||
self.array.keys()
|
||||
}
|
||||
|
||||
@@ -74,7 +74,7 @@ impl DictionaryVector {
|
||||
impl Vector for DictionaryVector {
|
||||
fn data_type(&self) -> ConcreteDataType {
|
||||
ConcreteDataType::Dictionary(DictionaryType::new(
|
||||
ConcreteDataType::int64_datatype(),
|
||||
ConcreteDataType::int32_datatype(),
|
||||
self.item_type.clone(),
|
||||
))
|
||||
}
|
||||
@@ -163,10 +163,10 @@ impl Serializable for DictionaryVector {
|
||||
}
|
||||
}
|
||||
|
||||
impl TryFrom<DictionaryArray<Int64Type>> for DictionaryVector {
|
||||
impl TryFrom<DictionaryArray<Int32Type>> for DictionaryVector {
|
||||
type Error = crate::error::Error;
|
||||
|
||||
fn try_from(array: DictionaryArray<Int64Type>) -> Result<Self> {
|
||||
fn try_from(array: DictionaryArray<Int32Type>) -> Result<Self> {
|
||||
let item_type = ConcreteDataType::from_arrow_type(array.values().data_type());
|
||||
let item_vector = Helper::try_into_vector(array.values())?;
|
||||
|
||||
@@ -243,7 +243,7 @@ impl VectorOp for DictionaryVector {
|
||||
previous_offset = offset;
|
||||
}
|
||||
|
||||
let new_keys = Int64Array::from(replicated_keys);
|
||||
let new_keys = Int32Array::from(replicated_keys);
|
||||
let new_array = DictionaryArray::try_new(new_keys, self.values().clone())
|
||||
.expect("Failed to create replicated dictionary array");
|
||||
|
||||
@@ -261,7 +261,7 @@ impl VectorOp for DictionaryVector {
|
||||
let filtered_key_array = filtered_key_vector.to_arrow_array();
|
||||
let filtered_key_array = filtered_key_array
|
||||
.as_any()
|
||||
.downcast_ref::<Int64Array>()
|
||||
.downcast_ref::<Int32Array>()
|
||||
.unwrap();
|
||||
|
||||
let new_array = DictionaryArray::try_new(filtered_key_array.clone(), self.values().clone())
|
||||
@@ -291,7 +291,7 @@ impl VectorOp for DictionaryVector {
|
||||
let key_vector = Helper::try_into_vector(&key_array)?;
|
||||
let new_key_vector = key_vector.take(indices)?;
|
||||
let new_key_array = new_key_vector.to_arrow_array();
|
||||
let new_key_array = new_key_array.as_any().downcast_ref::<Int64Array>().unwrap();
|
||||
let new_key_array = new_key_array.as_any().downcast_ref::<Int32Array>().unwrap();
|
||||
|
||||
let new_array = DictionaryArray::try_new(new_key_array.clone(), self.values().clone())
|
||||
.expect("Failed to create filtered dictionary array");
|
||||
@@ -318,7 +318,7 @@ mod tests {
|
||||
// Keys: [0, 1, 2, null, 1, 3]
|
||||
// Resulting in: ["a", "b", "c", null, "b", "d"]
|
||||
let values = StringArray::from(vec!["a", "b", "c", "d"]);
|
||||
let keys = Int64Array::from(vec![Some(0), Some(1), Some(2), None, Some(1), Some(3)]);
|
||||
let keys = Int32Array::from(vec![Some(0), Some(1), Some(2), None, Some(1), Some(3)]);
|
||||
let dict_array = DictionaryArray::new(keys, Arc::new(values));
|
||||
DictionaryVector::try_from(dict_array).unwrap()
|
||||
}
|
||||
@@ -404,7 +404,7 @@ mod tests {
|
||||
assert_eq!(
|
||||
casted.data_type(),
|
||||
ConcreteDataType::Dictionary(DictionaryType::new(
|
||||
ConcreteDataType::int64_datatype(),
|
||||
ConcreteDataType::int32_datatype(),
|
||||
ConcreteDataType::string_datatype(),
|
||||
))
|
||||
);
|
||||
|
||||
@@ -20,7 +20,7 @@ use std::sync::Arc;
|
||||
use arrow::array::{Array, ArrayRef, StringArray};
|
||||
use arrow::compute;
|
||||
use arrow::compute::kernels::comparison;
|
||||
use arrow::datatypes::{DataType as ArrowDataType, Int64Type, TimeUnit};
|
||||
use arrow::datatypes::{DataType as ArrowDataType, Int32Type, TimeUnit};
|
||||
use arrow_array::DictionaryArray;
|
||||
use arrow_schema::IntervalUnit;
|
||||
use datafusion_common::ScalarValue;
|
||||
@@ -348,11 +348,11 @@ impl Helper {
|
||||
ArrowDataType::Decimal128(_, _) => {
|
||||
Arc::new(Decimal128Vector::try_from_arrow_array(array)?)
|
||||
}
|
||||
ArrowDataType::Dictionary(key, value) if matches!(&**key, ArrowDataType::Int64) => {
|
||||
ArrowDataType::Dictionary(key, value) if matches!(&**key, ArrowDataType::Int32) => {
|
||||
let array = array
|
||||
.as_ref()
|
||||
.as_any()
|
||||
.downcast_ref::<DictionaryArray<Int64Type>>()
|
||||
.downcast_ref::<DictionaryArray<Int32Type>>()
|
||||
.unwrap(); // Safety: the type is guarded by match arm condition
|
||||
Arc::new(DictionaryVector::new(
|
||||
array.clone(),
|
||||
|
||||
@@ -120,7 +120,9 @@ impl fmt::Debug for NullVector {
|
||||
|
||||
impl Serializable for NullVector {
|
||||
fn serialize_to_json(&self) -> Result<Vec<serde_json::Value>> {
|
||||
Ok(std::iter::repeat_n(serde_json::Value::Null, self.len()).collect())
|
||||
Ok(std::iter::repeat(serde_json::Value::Null)
|
||||
.take(self.len())
|
||||
.collect())
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
@@ -388,7 +388,7 @@ pub(crate) fn replicate_primitive<T: LogicalPrimitiveType>(
|
||||
// Safety: std::iter::Repeat and std::iter::Take implement TrustedLen.
|
||||
builder
|
||||
.mutable_array
|
||||
.append_trusted_len_iter(std::iter::repeat_n(data, repeat_times));
|
||||
.append_trusted_len_iter(std::iter::repeat(data).take(repeat_times));
|
||||
}
|
||||
}
|
||||
None => {
|
||||
|
||||
@@ -135,13 +135,14 @@ impl Configurable for FlownodeOptions {
|
||||
}
|
||||
|
||||
/// Arc-ed FlowNodeManager, cheaper to clone
|
||||
pub type FlowStreamingEngineRef = Arc<StreamingEngine>;
|
||||
pub type FlowWorkerManagerRef = Arc<FlowStreamingEngine>;
|
||||
|
||||
/// FlowNodeManager manages the state of all tasks in the flow node, which should be run on the same thread
|
||||
///
|
||||
/// The choice of timestamp is just using current system timestamp for now
|
||||
///
|
||||
pub struct StreamingEngine {
|
||||
/// TODO(discord9): rename to FlowStreamingEngine
|
||||
pub struct FlowStreamingEngine {
|
||||
/// The handler to the worker that will run the dataflow
|
||||
/// which is `!Send` so a handle is used
|
||||
pub worker_handles: Vec<WorkerHandle>,
|
||||
@@ -170,7 +171,7 @@ pub struct StreamingEngine {
|
||||
}
|
||||
|
||||
/// Building FlownodeManager
|
||||
impl StreamingEngine {
|
||||
impl FlowStreamingEngine {
|
||||
/// set frontend invoker
|
||||
pub async fn set_frontend_invoker(&self, frontend: FrontendInvoker) {
|
||||
*self.frontend_invoker.write().await = Some(frontend);
|
||||
@@ -189,7 +190,7 @@ impl StreamingEngine {
|
||||
let node_context = FlownodeContext::new(Box::new(srv_map.clone()) as _);
|
||||
let tick_manager = FlowTickManager::new();
|
||||
let worker_handles = Vec::new();
|
||||
StreamingEngine {
|
||||
FlowStreamingEngine {
|
||||
worker_handles,
|
||||
worker_selector: Mutex::new(0),
|
||||
query_engine,
|
||||
@@ -265,7 +266,7 @@ pub fn batches_to_rows_req(batches: Vec<Batch>) -> Result<Vec<DiffRequest>, Erro
|
||||
}
|
||||
|
||||
/// This impl block contains methods to send writeback requests to frontend
|
||||
impl StreamingEngine {
|
||||
impl FlowStreamingEngine {
|
||||
/// Return the number of requests it made
|
||||
pub async fn send_writeback_requests(&self) -> Result<usize, Error> {
|
||||
let all_reqs = self.generate_writeback_request().await?;
|
||||
@@ -536,7 +537,7 @@ impl StreamingEngine {
|
||||
}
|
||||
|
||||
/// Flow Runtime related methods
|
||||
impl StreamingEngine {
|
||||
impl FlowStreamingEngine {
|
||||
/// Start state report handler, which will receive a sender from HeartbeatTask to send state size report back
|
||||
///
|
||||
/// if heartbeat task is shutdown, this future will exit too
|
||||
@@ -661,7 +662,7 @@ impl StreamingEngine {
|
||||
}
|
||||
// flow is now shutdown, drop frontend_invoker early so a ref cycle(in standalone mode) can be prevent:
|
||||
// FlowWorkerManager.frontend_invoker -> FrontendInvoker.inserter
|
||||
// -> Inserter.node_manager -> NodeManager.flownode -> Flownode.flow_streaming_engine.frontend_invoker
|
||||
// -> Inserter.node_manager -> NodeManager.flownode -> Flownode.flow_worker_manager.frontend_invoker
|
||||
self.frontend_invoker.write().await.take();
|
||||
}
|
||||
|
||||
@@ -730,7 +731,7 @@ impl StreamingEngine {
|
||||
}
|
||||
|
||||
/// Create&Remove flow
|
||||
impl StreamingEngine {
|
||||
impl FlowStreamingEngine {
|
||||
/// remove a flow by it's id
|
||||
pub async fn remove_flow_inner(&self, flow_id: FlowId) -> Result<(), Error> {
|
||||
for handle in self.worker_handles.iter() {
|
||||
@@ -748,6 +749,7 @@ impl StreamingEngine {
|
||||
/// steps to create task:
|
||||
/// 1. parse query into typed plan(and optional parse expire_after expr)
|
||||
/// 2. render source/sink with output table id and used input table id
|
||||
#[allow(clippy::too_many_arguments)]
|
||||
pub async fn create_flow_inner(&self, args: CreateFlowArgs) -> Result<Option<FlowId>, Error> {
|
||||
let CreateFlowArgs {
|
||||
flow_id,
|
||||
|
||||
@@ -35,7 +35,7 @@ use snafu::{ensure, IntoError, OptionExt, ResultExt};
|
||||
use store_api::storage::{RegionId, TableId};
|
||||
use tokio::sync::{Mutex, RwLock};
|
||||
|
||||
use crate::adapter::{CreateFlowArgs, StreamingEngine};
|
||||
use crate::adapter::{CreateFlowArgs, FlowStreamingEngine};
|
||||
use crate::batching_mode::engine::BatchingEngine;
|
||||
use crate::engine::FlowEngine;
|
||||
use crate::error::{
|
||||
@@ -55,7 +55,7 @@ pub type FlowDualEngineRef = Arc<FlowDualEngine>;
|
||||
/// including create/drop/flush flow
|
||||
/// and redirect insert requests to the appropriate engine
|
||||
pub struct FlowDualEngine {
|
||||
streaming_engine: Arc<StreamingEngine>,
|
||||
streaming_engine: Arc<FlowStreamingEngine>,
|
||||
batching_engine: Arc<BatchingEngine>,
|
||||
/// helper struct for faster query flow by table id or vice versa
|
||||
src_table2flow: RwLock<SrcTableToFlow>,
|
||||
@@ -66,7 +66,7 @@ pub struct FlowDualEngine {
|
||||
|
||||
impl FlowDualEngine {
|
||||
pub fn new(
|
||||
streaming_engine: Arc<StreamingEngine>,
|
||||
streaming_engine: Arc<FlowStreamingEngine>,
|
||||
batching_engine: Arc<BatchingEngine>,
|
||||
flow_metadata_manager: Arc<FlowMetadataManager>,
|
||||
catalog_manager: Arc<dyn CatalogManager>,
|
||||
@@ -81,7 +81,7 @@ impl FlowDualEngine {
|
||||
}
|
||||
}
|
||||
|
||||
pub fn streaming_engine(&self) -> Arc<StreamingEngine> {
|
||||
pub fn streaming_engine(&self) -> Arc<FlowStreamingEngine> {
|
||||
self.streaming_engine.clone()
|
||||
}
|
||||
|
||||
@@ -225,24 +225,11 @@ impl FlowDualEngine {
|
||||
comment: Some(info.comment().clone()),
|
||||
sql: info.raw_sql().clone(),
|
||||
flow_options: info.options().clone(),
|
||||
query_ctx: info
|
||||
.query_context()
|
||||
.clone()
|
||||
.map(|ctx| {
|
||||
ctx.try_into()
|
||||
.map_err(BoxedError::new)
|
||||
.context(ExternalSnafu)
|
||||
})
|
||||
.transpose()?
|
||||
// or use default QueryContext with catalog_name from info
|
||||
// to keep compatibility with old version
|
||||
.or_else(|| {
|
||||
Some(
|
||||
QueryContextBuilder::default()
|
||||
.current_catalog(info.catalog_name().to_string())
|
||||
.build(),
|
||||
)
|
||||
}),
|
||||
query_ctx: Some(
|
||||
QueryContextBuilder::default()
|
||||
.current_catalog(info.catalog_name().clone())
|
||||
.build(),
|
||||
),
|
||||
};
|
||||
if let Err(err) = self
|
||||
.create_flow(args)
|
||||
@@ -313,12 +300,11 @@ impl FlowDualEngine {
|
||||
}
|
||||
);
|
||||
|
||||
check_task.take().unwrap().stop().await?;
|
||||
check_task.take().expect("Already checked").stop().await?;
|
||||
info!("Stopped flow consistent check task");
|
||||
Ok(())
|
||||
}
|
||||
|
||||
/// TODO(discord9): also add a `exists` api using flow metadata manager's `exists` method
|
||||
async fn flow_exist_in_metadata(&self, flow_id: FlowId) -> Result<bool, Error> {
|
||||
self.flow_metadata_manager
|
||||
.flow_info_manager()
|
||||
@@ -346,24 +332,23 @@ impl ConsistentCheckTask {
|
||||
let (trigger_tx, mut trigger_rx) =
|
||||
tokio::sync::mpsc::channel::<(bool, bool, tokio::sync::oneshot::Sender<()>)>(10);
|
||||
let handle = common_runtime::spawn_global(async move {
|
||||
let (mut allow_create, mut allow_drop) = (false, false);
|
||||
let mut args = (false, false);
|
||||
let mut ret_signal: Option<tokio::sync::oneshot::Sender<()>> = None;
|
||||
loop {
|
||||
if let Err(err) = inner.check_flow_consistent(allow_create, allow_drop).await {
|
||||
if let Err(err) = inner.check_flow_consistent(args.0, args.1).await {
|
||||
error!(err; "Failed to check flow consistent");
|
||||
}
|
||||
if let Some(done) = ret_signal.take() {
|
||||
let _ = done.send(());
|
||||
}
|
||||
|
||||
tokio::select! {
|
||||
_ = rx.recv() => break,
|
||||
incoming = trigger_rx.recv() => if let Some(incoming) = incoming {
|
||||
(allow_create, allow_drop) = (incoming.0, incoming.1);
|
||||
args = (incoming.0, incoming.1);
|
||||
ret_signal = Some(incoming.2);
|
||||
},
|
||||
_ = tokio::time::sleep(std::time::Duration::from_secs(10)) => {
|
||||
(allow_create, allow_drop) = (false, false);
|
||||
},
|
||||
_ = tokio::time::sleep(std::time::Duration::from_secs(10)) => args=(false,false),
|
||||
}
|
||||
}
|
||||
});
|
||||
@@ -707,7 +692,7 @@ fn to_meta_err(
|
||||
}
|
||||
|
||||
#[async_trait::async_trait]
|
||||
impl common_meta::node_manager::Flownode for StreamingEngine {
|
||||
impl common_meta::node_manager::Flownode for FlowStreamingEngine {
|
||||
async fn handle(&self, request: FlowRequest) -> MetaResult<FlowResponse> {
|
||||
let query_ctx = request
|
||||
.header
|
||||
@@ -793,7 +778,7 @@ impl common_meta::node_manager::Flownode for StreamingEngine {
|
||||
}
|
||||
}
|
||||
|
||||
impl FlowEngine for StreamingEngine {
|
||||
impl FlowEngine for FlowStreamingEngine {
|
||||
async fn create_flow(&self, args: CreateFlowArgs) -> Result<Option<FlowId>, Error> {
|
||||
self.create_flow_inner(args).await
|
||||
}
|
||||
@@ -845,7 +830,7 @@ impl FetchFromRow {
|
||||
}
|
||||
}
|
||||
|
||||
impl StreamingEngine {
|
||||
impl FlowStreamingEngine {
|
||||
async fn handle_inserts_inner(
|
||||
&self,
|
||||
request: InsertRequests,
|
||||
|
||||
@@ -31,7 +31,7 @@ use snafu::{ensure, OptionExt, ResultExt};
|
||||
use table::metadata::TableId;
|
||||
|
||||
use crate::adapter::table_source::ManagedTableSource;
|
||||
use crate::adapter::{FlowId, FlowStreamingEngineRef, StreamingEngine};
|
||||
use crate::adapter::{FlowId, FlowStreamingEngine, FlowWorkerManagerRef};
|
||||
use crate::error::{FlowNotFoundSnafu, JoinTaskSnafu, UnexpectedSnafu};
|
||||
use crate::expr::error::ExternalSnafu;
|
||||
use crate::expr::utils::find_plan_time_window_expr_lower_bound;
|
||||
@@ -39,10 +39,10 @@ use crate::repr::RelationDesc;
|
||||
use crate::server::get_all_flow_ids;
|
||||
use crate::{Error, FrontendInvoker};
|
||||
|
||||
impl StreamingEngine {
|
||||
impl FlowStreamingEngine {
|
||||
/// Create and start refill flow tasks in background
|
||||
pub async fn create_and_start_refill_flow_tasks(
|
||||
self: &FlowStreamingEngineRef,
|
||||
self: &FlowWorkerManagerRef,
|
||||
flow_metadata_manager: &FlowMetadataManagerRef,
|
||||
catalog_manager: &CatalogManagerRef,
|
||||
) -> Result<(), Error> {
|
||||
@@ -130,7 +130,7 @@ impl StreamingEngine {
|
||||
|
||||
/// Starting to refill flows, if any error occurs, will rebuild the flow and retry
|
||||
pub(crate) async fn starting_refill_flows(
|
||||
self: &FlowStreamingEngineRef,
|
||||
self: &FlowWorkerManagerRef,
|
||||
tasks: Vec<RefillTask>,
|
||||
) -> Result<(), Error> {
|
||||
// TODO(discord9): add a back pressure mechanism
|
||||
@@ -266,7 +266,7 @@ impl TaskState<()> {
|
||||
fn start_running(
|
||||
&mut self,
|
||||
task_data: &TaskData,
|
||||
manager: FlowStreamingEngineRef,
|
||||
manager: FlowWorkerManagerRef,
|
||||
mut output_stream: SendableRecordBatchStream,
|
||||
) -> Result<(), Error> {
|
||||
let data = (*task_data).clone();
|
||||
@@ -383,7 +383,7 @@ impl RefillTask {
|
||||
/// Start running the task in background, non-blocking
|
||||
pub async fn start_running(
|
||||
&mut self,
|
||||
manager: FlowStreamingEngineRef,
|
||||
manager: FlowWorkerManagerRef,
|
||||
invoker: &FrontendInvoker,
|
||||
) -> Result<(), Error> {
|
||||
let TaskState::Prepared { sql } = &mut self.state else {
|
||||
|
||||
@@ -16,9 +16,9 @@ use std::collections::BTreeMap;
|
||||
|
||||
use common_meta::key::flow::flow_state::FlowStat;
|
||||
|
||||
use crate::StreamingEngine;
|
||||
use crate::FlowStreamingEngine;
|
||||
|
||||
impl StreamingEngine {
|
||||
impl FlowStreamingEngine {
|
||||
pub async fn gen_state_report(&self) -> FlowStat {
|
||||
let mut full_report = BTreeMap::new();
|
||||
let mut last_exec_time_map = BTreeMap::new();
|
||||
|
||||
@@ -33,8 +33,8 @@ use crate::adapter::table_source::TableDesc;
|
||||
use crate::adapter::{TableName, WorkerHandle, AUTO_CREATED_PLACEHOLDER_TS_COL};
|
||||
use crate::error::{Error, ExternalSnafu, UnexpectedSnafu};
|
||||
use crate::repr::{ColumnType, RelationDesc, RelationType};
|
||||
use crate::StreamingEngine;
|
||||
impl StreamingEngine {
|
||||
use crate::FlowStreamingEngine;
|
||||
impl FlowStreamingEngine {
|
||||
/// Get a worker handle for creating flow, using round robin to select a worker
|
||||
pub(crate) async fn get_worker_handle_for_create_flow(&self) -> &WorkerHandle {
|
||||
let use_idx = {
|
||||
|
||||
@@ -32,9 +32,3 @@ pub const SLOW_QUERY_THRESHOLD: Duration = Duration::from_secs(60);
|
||||
|
||||
/// The minimum duration between two queries execution by batching mode task
|
||||
const MIN_REFRESH_DURATION: Duration = Duration::new(5, 0);
|
||||
|
||||
/// Grpc connection timeout
|
||||
const GRPC_CONN_TIMEOUT: Duration = Duration::from_secs(5);
|
||||
|
||||
/// Grpc max retry number
|
||||
const GRPC_MAX_RETRIES: u32 = 3;
|
||||
|
||||
@@ -267,8 +267,7 @@ impl BatchingEngine {
|
||||
// also check table option to see if ttl!=instant
|
||||
let table_name = get_table_name(self.table_meta.table_info_manager(), &src_id).await?;
|
||||
let table_info = get_table_info(self.table_meta.table_info_manager(), &src_id).await?;
|
||||
ensure!(
|
||||
table_info.table_info.meta.options.ttl != Some(TimeToLive::Instant),
|
||||
if table_info.table_info.meta.options.ttl == Some(TimeToLive::Instant) {
|
||||
UnsupportedSnafu {
|
||||
reason: format!(
|
||||
"Source table `{}`(id={}) has instant TTL, Instant TTL is not supported under batching mode. Consider using a TTL longer than flush interval",
|
||||
@@ -276,8 +275,8 @@ impl BatchingEngine {
|
||||
src_id
|
||||
),
|
||||
}
|
||||
);
|
||||
|
||||
.fail()?;
|
||||
}
|
||||
source_table_names.push(table_name);
|
||||
}
|
||||
|
||||
|
||||
@@ -25,15 +25,12 @@ use common_meta::cluster::{NodeInfo, NodeInfoKey, Role};
|
||||
use common_meta::peer::Peer;
|
||||
use common_meta::rpc::store::RangeRequest;
|
||||
use common_query::Output;
|
||||
use common_telemetry::warn;
|
||||
use meta_client::client::MetaClient;
|
||||
use servers::query_handler::grpc::GrpcQueryHandler;
|
||||
use session::context::{QueryContextBuilder, QueryContextRef};
|
||||
use snafu::{OptionExt, ResultExt};
|
||||
|
||||
use crate::batching_mode::{
|
||||
DEFAULT_BATCHING_ENGINE_QUERY_TIMEOUT, GRPC_CONN_TIMEOUT, GRPC_MAX_RETRIES,
|
||||
};
|
||||
use crate::batching_mode::DEFAULT_BATCHING_ENGINE_QUERY_TIMEOUT;
|
||||
use crate::error::{ExternalSnafu, InvalidRequestSnafu, UnexpectedSnafu};
|
||||
use crate::Error;
|
||||
|
||||
@@ -82,6 +79,7 @@ pub enum FrontendClient {
|
||||
Standalone {
|
||||
/// for the sake of simplicity still use grpc even in standalone mode
|
||||
/// notice the client here should all be lazy, so that can wait after frontend is booted then make conn
|
||||
/// TODO(discord9): not use grpc under standalone mode
|
||||
database_client: HandlerMutable,
|
||||
},
|
||||
}
|
||||
@@ -102,9 +100,7 @@ impl FrontendClient {
|
||||
Self::Distributed {
|
||||
meta_client,
|
||||
chnl_mgr: {
|
||||
let cfg = ChannelConfig::new()
|
||||
.connect_timeout(GRPC_CONN_TIMEOUT)
|
||||
.timeout(DEFAULT_BATCHING_ENGINE_QUERY_TIMEOUT);
|
||||
let cfg = ChannelConfig::new().timeout(DEFAULT_BATCHING_ENGINE_QUERY_TIMEOUT);
|
||||
ChannelManager::with_config(cfg)
|
||||
},
|
||||
}
|
||||
@@ -228,32 +224,12 @@ impl FrontendClient {
|
||||
peer: db.peer.clone(),
|
||||
});
|
||||
|
||||
let mut retry = 0;
|
||||
|
||||
loop {
|
||||
let ret = db.database.handle(req.clone()).await.with_context(|_| {
|
||||
InvalidRequestSnafu {
|
||||
context: format!("Failed to handle request: {:?}", req),
|
||||
}
|
||||
});
|
||||
if let Err(err) = ret {
|
||||
if retry < GRPC_MAX_RETRIES {
|
||||
retry += 1;
|
||||
warn!(
|
||||
"Failed to send request to grpc handle at Peer={:?}, retry = {}, error = {:?}",
|
||||
db.peer, retry, err
|
||||
);
|
||||
continue;
|
||||
} else {
|
||||
common_telemetry::error!(
|
||||
"Failed to send request to grpc handle at Peer={:?} after {} retries, error = {:?}",
|
||||
db.peer, retry, err
|
||||
);
|
||||
return Err(err);
|
||||
}
|
||||
}
|
||||
return ret;
|
||||
}
|
||||
db.database
|
||||
.handle(req.clone())
|
||||
.await
|
||||
.with_context(|_| InvalidRequestSnafu {
|
||||
context: format!("Failed to handle request: {:?}", req),
|
||||
})
|
||||
}
|
||||
FrontendClient::Standalone { database_client } => {
|
||||
let ctx = QueryContextBuilder::default()
|
||||
|
||||
@@ -36,7 +36,7 @@ use operator::expr_helper::column_schemas_to_defs;
|
||||
use query::query_engine::DefaultSerializer;
|
||||
use query::QueryEngineRef;
|
||||
use session::context::QueryContextRef;
|
||||
use snafu::{ensure, OptionExt, ResultExt};
|
||||
use snafu::{OptionExt, ResultExt};
|
||||
use substrait::{DFLogicalSubstraitConvertor, SubstraitPlan};
|
||||
use tokio::sync::oneshot;
|
||||
use tokio::sync::oneshot::error::TryRecvError;
|
||||
@@ -53,7 +53,6 @@ use crate::batching_mode::utils::{
|
||||
use crate::batching_mode::{
|
||||
DEFAULT_BATCHING_ENGINE_QUERY_TIMEOUT, MIN_REFRESH_DURATION, SLOW_QUERY_THRESHOLD,
|
||||
};
|
||||
use crate::df_optimizer::apply_df_optimizer;
|
||||
use crate::error::{
|
||||
ConvertColumnSchemaSnafu, DatafusionSnafu, ExternalSnafu, InvalidQuerySnafu,
|
||||
SubstraitEncodeLogicalPlanSnafu, UnexpectedSnafu,
|
||||
@@ -223,15 +222,15 @@ impl BatchingTask {
|
||||
.map(|c| c.name)
|
||||
.collect::<BTreeSet<_>>();
|
||||
for column in new_query.schema().columns() {
|
||||
ensure!(
|
||||
table_columns.contains(column.name()),
|
||||
InvalidQuerySnafu {
|
||||
if !table_columns.contains(column.name()) {
|
||||
return InvalidQuerySnafu {
|
||||
reason: format!(
|
||||
"Column {} not found in sink table with columns {:?}",
|
||||
column, table_columns
|
||||
),
|
||||
}
|
||||
);
|
||||
.fail();
|
||||
}
|
||||
}
|
||||
// update_at& time index placeholder (if exists) should have default value
|
||||
LogicalPlan::Dml(DmlStatement::new(
|
||||
@@ -542,10 +541,7 @@ impl BatchingTask {
|
||||
.clone()
|
||||
.rewrite(&mut add_auto_column)
|
||||
.with_context(|_| DatafusionSnafu {
|
||||
context: format!(
|
||||
"Failed to rewrite plan:\n {}\n",
|
||||
self.config.plan
|
||||
),
|
||||
context: format!("Failed to rewrite plan {:?}", self.config.plan),
|
||||
})?
|
||||
.data;
|
||||
let schema_len = plan.schema().fields().len();
|
||||
@@ -577,19 +573,16 @@ impl BatchingTask {
|
||||
|
||||
let mut add_filter = AddFilterRewriter::new(expr);
|
||||
let mut add_auto_column = AddAutoColumnRewriter::new(sink_table_schema.clone());
|
||||
|
||||
// make a not optimized plan for clearer unparse
|
||||
let plan = sql_to_df_plan(query_ctx.clone(), engine.clone(), &self.config.query, false)
|
||||
.await?;
|
||||
let rewrite = plan
|
||||
.clone()
|
||||
plan.clone()
|
||||
.rewrite(&mut add_filter)
|
||||
.and_then(|p| p.data.rewrite(&mut add_auto_column))
|
||||
.with_context(|_| DatafusionSnafu {
|
||||
context: format!("Failed to rewrite plan:\n {}\n", plan),
|
||||
context: format!("Failed to rewrite plan {plan:?}"),
|
||||
})?
|
||||
.data;
|
||||
// only apply optimize after complex rewrite is done
|
||||
apply_df_optimizer(rewrite).await?
|
||||
.data
|
||||
};
|
||||
|
||||
Ok(Some((new_plan, schema_len)))
|
||||
|
||||
@@ -704,28 +704,6 @@ mod test {
|
||||
),
|
||||
"SELECT arrow_cast(date_bin(INTERVAL '1 MINS', numbers_with_ts.ts), 'Timestamp(Second, None)') AS time_window FROM numbers_with_ts WHERE ((ts >= CAST('2025-02-24 10:48:00' AS TIMESTAMP)) AND (ts <= CAST('2025-02-24 10:49:00' AS TIMESTAMP))) GROUP BY arrow_cast(date_bin(INTERVAL '1 MINS', numbers_with_ts.ts), 'Timestamp(Second, None)')"
|
||||
),
|
||||
// complex time window index with where
|
||||
(
|
||||
"SELECT arrow_cast(date_bin(INTERVAL '1 MINS', numbers_with_ts.ts), 'Timestamp(Second, None)') AS time_window FROM numbers_with_ts WHERE number in (2, 3, 4) GROUP BY time_window;",
|
||||
Timestamp::new(1740394109, TimeUnit::Second),
|
||||
(
|
||||
"ts".to_string(),
|
||||
Some(Timestamp::new(1740394080, TimeUnit::Second)),
|
||||
Some(Timestamp::new(1740394140, TimeUnit::Second)),
|
||||
),
|
||||
"SELECT arrow_cast(date_bin(INTERVAL '1 MINS', numbers_with_ts.ts), 'Timestamp(Second, None)') AS time_window FROM numbers_with_ts WHERE numbers_with_ts.number IN (2, 3, 4) AND ((ts >= CAST('2025-02-24 10:48:00' AS TIMESTAMP)) AND (ts <= CAST('2025-02-24 10:49:00' AS TIMESTAMP))) GROUP BY arrow_cast(date_bin(INTERVAL '1 MINS', numbers_with_ts.ts), 'Timestamp(Second, None)')"
|
||||
),
|
||||
// complex time window index with between and
|
||||
(
|
||||
"SELECT arrow_cast(date_bin(INTERVAL '1 MINS', numbers_with_ts.ts), 'Timestamp(Second, None)') AS time_window FROM numbers_with_ts WHERE number BETWEEN 2 AND 4 GROUP BY time_window;",
|
||||
Timestamp::new(1740394109, TimeUnit::Second),
|
||||
(
|
||||
"ts".to_string(),
|
||||
Some(Timestamp::new(1740394080, TimeUnit::Second)),
|
||||
Some(Timestamp::new(1740394140, TimeUnit::Second)),
|
||||
),
|
||||
"SELECT arrow_cast(date_bin(INTERVAL '1 MINS', numbers_with_ts.ts), 'Timestamp(Second, None)') AS time_window FROM numbers_with_ts WHERE (numbers_with_ts.number BETWEEN 2 AND 4) AND ((ts >= CAST('2025-02-24 10:48:00' AS TIMESTAMP)) AND (ts <= CAST('2025-02-24 10:49:00' AS TIMESTAMP))) GROUP BY arrow_cast(date_bin(INTERVAL '1 MINS', numbers_with_ts.ts), 'Timestamp(Second, None)')"
|
||||
),
|
||||
// no time index
|
||||
(
|
||||
"SELECT date_bin('5 minutes', ts) FROM numbers_with_ts;",
|
||||
|
||||
@@ -50,8 +50,8 @@ pub async fn get_table_info_df_schema(
|
||||
.await
|
||||
.map_err(BoxedError::new)
|
||||
.context(ExternalSnafu)?
|
||||
.context(TableNotFoundSnafu {
|
||||
name: &full_table_name,
|
||||
.with_context(|| TableNotFoundSnafu {
|
||||
name: full_table_name.clone(),
|
||||
})?;
|
||||
let table_info = table.table_info().clone();
|
||||
|
||||
@@ -342,8 +342,8 @@ impl TreeNodeRewriter for AddAutoColumnRewriter {
|
||||
}
|
||||
} else {
|
||||
return Err(DataFusionError::Plan(format!(
|
||||
"Expect table have 0,1 or 2 columns more than query columns, found {} query columns {:?}, {} table columns {:?}",
|
||||
query_col_cnt, exprs, table_col_cnt, self.schema.column_schemas()
|
||||
"Expect table have 0,1 or 2 columns more than query columns, found {} query columns {:?}, {} table columns {:?} at node {:?}",
|
||||
query_col_cnt, exprs, table_col_cnt, self.schema.column_schemas(), node
|
||||
)));
|
||||
}
|
||||
|
||||
@@ -358,6 +358,8 @@ impl TreeNodeRewriter for AddAutoColumnRewriter {
|
||||
}
|
||||
}
|
||||
|
||||
// TODO(discord9): a method to found out the precise time window
|
||||
|
||||
/// Find out the `Filter` Node corresponding to innermost(deepest) `WHERE` and add a new filter expr to it
|
||||
#[derive(Debug)]
|
||||
pub struct AddFilterRewriter {
|
||||
@@ -406,9 +408,7 @@ mod test {
|
||||
use datatypes::prelude::ConcreteDataType;
|
||||
use datatypes::schema::{ColumnSchema, Schema};
|
||||
use pretty_assertions::assert_eq;
|
||||
use query::query_engine::DefaultSerializer;
|
||||
use session::context::QueryContext;
|
||||
use substrait::{DFLogicalSubstraitConvertor, SubstraitPlan};
|
||||
|
||||
use super::*;
|
||||
use crate::test_utils::create_test_query_engine;
|
||||
@@ -703,18 +703,4 @@ mod test {
|
||||
);
|
||||
}
|
||||
}
|
||||
|
||||
#[tokio::test]
|
||||
async fn test_null_cast() {
|
||||
let query_engine = create_test_query_engine();
|
||||
let ctx = QueryContext::arc();
|
||||
let sql = "SELECT NULL::DOUBLE FROM numbers_with_ts";
|
||||
let plan = sql_to_df_plan(ctx, query_engine.clone(), sql, false)
|
||||
.await
|
||||
.unwrap();
|
||||
|
||||
let _sub_plan = DFLogicalSubstraitConvertor {}
|
||||
.encode(&plan, DefaultSerializer)
|
||||
.unwrap();
|
||||
}
|
||||
}
|
||||
|
||||
@@ -25,6 +25,7 @@ use datafusion::config::ConfigOptions;
|
||||
use datafusion::error::DataFusionError;
|
||||
use datafusion::functions_aggregate::count::count_udaf;
|
||||
use datafusion::functions_aggregate::sum::sum_udaf;
|
||||
use datafusion::optimizer::analyzer::count_wildcard_rule::CountWildcardRule;
|
||||
use datafusion::optimizer::analyzer::type_coercion::TypeCoercion;
|
||||
use datafusion::optimizer::common_subexpr_eliminate::CommonSubexprEliminate;
|
||||
use datafusion::optimizer::optimize_projections::OptimizeProjections;
|
||||
@@ -41,7 +42,6 @@ use datafusion_expr::{
|
||||
BinaryExpr, ColumnarValue, Expr, Operator, Projection, ScalarFunctionArgs, ScalarUDFImpl,
|
||||
Signature, TypeSignature, Volatility,
|
||||
};
|
||||
use query::optimizer::count_wildcard::CountWildcardToTimeIndexRule;
|
||||
use query::parser::QueryLanguageParser;
|
||||
use query::query_engine::DefaultSerializer;
|
||||
use query::QueryEngine;
|
||||
@@ -61,9 +61,9 @@ pub async fn apply_df_optimizer(
|
||||
) -> Result<datafusion_expr::LogicalPlan, Error> {
|
||||
let cfg = ConfigOptions::new();
|
||||
let analyzer = Analyzer::with_rules(vec![
|
||||
Arc::new(CountWildcardToTimeIndexRule),
|
||||
Arc::new(AvgExpandRule),
|
||||
Arc::new(TumbleExpandRule),
|
||||
Arc::new(CountWildcardRule::new()),
|
||||
Arc::new(AvgExpandRule::new()),
|
||||
Arc::new(TumbleExpandRule::new()),
|
||||
Arc::new(CheckGroupByRule::new()),
|
||||
Arc::new(TypeCoercion::new()),
|
||||
]);
|
||||
@@ -128,7 +128,13 @@ pub async fn sql_to_flow_plan(
|
||||
}
|
||||
|
||||
#[derive(Debug)]
|
||||
struct AvgExpandRule;
|
||||
struct AvgExpandRule {}
|
||||
|
||||
impl AvgExpandRule {
|
||||
pub fn new() -> Self {
|
||||
Self {}
|
||||
}
|
||||
}
|
||||
|
||||
impl AnalyzerRule for AvgExpandRule {
|
||||
fn analyze(
|
||||
@@ -325,7 +331,13 @@ impl TreeNodeRewriter for ExpandAvgRewriter<'_> {
|
||||
|
||||
/// expand tumble in aggr expr to tumble_start and tumble_end with column name like `window_start`
|
||||
#[derive(Debug)]
|
||||
struct TumbleExpandRule;
|
||||
struct TumbleExpandRule {}
|
||||
|
||||
impl TumbleExpandRule {
|
||||
pub fn new() -> Self {
|
||||
Self {}
|
||||
}
|
||||
}
|
||||
|
||||
impl AnalyzerRule for TumbleExpandRule {
|
||||
fn analyze(
|
||||
|
||||
@@ -43,7 +43,7 @@ mod utils;
|
||||
#[cfg(test)]
|
||||
mod test_utils;
|
||||
|
||||
pub use adapter::{FlowConfig, FlowStreamingEngineRef, FlownodeOptions, StreamingEngine};
|
||||
pub use adapter::{FlowConfig, FlowStreamingEngine, FlowWorkerManagerRef, FlownodeOptions};
|
||||
pub use batching_mode::frontend_client::{FrontendClient, GrpcQueryHandlerWithBoxedError};
|
||||
pub(crate) use engine::{CreateFlowArgs, FlowId, TableName};
|
||||
pub use error::{Error, Result};
|
||||
|
||||
@@ -52,7 +52,7 @@ use tonic::transport::server::TcpIncoming;
|
||||
use tonic::{Request, Response, Status};
|
||||
|
||||
use crate::adapter::flownode_impl::{FlowDualEngine, FlowDualEngineRef};
|
||||
use crate::adapter::{create_worker, FlowStreamingEngineRef};
|
||||
use crate::adapter::{create_worker, FlowWorkerManagerRef};
|
||||
use crate::batching_mode::engine::BatchingEngine;
|
||||
use crate::engine::FlowEngine;
|
||||
use crate::error::{
|
||||
@@ -63,12 +63,13 @@ use crate::heartbeat::HeartbeatTask;
|
||||
use crate::metrics::{METRIC_FLOW_PROCESSING_TIME, METRIC_FLOW_ROWS};
|
||||
use crate::transform::register_function_to_query_engine;
|
||||
use crate::utils::{SizeReportSender, StateReportHandler};
|
||||
use crate::{CreateFlowArgs, Error, FlownodeOptions, FrontendClient, StreamingEngine};
|
||||
use crate::{CreateFlowArgs, Error, FlowStreamingEngine, FlownodeOptions, FrontendClient};
|
||||
|
||||
pub const FLOW_NODE_SERVER_NAME: &str = "FLOW_NODE_SERVER";
|
||||
/// wrapping flow node manager to avoid orphan rule with Arc<...>
|
||||
#[derive(Clone)]
|
||||
pub struct FlowService {
|
||||
/// TODO(discord9): replace with dual engine
|
||||
pub dual_engine: FlowDualEngineRef,
|
||||
}
|
||||
|
||||
@@ -437,7 +438,6 @@ impl FlownodeBuilder {
|
||||
let cnt = to_be_recovered.len();
|
||||
|
||||
// TODO(discord9): recover in parallel
|
||||
info!("Recovering {} flows: {:?}", cnt, to_be_recovered);
|
||||
for flow_id in to_be_recovered {
|
||||
let info = self
|
||||
.flow_metadata_manager
|
||||
@@ -453,7 +453,6 @@ impl FlownodeBuilder {
|
||||
info.sink_table_name().schema_name.clone(),
|
||||
info.sink_table_name().table_name.clone(),
|
||||
];
|
||||
|
||||
let args = CreateFlowArgs {
|
||||
flow_id: flow_id as _,
|
||||
sink_table_name,
|
||||
@@ -467,24 +466,11 @@ impl FlownodeBuilder {
|
||||
comment: Some(info.comment().clone()),
|
||||
sql: info.raw_sql().clone(),
|
||||
flow_options: info.options().clone(),
|
||||
query_ctx: info
|
||||
.query_context()
|
||||
.clone()
|
||||
.map(|ctx| {
|
||||
ctx.try_into()
|
||||
.map_err(BoxedError::new)
|
||||
.context(ExternalSnafu)
|
||||
})
|
||||
.transpose()?
|
||||
// or use default QueryContext with catalog_name from info
|
||||
// to keep compatibility with old version
|
||||
.or_else(|| {
|
||||
Some(
|
||||
QueryContextBuilder::default()
|
||||
.current_catalog(info.catalog_name().to_string())
|
||||
.build(),
|
||||
)
|
||||
}),
|
||||
query_ctx: Some(
|
||||
QueryContextBuilder::default()
|
||||
.current_catalog(info.catalog_name().clone())
|
||||
.build(),
|
||||
),
|
||||
};
|
||||
manager
|
||||
.create_flow(args)
|
||||
@@ -503,7 +489,7 @@ impl FlownodeBuilder {
|
||||
async fn build_manager(
|
||||
&mut self,
|
||||
query_engine: Arc<dyn QueryEngine>,
|
||||
) -> Result<StreamingEngine, Error> {
|
||||
) -> Result<FlowStreamingEngine, Error> {
|
||||
let table_meta = self.table_meta.clone();
|
||||
|
||||
register_function_to_query_engine(&query_engine);
|
||||
@@ -512,7 +498,7 @@ impl FlownodeBuilder {
|
||||
|
||||
let node_id = self.opts.node_id.map(|id| id as u32);
|
||||
|
||||
let mut man = StreamingEngine::new(node_id, query_engine, table_meta);
|
||||
let mut man = FlowStreamingEngine::new(node_id, query_engine, table_meta);
|
||||
for worker_id in 0..num_workers {
|
||||
let (tx, rx) = oneshot::channel();
|
||||
|
||||
@@ -619,7 +605,7 @@ impl FrontendInvoker {
|
||||
}
|
||||
|
||||
pub async fn build_from(
|
||||
flow_streaming_engine: FlowStreamingEngineRef,
|
||||
flow_worker_manager: FlowWorkerManagerRef,
|
||||
catalog_manager: CatalogManagerRef,
|
||||
kv_backend: KvBackendRef,
|
||||
layered_cache_registry: LayeredCacheRegistryRef,
|
||||
@@ -654,7 +640,7 @@ impl FrontendInvoker {
|
||||
node_manager.clone(),
|
||||
));
|
||||
|
||||
let query_engine = flow_streaming_engine.query_engine.clone();
|
||||
let query_engine = flow_worker_manager.query_engine.clone();
|
||||
|
||||
let statement_executor = Arc::new(StatementExecutor::new(
|
||||
catalog_manager.clone(),
|
||||
|
||||
@@ -46,11 +46,7 @@ pub struct ChineseTokenizer;
|
||||
|
||||
impl Tokenizer for ChineseTokenizer {
|
||||
fn tokenize<'a>(&self, text: &'a str) -> Vec<&'a str> {
|
||||
if text.is_ascii() {
|
||||
EnglishTokenizer {}.tokenize(text)
|
||||
} else {
|
||||
JIEBA.cut(text, false)
|
||||
}
|
||||
JIEBA.cut(text, false)
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
@@ -481,7 +481,7 @@ mod tests {
|
||||
|
||||
let mock_values = dic_values
|
||||
.iter()
|
||||
.flat_map(|(value, size)| std::iter::repeat_n(value.clone(), *size))
|
||||
.flat_map(|(value, size)| iter::repeat(value.clone()).take(*size))
|
||||
.collect::<Vec<_>>();
|
||||
|
||||
let sorted_result = sorted_result(&mock_values, segment_row_count);
|
||||
|
||||
@@ -66,12 +66,10 @@ use crate::election::postgres::PgElection;
|
||||
#[cfg(any(feature = "pg_kvbackend", feature = "mysql_kvbackend"))]
|
||||
use crate::election::CANDIDATE_LEASE_SECS;
|
||||
use crate::metasrv::builder::MetasrvBuilder;
|
||||
use crate::metasrv::{BackendImpl, Metasrv, MetasrvOptions, SelectTarget, SelectorRef};
|
||||
use crate::node_excluder::NodeExcluderRef;
|
||||
use crate::metasrv::{BackendImpl, Metasrv, MetasrvOptions, SelectorRef};
|
||||
use crate::selector::lease_based::LeaseBasedSelector;
|
||||
use crate::selector::load_based::LoadBasedSelector;
|
||||
use crate::selector::round_robin::RoundRobinSelector;
|
||||
use crate::selector::weight_compute::RegionNumsBasedWeightCompute;
|
||||
use crate::selector::SelectorType;
|
||||
use crate::service::admin;
|
||||
use crate::{error, Result};
|
||||
@@ -296,25 +294,14 @@ pub async fn metasrv_builder(
|
||||
|
||||
let in_memory = Arc::new(MemoryKvBackend::new()) as ResettableKvBackendRef;
|
||||
|
||||
let node_excluder = plugins
|
||||
.get::<NodeExcluderRef>()
|
||||
.unwrap_or_else(|| Arc::new(Vec::new()) as NodeExcluderRef);
|
||||
let selector = if let Some(selector) = plugins.get::<SelectorRef>() {
|
||||
info!("Using selector from plugins");
|
||||
selector
|
||||
} else {
|
||||
let selector = match opts.selector {
|
||||
SelectorType::LoadBased => Arc::new(LoadBasedSelector::new(
|
||||
RegionNumsBasedWeightCompute,
|
||||
node_excluder,
|
||||
)) as SelectorRef,
|
||||
SelectorType::LeaseBased => {
|
||||
Arc::new(LeaseBasedSelector::new(node_excluder)) as SelectorRef
|
||||
}
|
||||
SelectorType::RoundRobin => Arc::new(RoundRobinSelector::new(
|
||||
SelectTarget::Datanode,
|
||||
node_excluder,
|
||||
)) as SelectorRef,
|
||||
SelectorType::LoadBased => Arc::new(LoadBasedSelector::default()) as SelectorRef,
|
||||
SelectorType::LeaseBased => Arc::new(LeaseBasedSelector) as SelectorRef,
|
||||
SelectorType::RoundRobin => Arc::new(RoundRobinSelector::default()) as SelectorRef,
|
||||
};
|
||||
info!(
|
||||
"Using selector from options, selector type: {}",
|
||||
|
||||
@@ -14,6 +14,7 @@
|
||||
|
||||
#![feature(result_flattening)]
|
||||
#![feature(assert_matches)]
|
||||
#![feature(extract_if)]
|
||||
#![feature(hash_set_entry)]
|
||||
|
||||
pub mod bootstrap;
|
||||
@@ -30,7 +31,6 @@ pub mod metasrv;
|
||||
pub mod metrics;
|
||||
#[cfg(feature = "mock")]
|
||||
pub mod mocks;
|
||||
pub mod node_excluder;
|
||||
pub mod procedure;
|
||||
pub mod pubsub;
|
||||
pub mod region;
|
||||
|
||||
@@ -111,11 +111,6 @@ pub struct MetasrvOptions {
|
||||
pub use_memory_store: bool,
|
||||
/// Whether to enable region failover.
|
||||
pub enable_region_failover: bool,
|
||||
/// Whether to allow region failover on local WAL.
|
||||
///
|
||||
/// If it's true, the region failover will be allowed even if the local WAL is used.
|
||||
/// Note that this option is not recommended to be set to true, because it may lead to data loss during failover.
|
||||
pub allow_region_failover_on_local_wal: bool,
|
||||
/// The HTTP server options.
|
||||
pub http: HttpOptions,
|
||||
/// The logging options.
|
||||
@@ -178,7 +173,6 @@ impl Default for MetasrvOptions {
|
||||
selector: SelectorType::default(),
|
||||
use_memory_store: false,
|
||||
enable_region_failover: false,
|
||||
allow_region_failover_on_local_wal: false,
|
||||
http: HttpOptions::default(),
|
||||
logging: LoggingOptions {
|
||||
dir: format!("{METASRV_HOME}/logs"),
|
||||
|
||||
@@ -40,8 +40,7 @@ use common_meta::state_store::KvStateStore;
|
||||
use common_meta::wal_options_allocator::{build_kafka_client, build_wal_options_allocator};
|
||||
use common_procedure::local::{LocalManager, ManagerConfig};
|
||||
use common_procedure::ProcedureManagerRef;
|
||||
use common_telemetry::warn;
|
||||
use snafu::{ensure, ResultExt};
|
||||
use snafu::ResultExt;
|
||||
|
||||
use crate::cache_invalidator::MetasrvCacheInvalidator;
|
||||
use crate::cluster::{MetaPeerClientBuilder, MetaPeerClientRef};
|
||||
@@ -191,7 +190,7 @@ impl MetasrvBuilder {
|
||||
|
||||
let meta_peer_client = meta_peer_client
|
||||
.unwrap_or_else(|| build_default_meta_peer_client(&election, &in_memory));
|
||||
let selector = selector.unwrap_or_else(|| Arc::new(LeaseBasedSelector::default()));
|
||||
let selector = selector.unwrap_or_else(|| Arc::new(LeaseBasedSelector));
|
||||
let pushers = Pushers::default();
|
||||
let mailbox = build_mailbox(&kv_backend, &pushers);
|
||||
let procedure_manager = build_procedure_manager(&options, &kv_backend);
|
||||
@@ -235,17 +234,13 @@ impl MetasrvBuilder {
|
||||
))
|
||||
});
|
||||
|
||||
let flow_selector = Arc::new(RoundRobinSelector::new(
|
||||
SelectTarget::Flownode,
|
||||
Arc::new(Vec::new()),
|
||||
)) as SelectorRef;
|
||||
|
||||
let flow_metadata_allocator = {
|
||||
// for now flownode just use round-robin selector
|
||||
let flow_selector = RoundRobinSelector::new(SelectTarget::Flownode);
|
||||
let flow_selector_ctx = selector_ctx.clone();
|
||||
let peer_allocator = Arc::new(FlowPeerAllocator::new(
|
||||
flow_selector_ctx,
|
||||
flow_selector.clone(),
|
||||
Arc::new(flow_selector),
|
||||
));
|
||||
let seq = Arc::new(
|
||||
SequenceBuilder::new(FLOW_ID_SEQ, kv_backend.clone())
|
||||
@@ -277,25 +272,18 @@ impl MetasrvBuilder {
|
||||
},
|
||||
));
|
||||
let peer_lookup_service = Arc::new(MetaPeerLookupService::new(meta_peer_client.clone()));
|
||||
|
||||
if !is_remote_wal && options.enable_region_failover {
|
||||
ensure!(
|
||||
options.allow_region_failover_on_local_wal,
|
||||
error::UnexpectedSnafu {
|
||||
violated: "Region failover is not supported in the local WAL implementation!
|
||||
If you want to enable region failover for local WAL, please set `allow_region_failover_on_local_wal` to true.",
|
||||
}
|
||||
);
|
||||
if options.allow_region_failover_on_local_wal {
|
||||
warn!("Region failover is force enabled in the local WAL implementation! This may lead to data loss during failover!");
|
||||
return error::UnexpectedSnafu {
|
||||
violated: "Region failover is not supported in the local WAL implementation!",
|
||||
}
|
||||
.fail();
|
||||
}
|
||||
|
||||
let (tx, rx) = RegionSupervisor::channel();
|
||||
let (region_failure_detector_controller, region_supervisor_ticker): (
|
||||
RegionFailureDetectorControllerRef,
|
||||
Option<std::sync::Arc<RegionSupervisorTicker>>,
|
||||
) = if options.enable_region_failover {
|
||||
) = if options.enable_region_failover && is_remote_wal {
|
||||
(
|
||||
Arc::new(RegionFailureDetectorControl::new(tx.clone())) as _,
|
||||
Some(Arc::new(RegionSupervisorTicker::new(
|
||||
@@ -321,7 +309,7 @@ impl MetasrvBuilder {
|
||||
));
|
||||
region_migration_manager.try_start()?;
|
||||
|
||||
let region_failover_handler = if options.enable_region_failover {
|
||||
let region_failover_handler = if options.enable_region_failover && is_remote_wal {
|
||||
let region_supervisor = RegionSupervisor::new(
|
||||
rx,
|
||||
options.failure_detector,
|
||||
@@ -432,7 +420,7 @@ impl MetasrvBuilder {
|
||||
meta_peer_client: meta_peer_client.clone(),
|
||||
selector,
|
||||
// TODO(jeremy): We do not allow configuring the flow selector.
|
||||
flow_selector,
|
||||
flow_selector: Arc::new(RoundRobinSelector::new(SelectTarget::Flownode)),
|
||||
handler_group: RwLock::new(None),
|
||||
handler_group_builder: Mutex::new(Some(handler_group_builder)),
|
||||
election,
|
||||
|
||||
@@ -71,13 +71,4 @@ lazy_static! {
|
||||
/// The remote WAL prune execute counter.
|
||||
pub static ref METRIC_META_REMOTE_WAL_PRUNE_EXECUTE: IntCounterVec =
|
||||
register_int_counter_vec!("greptime_meta_remote_wal_prune_execute", "meta remote wal prune execute", &["topic_name"]).unwrap();
|
||||
/// The migration stage elapsed histogram.
|
||||
pub static ref METRIC_META_REGION_MIGRATION_STAGE_ELAPSED: HistogramVec = register_histogram_vec!(
|
||||
"greptime_meta_region_migration_stage_elapsed",
|
||||
"meta region migration stage elapsed",
|
||||
&["stage"],
|
||||
// 0.01 ~ 1000
|
||||
exponential_buckets(0.01, 10.0, 7).unwrap(),
|
||||
)
|
||||
.unwrap();
|
||||
}
|
||||
|
||||
@@ -141,7 +141,10 @@ pub async fn mock(
|
||||
if let Some(client) = client {
|
||||
Ok(TokioIo::new(client))
|
||||
} else {
|
||||
Err(std::io::Error::other("Client already taken"))
|
||||
Err(std::io::Error::new(
|
||||
std::io::ErrorKind::Other,
|
||||
"Client already taken",
|
||||
))
|
||||
}
|
||||
}
|
||||
}),
|
||||
|
||||
@@ -1,32 +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::sync::Arc;
|
||||
|
||||
use common_meta::DatanodeId;
|
||||
|
||||
pub type NodeExcluderRef = Arc<dyn NodeExcluder>;
|
||||
|
||||
/// [NodeExcluder] is used to help decide whether some nodes should be excluded (out of consideration)
|
||||
/// in certain situations. For example, in some node selectors.
|
||||
pub trait NodeExcluder: Send + Sync {
|
||||
/// Returns the excluded datanode ids.
|
||||
fn excluded_datanode_ids(&self) -> &Vec<DatanodeId>;
|
||||
}
|
||||
|
||||
impl NodeExcluder for Vec<DatanodeId> {
|
||||
fn excluded_datanode_ids(&self) -> &Vec<DatanodeId> {
|
||||
self
|
||||
}
|
||||
}
|
||||
@@ -25,7 +25,7 @@ pub(crate) mod update_metadata;
|
||||
pub(crate) mod upgrade_candidate_region;
|
||||
|
||||
use std::any::Any;
|
||||
use std::fmt::{Debug, Display};
|
||||
use std::fmt::Debug;
|
||||
use std::time::Duration;
|
||||
|
||||
use common_error::ext::BoxedError;
|
||||
@@ -43,7 +43,7 @@ use common_procedure::error::{
|
||||
Error as ProcedureError, FromJsonSnafu, Result as ProcedureResult, ToJsonSnafu,
|
||||
};
|
||||
use common_procedure::{Context as ProcedureContext, LockKey, Procedure, Status, StringKey};
|
||||
use common_telemetry::{error, info};
|
||||
use common_telemetry::info;
|
||||
use manager::RegionMigrationProcedureGuard;
|
||||
pub use manager::{
|
||||
RegionMigrationManagerRef, RegionMigrationProcedureTask, RegionMigrationProcedureTracker,
|
||||
@@ -55,10 +55,7 @@ use tokio::time::Instant;
|
||||
|
||||
use self::migration_start::RegionMigrationStart;
|
||||
use crate::error::{self, Result};
|
||||
use crate::metrics::{
|
||||
METRIC_META_REGION_MIGRATION_ERROR, METRIC_META_REGION_MIGRATION_EXECUTE,
|
||||
METRIC_META_REGION_MIGRATION_STAGE_ELAPSED,
|
||||
};
|
||||
use crate::metrics::{METRIC_META_REGION_MIGRATION_ERROR, METRIC_META_REGION_MIGRATION_EXECUTE};
|
||||
use crate::service::mailbox::MailboxRef;
|
||||
|
||||
/// The default timeout for region migration.
|
||||
@@ -106,82 +103,6 @@ impl PersistentContext {
|
||||
}
|
||||
}
|
||||
|
||||
/// Metrics of region migration.
|
||||
#[derive(Debug, Clone, Default)]
|
||||
pub struct Metrics {
|
||||
/// Elapsed time of downgrading region and upgrading region.
|
||||
operations_elapsed: Duration,
|
||||
/// Elapsed time of downgrading leader region.
|
||||
downgrade_leader_region_elapsed: Duration,
|
||||
/// Elapsed time of open candidate region.
|
||||
open_candidate_region_elapsed: Duration,
|
||||
/// Elapsed time of upgrade candidate region.
|
||||
upgrade_candidate_region_elapsed: Duration,
|
||||
}
|
||||
|
||||
impl Display for Metrics {
|
||||
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
|
||||
write!(
|
||||
f,
|
||||
"operations_elapsed: {:?}, downgrade_leader_region_elapsed: {:?}, open_candidate_region_elapsed: {:?}, upgrade_candidate_region_elapsed: {:?}",
|
||||
self.operations_elapsed,
|
||||
self.downgrade_leader_region_elapsed,
|
||||
self.open_candidate_region_elapsed,
|
||||
self.upgrade_candidate_region_elapsed
|
||||
)
|
||||
}
|
||||
}
|
||||
|
||||
impl Metrics {
|
||||
/// Updates the elapsed time of downgrading region and upgrading region.
|
||||
pub fn update_operations_elapsed(&mut self, elapsed: Duration) {
|
||||
self.operations_elapsed += elapsed;
|
||||
}
|
||||
|
||||
/// Updates the elapsed time of downgrading leader region.
|
||||
pub fn update_downgrade_leader_region_elapsed(&mut self, elapsed: Duration) {
|
||||
self.downgrade_leader_region_elapsed += elapsed;
|
||||
}
|
||||
|
||||
/// Updates the elapsed time of open candidate region.
|
||||
pub fn update_open_candidate_region_elapsed(&mut self, elapsed: Duration) {
|
||||
self.open_candidate_region_elapsed += elapsed;
|
||||
}
|
||||
|
||||
/// Updates the elapsed time of upgrade candidate region.
|
||||
pub fn update_upgrade_candidate_region_elapsed(&mut self, elapsed: Duration) {
|
||||
self.upgrade_candidate_region_elapsed += elapsed;
|
||||
}
|
||||
}
|
||||
|
||||
impl Drop for Metrics {
|
||||
fn drop(&mut self) {
|
||||
if !self.operations_elapsed.is_zero() {
|
||||
METRIC_META_REGION_MIGRATION_STAGE_ELAPSED
|
||||
.with_label_values(&["operations"])
|
||||
.observe(self.operations_elapsed.as_secs_f64());
|
||||
}
|
||||
|
||||
if !self.downgrade_leader_region_elapsed.is_zero() {
|
||||
METRIC_META_REGION_MIGRATION_STAGE_ELAPSED
|
||||
.with_label_values(&["downgrade_leader_region"])
|
||||
.observe(self.downgrade_leader_region_elapsed.as_secs_f64());
|
||||
}
|
||||
|
||||
if !self.open_candidate_region_elapsed.is_zero() {
|
||||
METRIC_META_REGION_MIGRATION_STAGE_ELAPSED
|
||||
.with_label_values(&["open_candidate_region"])
|
||||
.observe(self.open_candidate_region_elapsed.as_secs_f64());
|
||||
}
|
||||
|
||||
if !self.upgrade_candidate_region_elapsed.is_zero() {
|
||||
METRIC_META_REGION_MIGRATION_STAGE_ELAPSED
|
||||
.with_label_values(&["upgrade_candidate_region"])
|
||||
.observe(self.upgrade_candidate_region_elapsed.as_secs_f64());
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
/// It's shared in each step and available in executing (including retrying).
|
||||
///
|
||||
/// It will be dropped if the procedure runner crashes.
|
||||
@@ -211,8 +132,8 @@ pub struct VolatileContext {
|
||||
leader_region_last_entry_id: Option<u64>,
|
||||
/// The last_entry_id of leader metadata region (Only used for metric engine).
|
||||
leader_region_metadata_last_entry_id: Option<u64>,
|
||||
/// Metrics of region migration.
|
||||
metrics: Metrics,
|
||||
/// Elapsed time of downgrading region and upgrading region.
|
||||
operations_elapsed: Duration,
|
||||
}
|
||||
|
||||
impl VolatileContext {
|
||||
@@ -310,35 +231,12 @@ impl Context {
|
||||
pub fn next_operation_timeout(&self) -> Option<Duration> {
|
||||
self.persistent_ctx
|
||||
.timeout
|
||||
.checked_sub(self.volatile_ctx.metrics.operations_elapsed)
|
||||
.checked_sub(self.volatile_ctx.operations_elapsed)
|
||||
}
|
||||
|
||||
/// Updates operations elapsed.
|
||||
pub fn update_operations_elapsed(&mut self, instant: Instant) {
|
||||
self.volatile_ctx
|
||||
.metrics
|
||||
.update_operations_elapsed(instant.elapsed());
|
||||
}
|
||||
|
||||
/// Updates the elapsed time of downgrading leader region.
|
||||
pub fn update_downgrade_leader_region_elapsed(&mut self, instant: Instant) {
|
||||
self.volatile_ctx
|
||||
.metrics
|
||||
.update_downgrade_leader_region_elapsed(instant.elapsed());
|
||||
}
|
||||
|
||||
/// Updates the elapsed time of open candidate region.
|
||||
pub fn update_open_candidate_region_elapsed(&mut self, instant: Instant) {
|
||||
self.volatile_ctx
|
||||
.metrics
|
||||
.update_open_candidate_region_elapsed(instant.elapsed());
|
||||
}
|
||||
|
||||
/// Updates the elapsed time of upgrade candidate region.
|
||||
pub fn update_upgrade_candidate_region_elapsed(&mut self, instant: Instant) {
|
||||
self.volatile_ctx
|
||||
.metrics
|
||||
.update_upgrade_candidate_region_elapsed(instant.elapsed());
|
||||
self.volatile_ctx.operations_elapsed += instant.elapsed();
|
||||
}
|
||||
|
||||
/// Returns address of meta server.
|
||||
@@ -652,14 +550,6 @@ impl Procedure for RegionMigrationProcedure {
|
||||
.inc();
|
||||
ProcedureError::retry_later(e)
|
||||
} else {
|
||||
error!(
|
||||
e;
|
||||
"Region migration procedure failed, region_id: {}, from_peer: {}, to_peer: {}, {}",
|
||||
self.context.region_id(),
|
||||
self.context.persistent_ctx.from_peer,
|
||||
self.context.persistent_ctx.to_peer,
|
||||
self.context.volatile_ctx.metrics,
|
||||
);
|
||||
METRIC_META_REGION_MIGRATION_ERROR
|
||||
.with_label_values(&[name, "external"])
|
||||
.inc();
|
||||
|
||||
@@ -46,13 +46,7 @@ impl State for CloseDowngradedRegion {
|
||||
let region_id = ctx.region_id();
|
||||
warn!(err; "Failed to close downgraded leader region: {region_id} on datanode {:?}", downgrade_leader_datanode);
|
||||
}
|
||||
info!(
|
||||
"Region migration is finished: region_id: {}, from_peer: {}, to_peer: {}, {}",
|
||||
ctx.region_id(),
|
||||
ctx.persistent_ctx.from_peer,
|
||||
ctx.persistent_ctx.to_peer,
|
||||
ctx.volatile_ctx.metrics,
|
||||
);
|
||||
|
||||
Ok((Box::new(RegionMigrationEnd), Status::done()))
|
||||
}
|
||||
|
||||
|
||||
@@ -54,7 +54,6 @@ impl Default for DowngradeLeaderRegion {
|
||||
#[typetag::serde]
|
||||
impl State for DowngradeLeaderRegion {
|
||||
async fn next(&mut self, ctx: &mut Context) -> Result<(Box<dyn State>, Status)> {
|
||||
let now = Instant::now();
|
||||
// Ensures the `leader_region_lease_deadline` must exist after recovering.
|
||||
ctx.volatile_ctx
|
||||
.set_leader_region_lease_deadline(Duration::from_secs(REGION_LEASE_SECS));
|
||||
@@ -78,7 +77,6 @@ impl State for DowngradeLeaderRegion {
|
||||
}
|
||||
}
|
||||
}
|
||||
ctx.update_downgrade_leader_region_elapsed(now);
|
||||
|
||||
Ok((
|
||||
Box::new(UpgradeCandidateRegion::default()),
|
||||
@@ -350,8 +348,7 @@ mod tests {
|
||||
let env = TestingEnv::new();
|
||||
let mut ctx = env.context_factory().new_context(persistent_context);
|
||||
prepare_table_metadata(&ctx, HashMap::default()).await;
|
||||
ctx.volatile_ctx.metrics.operations_elapsed =
|
||||
ctx.persistent_ctx.timeout + Duration::from_secs(1);
|
||||
ctx.volatile_ctx.operations_elapsed = ctx.persistent_ctx.timeout + Duration::from_secs(1);
|
||||
|
||||
let err = state.downgrade_region(&mut ctx).await.unwrap_err();
|
||||
|
||||
@@ -594,8 +591,7 @@ mod tests {
|
||||
let mut ctx = env.context_factory().new_context(persistent_context);
|
||||
let mailbox_ctx = env.mailbox_context();
|
||||
let mailbox = mailbox_ctx.mailbox().clone();
|
||||
ctx.volatile_ctx.metrics.operations_elapsed =
|
||||
ctx.persistent_ctx.timeout + Duration::from_secs(1);
|
||||
ctx.volatile_ctx.operations_elapsed = ctx.persistent_ctx.timeout + Duration::from_secs(1);
|
||||
|
||||
let (tx, rx) = tokio::sync::mpsc::channel(1);
|
||||
mailbox_ctx
|
||||
|
||||
@@ -15,7 +15,6 @@
|
||||
use std::any::Any;
|
||||
|
||||
use common_procedure::Status;
|
||||
use common_telemetry::warn;
|
||||
use serde::{Deserialize, Serialize};
|
||||
|
||||
use crate::error::{self, Result};
|
||||
@@ -38,15 +37,7 @@ impl RegionMigrationAbort {
|
||||
#[async_trait::async_trait]
|
||||
#[typetag::serde]
|
||||
impl State for RegionMigrationAbort {
|
||||
async fn next(&mut self, ctx: &mut Context) -> Result<(Box<dyn State>, Status)> {
|
||||
warn!(
|
||||
"Region migration is aborted: {}, region_id: {}, from_peer: {}, to_peer: {}, {}",
|
||||
self.reason,
|
||||
ctx.region_id(),
|
||||
ctx.persistent_ctx.from_peer,
|
||||
ctx.persistent_ctx.to_peer,
|
||||
ctx.volatile_ctx.metrics,
|
||||
);
|
||||
async fn next(&mut self, _: &mut Context) -> Result<(Box<dyn State>, Status)> {
|
||||
error::MigrationAbortSnafu {
|
||||
reason: &self.reason,
|
||||
}
|
||||
|
||||
@@ -13,7 +13,7 @@
|
||||
// limitations under the License.
|
||||
|
||||
use std::any::Any;
|
||||
use std::time::Duration;
|
||||
use std::time::{Duration, Instant};
|
||||
|
||||
use api::v1::meta::MailboxMessage;
|
||||
use common_meta::distributed_time_constants::REGION_LEASE_SECS;
|
||||
@@ -24,7 +24,6 @@ use common_procedure::Status;
|
||||
use common_telemetry::info;
|
||||
use serde::{Deserialize, Serialize};
|
||||
use snafu::{OptionExt, ResultExt};
|
||||
use tokio::time::Instant;
|
||||
|
||||
use crate::error::{self, Result};
|
||||
use crate::handler::HeartbeatMailbox;
|
||||
@@ -43,9 +42,7 @@ pub struct OpenCandidateRegion;
|
||||
impl State for OpenCandidateRegion {
|
||||
async fn next(&mut self, ctx: &mut Context) -> Result<(Box<dyn State>, Status)> {
|
||||
let instruction = self.build_open_region_instruction(ctx).await?;
|
||||
let now = Instant::now();
|
||||
self.open_candidate_region(ctx, instruction).await?;
|
||||
ctx.update_open_candidate_region_elapsed(now);
|
||||
|
||||
Ok((
|
||||
Box::new(UpdateMetadata::Downgrade),
|
||||
|
||||
@@ -54,12 +54,9 @@ impl Default for UpgradeCandidateRegion {
|
||||
#[typetag::serde]
|
||||
impl State for UpgradeCandidateRegion {
|
||||
async fn next(&mut self, ctx: &mut Context) -> Result<(Box<dyn State>, Status)> {
|
||||
let now = Instant::now();
|
||||
if self.upgrade_region_with_retry(ctx).await {
|
||||
ctx.update_upgrade_candidate_region_elapsed(now);
|
||||
Ok((Box::new(UpdateMetadata::Upgrade), Status::executing(false)))
|
||||
} else {
|
||||
ctx.update_upgrade_candidate_region_elapsed(now);
|
||||
Ok((Box::new(UpdateMetadata::Rollback), Status::executing(false)))
|
||||
}
|
||||
}
|
||||
@@ -291,8 +288,7 @@ mod tests {
|
||||
let persistent_context = new_persistent_context();
|
||||
let env = TestingEnv::new();
|
||||
let mut ctx = env.context_factory().new_context(persistent_context);
|
||||
ctx.volatile_ctx.metrics.operations_elapsed =
|
||||
ctx.persistent_ctx.timeout + Duration::from_secs(1);
|
||||
ctx.volatile_ctx.operations_elapsed = ctx.persistent_ctx.timeout + Duration::from_secs(1);
|
||||
|
||||
let err = state.upgrade_region(&ctx).await.unwrap_err();
|
||||
|
||||
@@ -562,8 +558,7 @@ mod tests {
|
||||
let mut ctx = env.context_factory().new_context(persistent_context);
|
||||
let mailbox_ctx = env.mailbox_context();
|
||||
let mailbox = mailbox_ctx.mailbox().clone();
|
||||
ctx.volatile_ctx.metrics.operations_elapsed =
|
||||
ctx.persistent_ctx.timeout + Duration::from_secs(1);
|
||||
ctx.volatile_ctx.operations_elapsed = ctx.persistent_ctx.timeout + Duration::from_secs(1);
|
||||
|
||||
let (tx, rx) = tokio::sync::mpsc::channel(1);
|
||||
mailbox_ctx
|
||||
|
||||
@@ -335,21 +335,22 @@ impl WalPruneProcedure {
|
||||
})?;
|
||||
partition_client
|
||||
.delete_records(
|
||||
// notice here no "+1" is needed because the offset arg is exclusive, and it's defensive programming just in case somewhere else have a off by one error, see https://kafka.apache.org/36/javadoc/org/apache/kafka/clients/consumer/KafkaConsumer.html#endOffsets(java.util.Collection) which we use to get the end offset from high watermark
|
||||
self.data.prunable_entry_id as i64,
|
||||
(self.data.prunable_entry_id + 1) as i64,
|
||||
DELETE_RECORDS_TIMEOUT.as_millis() as i32,
|
||||
)
|
||||
.await
|
||||
.context(DeleteRecordsSnafu {
|
||||
topic: &self.data.topic,
|
||||
partition: DEFAULT_PARTITION,
|
||||
offset: self.data.prunable_entry_id,
|
||||
offset: (self.data.prunable_entry_id + 1),
|
||||
})
|
||||
.map_err(BoxedError::new)
|
||||
.with_context(|_| error::RetryLaterWithSourceSnafu {
|
||||
reason: format!(
|
||||
"Failed to delete records for topic: {}, partition: {}, offset: {}",
|
||||
self.data.topic, DEFAULT_PARTITION, self.data.prunable_entry_id
|
||||
self.data.topic,
|
||||
DEFAULT_PARTITION,
|
||||
self.data.prunable_entry_id + 1
|
||||
),
|
||||
})?;
|
||||
info!(
|
||||
@@ -604,19 +605,19 @@ mod tests {
|
||||
// Step 3: Test `on_prune`.
|
||||
let status = procedure.on_prune().await.unwrap();
|
||||
assert_matches!(status, Status::Done { output: None });
|
||||
// Check if the entry ids after(include) `prunable_entry_id` still exist.
|
||||
// Check if the entry ids after `prunable_entry_id` still exist.
|
||||
check_entry_id_existence(
|
||||
procedure.context.client.clone(),
|
||||
&topic_name,
|
||||
procedure.data.prunable_entry_id as i64 + 1,
|
||||
true,
|
||||
)
|
||||
.await;
|
||||
// Check if the entry s before `prunable_entry_id` are deleted.
|
||||
check_entry_id_existence(
|
||||
procedure.context.client.clone(),
|
||||
&topic_name,
|
||||
procedure.data.prunable_entry_id as i64,
|
||||
true,
|
||||
)
|
||||
.await;
|
||||
// Check if the entry ids before `prunable_entry_id` are deleted.
|
||||
check_entry_id_existence(
|
||||
procedure.context.client.clone(),
|
||||
&topic_name,
|
||||
procedure.data.prunable_entry_id as i64 - 1,
|
||||
false,
|
||||
)
|
||||
.await;
|
||||
|
||||
@@ -12,7 +12,7 @@
|
||||
// See the License for the specific language governing permissions and
|
||||
// limitations under the License.
|
||||
|
||||
use std::collections::{HashMap, HashSet};
|
||||
use std::collections::HashSet;
|
||||
use std::fmt::Debug;
|
||||
use std::sync::{Arc, Mutex};
|
||||
use std::time::Duration;
|
||||
@@ -25,7 +25,7 @@ use common_meta::leadership_notifier::LeadershipChangeListener;
|
||||
use common_meta::peer::PeerLookupServiceRef;
|
||||
use common_meta::DatanodeId;
|
||||
use common_runtime::JoinHandle;
|
||||
use common_telemetry::{debug, error, info, warn};
|
||||
use common_telemetry::{error, info, warn};
|
||||
use common_time::util::current_time_millis;
|
||||
use error::Error::{LeaderPeerChanged, MigrationRunning, TableRouteNotFound};
|
||||
use snafu::{OptionExt, ResultExt};
|
||||
@@ -208,8 +208,6 @@ pub const DEFAULT_TICK_INTERVAL: Duration = Duration::from_secs(1);
|
||||
pub struct RegionSupervisor {
|
||||
/// Used to detect the failure of regions.
|
||||
failure_detector: RegionFailureDetector,
|
||||
/// Tracks the number of failovers for each region.
|
||||
failover_counts: HashMap<DetectingRegion, u32>,
|
||||
/// Receives [Event]s.
|
||||
receiver: Receiver<Event>,
|
||||
/// The context of [`SelectorRef`]
|
||||
@@ -295,7 +293,6 @@ impl RegionSupervisor {
|
||||
) -> Self {
|
||||
Self {
|
||||
failure_detector: RegionFailureDetector::new(options),
|
||||
failover_counts: HashMap::new(),
|
||||
receiver: event_receiver,
|
||||
selector_context,
|
||||
selector,
|
||||
@@ -339,14 +336,13 @@ impl RegionSupervisor {
|
||||
}
|
||||
}
|
||||
|
||||
async fn deregister_failure_detectors(&mut self, detecting_regions: Vec<DetectingRegion>) {
|
||||
async fn deregister_failure_detectors(&self, detecting_regions: Vec<DetectingRegion>) {
|
||||
for region in detecting_regions {
|
||||
self.failure_detector.remove(®ion);
|
||||
self.failover_counts.remove(®ion);
|
||||
self.failure_detector.remove(®ion)
|
||||
}
|
||||
}
|
||||
|
||||
async fn handle_region_failures(&mut self, mut regions: Vec<(DatanodeId, RegionId)>) {
|
||||
async fn handle_region_failures(&self, mut regions: Vec<(DatanodeId, RegionId)>) {
|
||||
if regions.is_empty() {
|
||||
return;
|
||||
}
|
||||
@@ -369,7 +365,8 @@ impl RegionSupervisor {
|
||||
.collect::<Vec<_>>();
|
||||
|
||||
for (datanode_id, region_id) in migrating_regions {
|
||||
debug!(
|
||||
self.failure_detector.remove(&(datanode_id, region_id));
|
||||
warn!(
|
||||
"Removed region failover for region: {region_id}, datanode: {datanode_id} because it's migrating"
|
||||
);
|
||||
}
|
||||
@@ -389,12 +386,7 @@ impl RegionSupervisor {
|
||||
.context(error::MaintenanceModeManagerSnafu)
|
||||
}
|
||||
|
||||
async fn do_failover(&mut self, datanode_id: DatanodeId, region_id: RegionId) -> Result<()> {
|
||||
let count = *self
|
||||
.failover_counts
|
||||
.entry((datanode_id, region_id))
|
||||
.and_modify(|count| *count += 1)
|
||||
.or_insert(1);
|
||||
async fn do_failover(&self, datanode_id: DatanodeId, region_id: RegionId) -> Result<()> {
|
||||
let from_peer = self
|
||||
.peer_lookup
|
||||
.datanode(datanode_id)
|
||||
@@ -423,14 +415,11 @@ impl RegionSupervisor {
|
||||
);
|
||||
return Ok(());
|
||||
}
|
||||
info!(
|
||||
"Failover for region: {region_id}, from_peer: {from_peer}, to_peer: {to_peer}, tries: {count}"
|
||||
);
|
||||
let task = RegionMigrationProcedureTask {
|
||||
region_id,
|
||||
from_peer,
|
||||
to_peer,
|
||||
timeout: DEFAULT_REGION_MIGRATION_TIMEOUT * count,
|
||||
timeout: DEFAULT_REGION_MIGRATION_TIMEOUT,
|
||||
};
|
||||
|
||||
if let Err(err) = self.region_migration_manager.submit_procedure(task).await {
|
||||
@@ -444,8 +433,7 @@ impl RegionSupervisor {
|
||||
Ok(())
|
||||
}
|
||||
TableRouteNotFound { .. } => {
|
||||
self.deregister_failure_detectors(vec![(datanode_id, region_id)])
|
||||
.await;
|
||||
self.failure_detector.remove(&(datanode_id, region_id));
|
||||
info!(
|
||||
"Table route is not found, the table is dropped, removed failover detector for region: {}, datanode: {}",
|
||||
region_id, datanode_id
|
||||
@@ -453,8 +441,7 @@ impl RegionSupervisor {
|
||||
Ok(())
|
||||
}
|
||||
LeaderPeerChanged { .. } => {
|
||||
self.deregister_failure_detectors(vec![(datanode_id, region_id)])
|
||||
.await;
|
||||
self.failure_detector.remove(&(datanode_id, region_id));
|
||||
info!(
|
||||
"Region's leader peer changed, removed failover detector for region: {}, datanode: {}",
|
||||
region_id, datanode_id
|
||||
|
||||
@@ -18,7 +18,7 @@ pub mod load_based;
|
||||
pub mod round_robin;
|
||||
#[cfg(test)]
|
||||
pub(crate) mod test_utils;
|
||||
pub mod weight_compute;
|
||||
mod weight_compute;
|
||||
pub mod weighted_choose;
|
||||
use std::collections::HashSet;
|
||||
|
||||
|
||||
@@ -12,37 +12,17 @@
|
||||
// See the License for the specific language governing permissions and
|
||||
// limitations under the License.
|
||||
|
||||
use std::collections::HashSet;
|
||||
use std::sync::Arc;
|
||||
|
||||
use common_meta::peer::Peer;
|
||||
|
||||
use crate::error::Result;
|
||||
use crate::lease;
|
||||
use crate::metasrv::SelectorContext;
|
||||
use crate::node_excluder::NodeExcluderRef;
|
||||
use crate::selector::common::{choose_items, filter_out_excluded_peers};
|
||||
use crate::selector::weighted_choose::{RandomWeightedChoose, WeightedItem};
|
||||
use crate::selector::{Selector, SelectorOptions};
|
||||
|
||||
/// Select all alive datanodes based using a random weighted choose.
|
||||
pub struct LeaseBasedSelector {
|
||||
node_excluder: NodeExcluderRef,
|
||||
}
|
||||
|
||||
impl LeaseBasedSelector {
|
||||
pub fn new(node_excluder: NodeExcluderRef) -> Self {
|
||||
Self { node_excluder }
|
||||
}
|
||||
}
|
||||
|
||||
impl Default for LeaseBasedSelector {
|
||||
fn default() -> Self {
|
||||
Self {
|
||||
node_excluder: Arc::new(Vec::new()),
|
||||
}
|
||||
}
|
||||
}
|
||||
pub struct LeaseBasedSelector;
|
||||
|
||||
#[async_trait::async_trait]
|
||||
impl Selector for LeaseBasedSelector {
|
||||
@@ -67,14 +47,7 @@ impl Selector for LeaseBasedSelector {
|
||||
.collect();
|
||||
|
||||
// 3. choose peers by weight_array.
|
||||
let mut exclude_peer_ids = self
|
||||
.node_excluder
|
||||
.excluded_datanode_ids()
|
||||
.iter()
|
||||
.cloned()
|
||||
.collect::<HashSet<_>>();
|
||||
exclude_peer_ids.extend(opts.exclude_peer_ids.iter());
|
||||
filter_out_excluded_peers(&mut weight_array, &exclude_peer_ids);
|
||||
filter_out_excluded_peers(&mut weight_array, &opts.exclude_peer_ids);
|
||||
let mut weighted_choose = RandomWeightedChoose::new(weight_array);
|
||||
let selected = choose_items(&opts, &mut weighted_choose)?;
|
||||
|
||||
|
||||
@@ -12,8 +12,7 @@
|
||||
// See the License for the specific language governing permissions and
|
||||
// limitations under the License.
|
||||
|
||||
use std::collections::{HashMap, HashSet};
|
||||
use std::sync::Arc;
|
||||
use std::collections::HashMap;
|
||||
|
||||
use common_meta::datanode::{DatanodeStatKey, DatanodeStatValue};
|
||||
use common_meta::key::TableMetadataManager;
|
||||
@@ -27,7 +26,6 @@ use crate::error::{self, Result};
|
||||
use crate::key::{DatanodeLeaseKey, LeaseValue};
|
||||
use crate::lease;
|
||||
use crate::metasrv::SelectorContext;
|
||||
use crate::node_excluder::NodeExcluderRef;
|
||||
use crate::selector::common::{choose_items, filter_out_excluded_peers};
|
||||
use crate::selector::weight_compute::{RegionNumsBasedWeightCompute, WeightCompute};
|
||||
use crate::selector::weighted_choose::RandomWeightedChoose;
|
||||
@@ -35,15 +33,11 @@ use crate::selector::{Selector, SelectorOptions};
|
||||
|
||||
pub struct LoadBasedSelector<C> {
|
||||
weight_compute: C,
|
||||
node_excluder: NodeExcluderRef,
|
||||
}
|
||||
|
||||
impl<C> LoadBasedSelector<C> {
|
||||
pub fn new(weight_compute: C, node_excluder: NodeExcluderRef) -> Self {
|
||||
Self {
|
||||
weight_compute,
|
||||
node_excluder,
|
||||
}
|
||||
pub fn new(weight_compute: C) -> Self {
|
||||
Self { weight_compute }
|
||||
}
|
||||
}
|
||||
|
||||
@@ -51,7 +45,6 @@ impl Default for LoadBasedSelector<RegionNumsBasedWeightCompute> {
|
||||
fn default() -> Self {
|
||||
Self {
|
||||
weight_compute: RegionNumsBasedWeightCompute,
|
||||
node_excluder: Arc::new(Vec::new()),
|
||||
}
|
||||
}
|
||||
}
|
||||
@@ -95,14 +88,7 @@ where
|
||||
let mut weight_array = self.weight_compute.compute(&stat_kvs);
|
||||
|
||||
// 5. choose peers by weight_array.
|
||||
let mut exclude_peer_ids = self
|
||||
.node_excluder
|
||||
.excluded_datanode_ids()
|
||||
.iter()
|
||||
.cloned()
|
||||
.collect::<HashSet<_>>();
|
||||
exclude_peer_ids.extend(opts.exclude_peer_ids.iter());
|
||||
filter_out_excluded_peers(&mut weight_array, &exclude_peer_ids);
|
||||
filter_out_excluded_peers(&mut weight_array, &opts.exclude_peer_ids);
|
||||
let mut weighted_choose = RandomWeightedChoose::new(weight_array);
|
||||
let selected = choose_items(&opts, &mut weighted_choose)?;
|
||||
|
||||
|
||||
@@ -12,9 +12,7 @@
|
||||
// See the License for the specific language governing permissions and
|
||||
// limitations under the License.
|
||||
|
||||
use std::collections::HashSet;
|
||||
use std::sync::atomic::AtomicUsize;
|
||||
use std::sync::Arc;
|
||||
|
||||
use common_meta::peer::Peer;
|
||||
use snafu::ensure;
|
||||
@@ -22,7 +20,6 @@ use snafu::ensure;
|
||||
use crate::error::{NoEnoughAvailableNodeSnafu, Result};
|
||||
use crate::lease;
|
||||
use crate::metasrv::{SelectTarget, SelectorContext};
|
||||
use crate::node_excluder::NodeExcluderRef;
|
||||
use crate::selector::{Selector, SelectorOptions};
|
||||
|
||||
/// Round-robin selector that returns the next peer in the list in sequence.
|
||||
@@ -35,7 +32,6 @@ use crate::selector::{Selector, SelectorOptions};
|
||||
pub struct RoundRobinSelector {
|
||||
select_target: SelectTarget,
|
||||
counter: AtomicUsize,
|
||||
node_excluder: NodeExcluderRef,
|
||||
}
|
||||
|
||||
impl Default for RoundRobinSelector {
|
||||
@@ -43,38 +39,32 @@ impl Default for RoundRobinSelector {
|
||||
Self {
|
||||
select_target: SelectTarget::Datanode,
|
||||
counter: AtomicUsize::new(0),
|
||||
node_excluder: Arc::new(Vec::new()),
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
impl RoundRobinSelector {
|
||||
pub fn new(select_target: SelectTarget, node_excluder: NodeExcluderRef) -> Self {
|
||||
pub fn new(select_target: SelectTarget) -> Self {
|
||||
Self {
|
||||
select_target,
|
||||
node_excluder,
|
||||
..Default::default()
|
||||
}
|
||||
}
|
||||
|
||||
async fn get_peers(&self, opts: &SelectorOptions, ctx: &SelectorContext) -> Result<Vec<Peer>> {
|
||||
async fn get_peers(
|
||||
&self,
|
||||
min_required_items: usize,
|
||||
ctx: &SelectorContext,
|
||||
) -> Result<Vec<Peer>> {
|
||||
let mut peers = match self.select_target {
|
||||
SelectTarget::Datanode => {
|
||||
// 1. get alive datanodes.
|
||||
let lease_kvs =
|
||||
lease::alive_datanodes(&ctx.meta_peer_client, ctx.datanode_lease_secs).await?;
|
||||
|
||||
let mut exclude_peer_ids = self
|
||||
.node_excluder
|
||||
.excluded_datanode_ids()
|
||||
.iter()
|
||||
.cloned()
|
||||
.collect::<HashSet<_>>();
|
||||
exclude_peer_ids.extend(opts.exclude_peer_ids.iter());
|
||||
// 2. map into peers
|
||||
lease_kvs
|
||||
.into_iter()
|
||||
.filter(|(k, _)| !exclude_peer_ids.contains(&k.node_id))
|
||||
.map(|(k, v)| Peer::new(k.node_id, v.node_addr))
|
||||
.collect::<Vec<_>>()
|
||||
}
|
||||
@@ -94,8 +84,8 @@ impl RoundRobinSelector {
|
||||
ensure!(
|
||||
!peers.is_empty(),
|
||||
NoEnoughAvailableNodeSnafu {
|
||||
required: opts.min_required_items,
|
||||
available: peers.len(),
|
||||
required: min_required_items,
|
||||
available: 0usize,
|
||||
select_target: self.select_target
|
||||
}
|
||||
);
|
||||
@@ -113,7 +103,7 @@ impl Selector for RoundRobinSelector {
|
||||
type Output = Vec<Peer>;
|
||||
|
||||
async fn select(&self, ctx: &Self::Context, opts: SelectorOptions) -> Result<Vec<Peer>> {
|
||||
let peers = self.get_peers(&opts, ctx).await?;
|
||||
let peers = self.get_peers(opts.min_required_items, ctx).await?;
|
||||
// choose peers
|
||||
let mut selected = Vec::with_capacity(opts.min_required_items);
|
||||
for _ in 0..opts.min_required_items {
|
||||
@@ -186,42 +176,4 @@ mod test {
|
||||
assert_eq!(peers.len(), 2);
|
||||
assert_eq!(peers, vec![peer2.clone(), peer3.clone()]);
|
||||
}
|
||||
|
||||
#[tokio::test]
|
||||
async fn test_round_robin_selector_with_exclude_peer_ids() {
|
||||
let selector = RoundRobinSelector::new(SelectTarget::Datanode, Arc::new(vec![5]));
|
||||
let ctx = create_selector_context();
|
||||
// add three nodes
|
||||
let peer1 = Peer {
|
||||
id: 2,
|
||||
addr: "node1".to_string(),
|
||||
};
|
||||
let peer2 = Peer {
|
||||
id: 5,
|
||||
addr: "node2".to_string(),
|
||||
};
|
||||
let peer3 = Peer {
|
||||
id: 8,
|
||||
addr: "node3".to_string(),
|
||||
};
|
||||
put_datanodes(
|
||||
&ctx.meta_peer_client,
|
||||
vec![peer1.clone(), peer2.clone(), peer3.clone()],
|
||||
)
|
||||
.await;
|
||||
|
||||
let peers = selector
|
||||
.select(
|
||||
&ctx,
|
||||
SelectorOptions {
|
||||
min_required_items: 1,
|
||||
allow_duplication: true,
|
||||
exclude_peer_ids: HashSet::from([2]),
|
||||
},
|
||||
)
|
||||
.await
|
||||
.unwrap();
|
||||
assert_eq!(peers.len(), 1);
|
||||
assert_eq!(peers, vec![peer3.clone()]);
|
||||
}
|
||||
}
|
||||
|
||||
@@ -278,7 +278,7 @@ impl KvBackend for LeaderCachedKvBackend {
|
||||
|
||||
let remote_res = self.store.batch_get(remote_req).await?;
|
||||
let put_req = BatchPutRequest {
|
||||
kvs: remote_res.kvs.clone().into_iter().collect(),
|
||||
kvs: remote_res.kvs.clone().into_iter().map(Into::into).collect(),
|
||||
..Default::default()
|
||||
};
|
||||
let _ = self.cache.batch_put(put_req).await?;
|
||||
|
||||
@@ -18,13 +18,11 @@ common-error.workspace = true
|
||||
common-macro.workspace = true
|
||||
common-query.workspace = true
|
||||
common-recordbatch.workspace = true
|
||||
common-runtime.workspace = true
|
||||
common-telemetry.workspace = true
|
||||
common-time.workspace = true
|
||||
datafusion.workspace = true
|
||||
datatypes.workspace = true
|
||||
futures-util.workspace = true
|
||||
humantime-serde.workspace = true
|
||||
itertools.workspace = true
|
||||
lazy_static = "1.4"
|
||||
mito2.workspace = true
|
||||
|
||||
@@ -12,49 +12,9 @@
|
||||
// See the License for the specific language governing permissions and
|
||||
// limitations under the License.
|
||||
|
||||
use std::time::Duration;
|
||||
|
||||
use common_telemetry::warn;
|
||||
use serde::{Deserialize, Serialize};
|
||||
|
||||
/// The default flush interval of the metadata region.
|
||||
pub(crate) const DEFAULT_FLUSH_METADATA_REGION_INTERVAL: Duration = Duration::from_secs(30);
|
||||
|
||||
/// Configuration for the metric engine.
|
||||
#[derive(Debug, Clone, Serialize, Deserialize, PartialEq, Eq)]
|
||||
#[derive(Debug, Clone, Default, Serialize, Deserialize, PartialEq, Eq)]
|
||||
pub struct EngineConfig {
|
||||
/// Experimental feature to use sparse primary key encoding.
|
||||
pub experimental_sparse_primary_key_encoding: bool,
|
||||
/// The flush interval of the metadata region.
|
||||
#[serde(
|
||||
with = "humantime_serde",
|
||||
default = "EngineConfig::default_flush_metadata_region_interval"
|
||||
)]
|
||||
pub flush_metadata_region_interval: Duration,
|
||||
}
|
||||
|
||||
impl Default for EngineConfig {
|
||||
fn default() -> Self {
|
||||
Self {
|
||||
flush_metadata_region_interval: DEFAULT_FLUSH_METADATA_REGION_INTERVAL,
|
||||
experimental_sparse_primary_key_encoding: false,
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
impl EngineConfig {
|
||||
fn default_flush_metadata_region_interval() -> Duration {
|
||||
DEFAULT_FLUSH_METADATA_REGION_INTERVAL
|
||||
}
|
||||
|
||||
/// Sanitizes the configuration.
|
||||
pub fn sanitize(&mut self) {
|
||||
if self.flush_metadata_region_interval.is_zero() {
|
||||
warn!(
|
||||
"Flush metadata region interval is zero, override with default value: {:?}. Disable metadata region flush is forbidden.",
|
||||
DEFAULT_FLUSH_METADATA_REGION_INTERVAL
|
||||
);
|
||||
self.flush_metadata_region_interval = DEFAULT_FLUSH_METADATA_REGION_INTERVAL;
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
@@ -34,11 +34,9 @@ use api::region::RegionResponse;
|
||||
use async_trait::async_trait;
|
||||
use common_error::ext::{BoxedError, ErrorExt};
|
||||
use common_error::status_code::StatusCode;
|
||||
use common_runtime::RepeatedTask;
|
||||
use mito2::engine::MitoEngine;
|
||||
pub(crate) use options::IndexOptions;
|
||||
use snafu::ResultExt;
|
||||
pub(crate) use state::MetricEngineState;
|
||||
use store_api::metadata::RegionMetadataRef;
|
||||
use store_api::metric_engine_consts::METRIC_ENGINE_NAME;
|
||||
use store_api::region_engine::{
|
||||
@@ -49,11 +47,11 @@ use store_api::region_engine::{
|
||||
use store_api::region_request::{BatchRegionDdlRequest, RegionRequest};
|
||||
use store_api::storage::{RegionId, ScanRequest, SequenceNumber};
|
||||
|
||||
use self::state::MetricEngineState;
|
||||
use crate::config::EngineConfig;
|
||||
use crate::data_region::DataRegion;
|
||||
use crate::error::{self, Error, Result, StartRepeatedTaskSnafu, UnsupportedRegionRequestSnafu};
|
||||
use crate::error::{self, Result, UnsupportedRegionRequestSnafu};
|
||||
use crate::metadata_region::MetadataRegion;
|
||||
use crate::repeated_task::FlushMetadataRegionTask;
|
||||
use crate::row_modifier::RowModifier;
|
||||
use crate::utils::{self, get_region_statistic};
|
||||
|
||||
@@ -361,32 +359,19 @@ impl RegionEngine for MetricEngine {
|
||||
}
|
||||
|
||||
impl MetricEngine {
|
||||
pub fn try_new(mito: MitoEngine, mut config: EngineConfig) -> Result<Self> {
|
||||
pub fn new(mito: MitoEngine, config: EngineConfig) -> Self {
|
||||
let metadata_region = MetadataRegion::new(mito.clone());
|
||||
let data_region = DataRegion::new(mito.clone());
|
||||
let state = Arc::new(RwLock::default());
|
||||
config.sanitize();
|
||||
let flush_interval = config.flush_metadata_region_interval;
|
||||
let inner = Arc::new(MetricEngineInner {
|
||||
mito: mito.clone(),
|
||||
metadata_region,
|
||||
data_region,
|
||||
state: state.clone(),
|
||||
config,
|
||||
row_modifier: RowModifier::new(),
|
||||
flush_task: RepeatedTask::new(
|
||||
flush_interval,
|
||||
Box::new(FlushMetadataRegionTask {
|
||||
state: state.clone(),
|
||||
mito: mito.clone(),
|
||||
}),
|
||||
),
|
||||
});
|
||||
inner
|
||||
.flush_task
|
||||
.start(common_runtime::global_runtime())
|
||||
.context(StartRepeatedTaskSnafu { name: "flush_task" })?;
|
||||
Ok(Self { inner })
|
||||
Self {
|
||||
inner: Arc::new(MetricEngineInner {
|
||||
mito,
|
||||
metadata_region,
|
||||
data_region,
|
||||
state: RwLock::default(),
|
||||
config,
|
||||
row_modifier: RowModifier::new(),
|
||||
}),
|
||||
}
|
||||
}
|
||||
|
||||
pub fn mito(&self) -> MitoEngine {
|
||||
@@ -441,21 +426,15 @@ impl MetricEngine {
|
||||
) -> Result<common_recordbatch::SendableRecordBatchStream, BoxedError> {
|
||||
self.inner.scan_to_stream(region_id, request).await
|
||||
}
|
||||
|
||||
/// Returns the configuration of the engine.
|
||||
pub fn config(&self) -> &EngineConfig {
|
||||
&self.inner.config
|
||||
}
|
||||
}
|
||||
|
||||
struct MetricEngineInner {
|
||||
mito: MitoEngine,
|
||||
metadata_region: MetadataRegion,
|
||||
data_region: DataRegion,
|
||||
state: Arc<RwLock<MetricEngineState>>,
|
||||
state: RwLock<MetricEngineState>,
|
||||
config: EngineConfig,
|
||||
row_modifier: RowModifier,
|
||||
flush_task: RepeatedTask<Error>,
|
||||
}
|
||||
|
||||
#[cfg(test)]
|
||||
|
||||
@@ -737,7 +737,7 @@ mod test {
|
||||
|
||||
// set up
|
||||
let env = TestEnv::new().await;
|
||||
let engine = MetricEngine::try_new(env.mito(), EngineConfig::default()).unwrap();
|
||||
let engine = MetricEngine::new(env.mito(), EngineConfig::default());
|
||||
let engine_inner = engine.inner;
|
||||
|
||||
// check create data region request
|
||||
|
||||
@@ -282,14 +282,6 @@ pub enum Error {
|
||||
#[snafu(implicit)]
|
||||
location: Location,
|
||||
},
|
||||
|
||||
#[snafu(display("Failed to start repeated task: {}", name))]
|
||||
StartRepeatedTask {
|
||||
name: String,
|
||||
source: common_runtime::error::Error,
|
||||
#[snafu(implicit)]
|
||||
location: Location,
|
||||
},
|
||||
}
|
||||
|
||||
pub type Result<T, E = Error> = std::result::Result<T, E>;
|
||||
@@ -343,8 +335,6 @@ impl ErrorExt for Error {
|
||||
|
||||
CollectRecordBatchStream { source, .. } => source.status_code(),
|
||||
|
||||
StartRepeatedTask { source, .. } => source.status_code(),
|
||||
|
||||
MetricManifestInfo { .. } => StatusCode::Internal,
|
||||
}
|
||||
}
|
||||
|
||||
@@ -59,7 +59,6 @@ pub mod engine;
|
||||
pub mod error;
|
||||
mod metadata_region;
|
||||
mod metrics;
|
||||
mod repeated_task;
|
||||
pub mod row_modifier;
|
||||
#[cfg(test)]
|
||||
mod test_util;
|
||||
|
||||
@@ -1,167 +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::sync::{Arc, RwLock};
|
||||
use std::time::Instant;
|
||||
|
||||
use common_runtime::TaskFunction;
|
||||
use common_telemetry::{debug, error};
|
||||
use mito2::engine::MitoEngine;
|
||||
use store_api::region_engine::{RegionEngine, RegionRole};
|
||||
use store_api::region_request::{RegionFlushRequest, RegionRequest};
|
||||
|
||||
use crate::engine::MetricEngineState;
|
||||
use crate::error::{Error, Result};
|
||||
use crate::utils;
|
||||
|
||||
/// Task to flush metadata regions.
|
||||
///
|
||||
/// This task is used to send flush requests to the metadata regions
|
||||
/// periodically.
|
||||
pub(crate) struct FlushMetadataRegionTask {
|
||||
pub(crate) state: Arc<RwLock<MetricEngineState>>,
|
||||
pub(crate) mito: MitoEngine,
|
||||
}
|
||||
|
||||
#[async_trait::async_trait]
|
||||
impl TaskFunction<Error> for FlushMetadataRegionTask {
|
||||
fn name(&self) -> &str {
|
||||
"FlushMetadataRegionTask"
|
||||
}
|
||||
|
||||
async fn call(&mut self) -> Result<()> {
|
||||
let region_ids = {
|
||||
let state = self.state.read().unwrap();
|
||||
state
|
||||
.physical_region_states()
|
||||
.keys()
|
||||
.cloned()
|
||||
.collect::<Vec<_>>()
|
||||
};
|
||||
|
||||
let num_region = region_ids.len();
|
||||
let now = Instant::now();
|
||||
for region_id in region_ids {
|
||||
let Some(role) = self.mito.role(region_id) else {
|
||||
continue;
|
||||
};
|
||||
if role == RegionRole::Follower {
|
||||
continue;
|
||||
}
|
||||
let metadata_region_id = utils::to_metadata_region_id(region_id);
|
||||
if let Err(e) = self
|
||||
.mito
|
||||
.handle_request(
|
||||
metadata_region_id,
|
||||
RegionRequest::Flush(RegionFlushRequest {
|
||||
row_group_size: None,
|
||||
}),
|
||||
)
|
||||
.await
|
||||
{
|
||||
error!(e; "Failed to flush metadata region {}", metadata_region_id);
|
||||
}
|
||||
}
|
||||
debug!(
|
||||
"Flushed {} metadata regions, elapsed: {:?}",
|
||||
num_region,
|
||||
now.elapsed()
|
||||
);
|
||||
|
||||
Ok(())
|
||||
}
|
||||
}
|
||||
|
||||
#[cfg(test)]
|
||||
mod tests {
|
||||
use std::assert_matches::assert_matches;
|
||||
use std::time::Duration;
|
||||
|
||||
use store_api::region_engine::{RegionEngine, RegionManifestInfo};
|
||||
|
||||
use crate::config::{EngineConfig, DEFAULT_FLUSH_METADATA_REGION_INTERVAL};
|
||||
use crate::test_util::TestEnv;
|
||||
|
||||
#[tokio::test]
|
||||
async fn test_flush_metadata_region_task() {
|
||||
let env = TestEnv::with_prefix_and_config(
|
||||
"test_flush_metadata_region_task",
|
||||
EngineConfig {
|
||||
flush_metadata_region_interval: Duration::from_millis(100),
|
||||
..Default::default()
|
||||
},
|
||||
)
|
||||
.await;
|
||||
env.init_metric_region().await;
|
||||
let engine = env.metric();
|
||||
// Wait for flush task run
|
||||
tokio::time::sleep(Duration::from_millis(200)).await;
|
||||
let physical_region_id = env.default_physical_region_id();
|
||||
let stat = engine.region_statistic(physical_region_id).unwrap();
|
||||
|
||||
assert_matches!(
|
||||
stat.manifest,
|
||||
RegionManifestInfo::Metric {
|
||||
metadata_manifest_version: 1,
|
||||
metadata_flushed_entry_id: 1,
|
||||
..
|
||||
}
|
||||
)
|
||||
}
|
||||
|
||||
#[tokio::test]
|
||||
async fn test_flush_metadata_region_task_with_long_interval() {
|
||||
let env = TestEnv::with_prefix_and_config(
|
||||
"test_flush_metadata_region_task_with_long_interval",
|
||||
EngineConfig {
|
||||
flush_metadata_region_interval: Duration::from_secs(60),
|
||||
..Default::default()
|
||||
},
|
||||
)
|
||||
.await;
|
||||
env.init_metric_region().await;
|
||||
let engine = env.metric();
|
||||
// Wait for flush task run, should not flush metadata region
|
||||
tokio::time::sleep(Duration::from_millis(200)).await;
|
||||
let physical_region_id = env.default_physical_region_id();
|
||||
let stat = engine.region_statistic(physical_region_id).unwrap();
|
||||
|
||||
assert_matches!(
|
||||
stat.manifest,
|
||||
RegionManifestInfo::Metric {
|
||||
metadata_manifest_version: 0,
|
||||
metadata_flushed_entry_id: 0,
|
||||
..
|
||||
}
|
||||
)
|
||||
}
|
||||
|
||||
#[tokio::test]
|
||||
async fn test_flush_metadata_region_sanitize() {
|
||||
let env = TestEnv::with_prefix_and_config(
|
||||
"test_flush_metadata_region_sanitize",
|
||||
EngineConfig {
|
||||
flush_metadata_region_interval: Duration::from_secs(0),
|
||||
..Default::default()
|
||||
},
|
||||
)
|
||||
.await;
|
||||
let metric = env.metric();
|
||||
let config = metric.config();
|
||||
assert_eq!(
|
||||
config.flush_metadata_region_interval,
|
||||
DEFAULT_FLUSH_METADATA_REGION_INTERVAL
|
||||
);
|
||||
}
|
||||
}
|
||||
@@ -54,14 +54,9 @@ impl TestEnv {
|
||||
|
||||
/// Returns a new env with specific `prefix` for test.
|
||||
pub async fn with_prefix(prefix: &str) -> Self {
|
||||
Self::with_prefix_and_config(prefix, EngineConfig::default()).await
|
||||
}
|
||||
|
||||
/// Returns a new env with specific `prefix` and `config` for test.
|
||||
pub async fn with_prefix_and_config(prefix: &str, config: EngineConfig) -> Self {
|
||||
let mut mito_env = MitoTestEnv::with_prefix(prefix);
|
||||
let mito = mito_env.create_engine(MitoConfig::default()).await;
|
||||
let metric = MetricEngine::try_new(mito.clone(), config).unwrap();
|
||||
let metric = MetricEngine::new(mito.clone(), EngineConfig::default());
|
||||
Self {
|
||||
mito_env,
|
||||
mito,
|
||||
@@ -89,7 +84,7 @@ impl TestEnv {
|
||||
.mito_env
|
||||
.create_follower_engine(MitoConfig::default())
|
||||
.await;
|
||||
let metric = MetricEngine::try_new(mito.clone(), EngineConfig::default()).unwrap();
|
||||
let metric = MetricEngine::new(mito.clone(), EngineConfig::default());
|
||||
|
||||
let region_id = self.default_physical_region_id();
|
||||
debug!("opening default physical region: {region_id}");
|
||||
|
||||
@@ -710,8 +710,8 @@ pub enum Error {
|
||||
error: std::io::Error,
|
||||
},
|
||||
|
||||
#[snafu(display("Record batch error"))]
|
||||
RecordBatch {
|
||||
#[snafu(display("Failed to filter record batch"))]
|
||||
FilterRecordBatch {
|
||||
source: common_recordbatch::error::Error,
|
||||
#[snafu(implicit)]
|
||||
location: Location,
|
||||
@@ -1032,20 +1032,6 @@ pub enum Error {
|
||||
#[snafu(implicit)]
|
||||
location: Location,
|
||||
},
|
||||
|
||||
#[snafu(display("Failed to scan series"))]
|
||||
ScanSeries {
|
||||
#[snafu(implicit)]
|
||||
location: Location,
|
||||
source: Arc<Error>,
|
||||
},
|
||||
|
||||
#[snafu(display("Partition {} scan multiple times", partition))]
|
||||
ScanMultiTimes {
|
||||
partition: usize,
|
||||
#[snafu(implicit)]
|
||||
location: Location,
|
||||
},
|
||||
}
|
||||
|
||||
pub type Result<T, E = Error> = std::result::Result<T, E>;
|
||||
@@ -1168,7 +1154,7 @@ impl ErrorExt for Error {
|
||||
|
||||
External { source, .. } => source.status_code(),
|
||||
|
||||
RecordBatch { source, .. } => source.status_code(),
|
||||
FilterRecordBatch { source, .. } => source.status_code(),
|
||||
|
||||
Download { .. } | Upload { .. } => StatusCode::StorageUnavailable,
|
||||
ChecksumMismatch { .. } => StatusCode::Unexpected,
|
||||
@@ -1197,12 +1183,7 @@ impl ErrorExt for Error {
|
||||
ManualCompactionOverride {} => StatusCode::Cancelled,
|
||||
|
||||
IncompatibleWalProviderChange { .. } => StatusCode::InvalidArguments,
|
||||
|
||||
ConvertDataType { .. } => StatusCode::Internal,
|
||||
|
||||
ScanSeries { source, .. } => source.status_code(),
|
||||
|
||||
ScanMultiTimes { .. } => StatusCode::InvalidArguments,
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
@@ -24,7 +24,6 @@ pub(crate) mod range;
|
||||
pub(crate) mod scan_region;
|
||||
pub(crate) mod scan_util;
|
||||
pub(crate) mod seq_scan;
|
||||
pub(crate) mod series_scan;
|
||||
pub(crate) mod unordered_scan;
|
||||
|
||||
use std::collections::{HashMap, HashSet};
|
||||
|
||||
@@ -363,9 +363,9 @@ mod tests {
|
||||
builder
|
||||
.push_field_array(
|
||||
*column_id,
|
||||
Arc::new(Int64Array::from_iter_values(std::iter::repeat_n(
|
||||
*field, num_rows,
|
||||
))),
|
||||
Arc::new(Int64Array::from_iter_values(
|
||||
std::iter::repeat(*field).take(num_rows),
|
||||
)),
|
||||
)
|
||||
.unwrap();
|
||||
}
|
||||
|
||||
@@ -21,7 +21,7 @@ use datatypes::arrow::array::BooleanArray;
|
||||
use datatypes::arrow::buffer::BooleanBuffer;
|
||||
use snafu::ResultExt;
|
||||
|
||||
use crate::error::{RecordBatchSnafu, Result};
|
||||
use crate::error::{FilterRecordBatchSnafu, Result};
|
||||
use crate::memtable::BoxedBatchIterator;
|
||||
use crate::read::last_row::RowGroupLastRowCachedReader;
|
||||
use crate::read::{Batch, BatchReader};
|
||||
@@ -201,7 +201,7 @@ impl PruneTimeIterator {
|
||||
for filter in filters.iter() {
|
||||
let result = filter
|
||||
.evaluate_vector(batch.timestamps())
|
||||
.context(RecordBatchSnafu)?;
|
||||
.context(FilterRecordBatchSnafu)?;
|
||||
mask = mask.bitand(&result);
|
||||
}
|
||||
|
||||
|
||||
@@ -46,7 +46,6 @@ use crate::read::compat::{self, CompatBatch};
|
||||
use crate::read::projection::ProjectionMapper;
|
||||
use crate::read::range::{FileRangeBuilder, MemRangeBuilder, RangeMeta, RowGroupIndex};
|
||||
use crate::read::seq_scan::SeqScan;
|
||||
use crate::read::series_scan::SeriesScan;
|
||||
use crate::read::unordered_scan::UnorderedScan;
|
||||
use crate::read::{Batch, Source};
|
||||
use crate::region::options::MergeMode;
|
||||
@@ -67,8 +66,6 @@ pub(crate) enum Scanner {
|
||||
Seq(SeqScan),
|
||||
/// Unordered scan.
|
||||
Unordered(UnorderedScan),
|
||||
/// Per-series scan.
|
||||
Series(SeriesScan),
|
||||
}
|
||||
|
||||
impl Scanner {
|
||||
@@ -78,7 +75,6 @@ impl Scanner {
|
||||
match self {
|
||||
Scanner::Seq(seq_scan) => seq_scan.build_stream(),
|
||||
Scanner::Unordered(unordered_scan) => unordered_scan.build_stream().await,
|
||||
Scanner::Series(series_scan) => series_scan.build_stream().await,
|
||||
}
|
||||
}
|
||||
}
|
||||
@@ -90,7 +86,6 @@ impl Scanner {
|
||||
match self {
|
||||
Scanner::Seq(seq_scan) => seq_scan.input().num_files(),
|
||||
Scanner::Unordered(unordered_scan) => unordered_scan.input().num_files(),
|
||||
Scanner::Series(series_scan) => series_scan.input().num_files(),
|
||||
}
|
||||
}
|
||||
|
||||
@@ -99,7 +94,6 @@ impl Scanner {
|
||||
match self {
|
||||
Scanner::Seq(seq_scan) => seq_scan.input().num_memtables(),
|
||||
Scanner::Unordered(unordered_scan) => unordered_scan.input().num_memtables(),
|
||||
Scanner::Series(series_scan) => series_scan.input().num_memtables(),
|
||||
}
|
||||
}
|
||||
|
||||
@@ -108,7 +102,6 @@ impl Scanner {
|
||||
match self {
|
||||
Scanner::Seq(seq_scan) => seq_scan.input().file_ids(),
|
||||
Scanner::Unordered(unordered_scan) => unordered_scan.input().file_ids(),
|
||||
Scanner::Series(series_scan) => series_scan.input().file_ids(),
|
||||
}
|
||||
}
|
||||
|
||||
@@ -120,7 +113,6 @@ impl Scanner {
|
||||
match self {
|
||||
Scanner::Seq(seq_scan) => seq_scan.prepare(request).unwrap(),
|
||||
Scanner::Unordered(unordered_scan) => unordered_scan.prepare(request).unwrap(),
|
||||
Scanner::Series(series_scan) => series_scan.prepare(request).unwrap(),
|
||||
}
|
||||
}
|
||||
}
|
||||
@@ -256,9 +248,7 @@ impl ScanRegion {
|
||||
|
||||
/// Returns a [Scanner] to scan the region.
|
||||
pub(crate) fn scanner(self) -> Result<Scanner> {
|
||||
if self.use_series_scan() {
|
||||
self.series_scan().map(Scanner::Series)
|
||||
} else if self.use_unordered_scan() {
|
||||
if self.use_unordered_scan() {
|
||||
// If table is append only and there is no series row selector, we use unordered scan in query.
|
||||
// We still use seq scan in compaction.
|
||||
self.unordered_scan().map(Scanner::Unordered)
|
||||
@@ -270,9 +260,7 @@ impl ScanRegion {
|
||||
/// Returns a [RegionScanner] to scan the region.
|
||||
#[tracing::instrument(level = tracing::Level::DEBUG, skip_all)]
|
||||
pub(crate) fn region_scanner(self) -> Result<RegionScannerRef> {
|
||||
if self.use_series_scan() {
|
||||
self.series_scan().map(|scanner| Box::new(scanner) as _)
|
||||
} else if self.use_unordered_scan() {
|
||||
if self.use_unordered_scan() {
|
||||
self.unordered_scan().map(|scanner| Box::new(scanner) as _)
|
||||
} else {
|
||||
self.seq_scan().map(|scanner| Box::new(scanner) as _)
|
||||
@@ -291,12 +279,6 @@ impl ScanRegion {
|
||||
Ok(UnorderedScan::new(input))
|
||||
}
|
||||
|
||||
/// Scans by series.
|
||||
pub(crate) fn series_scan(self) -> Result<SeriesScan> {
|
||||
let input = self.scan_input(true)?;
|
||||
Ok(SeriesScan::new(input))
|
||||
}
|
||||
|
||||
#[cfg(test)]
|
||||
pub(crate) fn scan_without_filter_deleted(self) -> Result<SeqScan> {
|
||||
let input = self.scan_input(false)?;
|
||||
@@ -317,11 +299,6 @@ impl ScanRegion {
|
||||
|| self.request.distribution == Some(TimeSeriesDistribution::TimeWindowed))
|
||||
}
|
||||
|
||||
/// Returns true if the region can use series scan for current request.
|
||||
fn use_series_scan(&self) -> bool {
|
||||
self.request.distribution == Some(TimeSeriesDistribution::PerSeries)
|
||||
}
|
||||
|
||||
/// Creates a scan input.
|
||||
fn scan_input(mut self, filter_deleted: bool) -> Result<ScanInput> {
|
||||
let time_range = self.build_time_range_predicate();
|
||||
|
||||
@@ -92,8 +92,6 @@ struct ScanMetricsSet {
|
||||
|
||||
/// Elapsed time before the first poll operation.
|
||||
first_poll: Duration,
|
||||
/// Number of send timeout in SeriesScan.
|
||||
num_series_send_timeout: usize,
|
||||
}
|
||||
|
||||
impl fmt::Debug for ScanMetricsSet {
|
||||
@@ -124,7 +122,6 @@ impl fmt::Debug for ScanMetricsSet {
|
||||
num_sst_batches,
|
||||
num_sst_rows,
|
||||
first_poll,
|
||||
num_series_send_timeout,
|
||||
} = self;
|
||||
|
||||
write!(
|
||||
@@ -153,8 +150,7 @@ impl fmt::Debug for ScanMetricsSet {
|
||||
num_sst_record_batches={num_sst_record_batches}, \
|
||||
num_sst_batches={num_sst_batches}, \
|
||||
num_sst_rows={num_sst_rows}, \
|
||||
first_poll={first_poll:?}, \
|
||||
num_series_send_timeout={num_series_send_timeout}}}"
|
||||
first_poll={first_poll:?}}}"
|
||||
)
|
||||
}
|
||||
}
|
||||
@@ -443,12 +439,6 @@ impl PartitionMetrics {
|
||||
pub(crate) fn on_finish(&self) {
|
||||
self.0.on_finish();
|
||||
}
|
||||
|
||||
/// Sets the `num_series_send_timeout`.
|
||||
pub(crate) fn set_num_series_send_timeout(&self, num_timeout: usize) {
|
||||
let mut metrics = self.0.metrics.lock().unwrap();
|
||||
metrics.num_series_send_timeout = num_timeout;
|
||||
}
|
||||
}
|
||||
|
||||
impl fmt::Debug for PartitionMetrics {
|
||||
|
||||
@@ -30,7 +30,7 @@ use datatypes::schema::SchemaRef;
|
||||
use snafu::ResultExt;
|
||||
use store_api::metadata::RegionMetadataRef;
|
||||
use store_api::region_engine::{PartitionRange, PrepareRequest, RegionScanner, ScannerProperties};
|
||||
use store_api::storage::TimeSeriesRowSelector;
|
||||
use store_api::storage::{TimeSeriesDistribution, TimeSeriesRowSelector};
|
||||
use tokio::sync::Semaphore;
|
||||
|
||||
use crate::error::{PartitionOutOfRangeSnafu, Result};
|
||||
@@ -149,7 +149,7 @@ impl SeqScan {
|
||||
/// Builds a reader to read sources. If `semaphore` is provided, reads sources in parallel
|
||||
/// if possible.
|
||||
#[tracing::instrument(level = tracing::Level::DEBUG, skip_all)]
|
||||
pub(crate) async fn build_reader_from_sources(
|
||||
async fn build_reader_from_sources(
|
||||
stream_ctx: &StreamContext,
|
||||
mut sources: Vec<Source>,
|
||||
semaphore: Option<Arc<Semaphore>>,
|
||||
@@ -206,13 +206,9 @@ impl SeqScan {
|
||||
.build(),
|
||||
));
|
||||
}
|
||||
if self.properties.partitions[partition].is_empty() {
|
||||
return Ok(Box::pin(RecordBatchStreamWrapper::new(
|
||||
self.stream_ctx.input.mapper.output_schema(),
|
||||
common_recordbatch::EmptyRecordBatchStream::new(
|
||||
self.stream_ctx.input.mapper.output_schema(),
|
||||
),
|
||||
)));
|
||||
|
||||
if self.stream_ctx.input.distribution == Some(TimeSeriesDistribution::PerSeries) {
|
||||
return self.scan_partition_by_series(metrics_set, partition);
|
||||
}
|
||||
|
||||
let stream_ctx = self.stream_ctx.clone();
|
||||
@@ -241,14 +237,14 @@ impl SeqScan {
|
||||
&mut sources,
|
||||
);
|
||||
|
||||
let mut metrics = ScannerMetrics::default();
|
||||
let mut fetch_start = Instant::now();
|
||||
let mut reader =
|
||||
Self::build_reader_from_sources(&stream_ctx, sources, semaphore.clone())
|
||||
.await
|
||||
.map_err(BoxedError::new)
|
||||
.context(ExternalSnafu)?;
|
||||
let cache = &stream_ctx.input.cache_strategy;
|
||||
let mut metrics = ScannerMetrics::default();
|
||||
let mut fetch_start = Instant::now();
|
||||
#[cfg(debug_assertions)]
|
||||
let mut checker = crate::read::BatchChecker::default()
|
||||
.with_start(Some(part_range.start))
|
||||
@@ -311,6 +307,97 @@ impl SeqScan {
|
||||
Ok(stream)
|
||||
}
|
||||
|
||||
/// Scans all ranges in the given partition and merge by time series.
|
||||
/// Otherwise the returned stream might not contains any data.
|
||||
fn scan_partition_by_series(
|
||||
&self,
|
||||
metrics_set: &ExecutionPlanMetricsSet,
|
||||
partition: usize,
|
||||
) -> Result<SendableRecordBatchStream, BoxedError> {
|
||||
let stream_ctx = self.stream_ctx.clone();
|
||||
let semaphore = self.new_semaphore();
|
||||
let partition_ranges = self.properties.partitions[partition].clone();
|
||||
let distinguish_range = self.properties.distinguish_partition_range;
|
||||
let part_metrics = self.new_partition_metrics(metrics_set, partition);
|
||||
debug_assert!(!self.compaction);
|
||||
|
||||
let stream = try_stream! {
|
||||
part_metrics.on_first_poll();
|
||||
|
||||
let range_builder_list = Arc::new(RangeBuilderList::new(
|
||||
stream_ctx.input.num_memtables(),
|
||||
stream_ctx.input.num_files(),
|
||||
));
|
||||
// Scans all parts.
|
||||
let mut sources = Vec::with_capacity(partition_ranges.len());
|
||||
for part_range in partition_ranges {
|
||||
build_sources(
|
||||
&stream_ctx,
|
||||
&part_range,
|
||||
false,
|
||||
&part_metrics,
|
||||
range_builder_list.clone(),
|
||||
&mut sources,
|
||||
);
|
||||
}
|
||||
|
||||
// Builds a reader that merge sources from all parts.
|
||||
let mut reader =
|
||||
Self::build_reader_from_sources(&stream_ctx, sources, semaphore.clone())
|
||||
.await
|
||||
.map_err(BoxedError::new)
|
||||
.context(ExternalSnafu)?;
|
||||
let cache = &stream_ctx.input.cache_strategy;
|
||||
let mut metrics = ScannerMetrics::default();
|
||||
let mut fetch_start = Instant::now();
|
||||
|
||||
while let Some(batch) = reader
|
||||
.next_batch()
|
||||
.await
|
||||
.map_err(BoxedError::new)
|
||||
.context(ExternalSnafu)?
|
||||
{
|
||||
metrics.scan_cost += fetch_start.elapsed();
|
||||
metrics.num_batches += 1;
|
||||
metrics.num_rows += batch.num_rows();
|
||||
|
||||
debug_assert!(!batch.is_empty());
|
||||
if batch.is_empty() {
|
||||
continue;
|
||||
}
|
||||
|
||||
let convert_start = Instant::now();
|
||||
let record_batch = stream_ctx.input.mapper.convert(&batch, cache)?;
|
||||
metrics.convert_cost += convert_start.elapsed();
|
||||
let yield_start = Instant::now();
|
||||
yield record_batch;
|
||||
metrics.yield_cost += yield_start.elapsed();
|
||||
|
||||
fetch_start = Instant::now();
|
||||
}
|
||||
|
||||
// Yields an empty part to indicate this range is terminated.
|
||||
// The query engine can use this to optimize some queries.
|
||||
if distinguish_range {
|
||||
let yield_start = Instant::now();
|
||||
yield stream_ctx.input.mapper.empty_record_batch();
|
||||
metrics.yield_cost += yield_start.elapsed();
|
||||
}
|
||||
|
||||
metrics.scan_cost += fetch_start.elapsed();
|
||||
part_metrics.merge_metrics(&metrics);
|
||||
|
||||
part_metrics.on_finish();
|
||||
};
|
||||
|
||||
let stream = Box::pin(RecordBatchStreamWrapper::new(
|
||||
self.stream_ctx.input.mapper.output_schema(),
|
||||
Box::pin(stream),
|
||||
));
|
||||
|
||||
Ok(stream)
|
||||
}
|
||||
|
||||
fn new_semaphore(&self) -> Option<Arc<Semaphore>> {
|
||||
if self.properties.target_partitions() > self.properties.num_partitions() {
|
||||
// We can use additional tasks to read the data if we have more target partitions than actual partitions.
|
||||
@@ -411,7 +498,7 @@ impl fmt::Debug for SeqScan {
|
||||
}
|
||||
|
||||
/// Builds sources for the partition range and push them to the `sources` vector.
|
||||
pub(crate) fn build_sources(
|
||||
fn build_sources(
|
||||
stream_ctx: &Arc<StreamContext>,
|
||||
part_range: &PartitionRange,
|
||||
compaction: bool,
|
||||
@@ -422,8 +509,8 @@ pub(crate) fn build_sources(
|
||||
// Gets range meta.
|
||||
let range_meta = &stream_ctx.ranges[part_range.identifier];
|
||||
#[cfg(debug_assertions)]
|
||||
if compaction {
|
||||
// Compaction expects input sources are not been split.
|
||||
if compaction || stream_ctx.input.distribution == Some(TimeSeriesDistribution::PerSeries) {
|
||||
// Compaction or per series distribution expects input sources are not been split.
|
||||
debug_assert_eq!(range_meta.indices.len(), range_meta.row_group_indices.len());
|
||||
for (i, row_group_idx) in range_meta.row_group_indices.iter().enumerate() {
|
||||
// It should scan all row groups.
|
||||
|
||||
@@ -1,547 +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.
|
||||
|
||||
//! Per-series scan implementation.
|
||||
|
||||
use std::fmt;
|
||||
use std::sync::{Arc, Mutex};
|
||||
use std::time::{Duration, Instant};
|
||||
|
||||
use async_stream::try_stream;
|
||||
use common_error::ext::BoxedError;
|
||||
use common_recordbatch::error::ExternalSnafu;
|
||||
use common_recordbatch::util::ChainedRecordBatchStream;
|
||||
use common_recordbatch::{RecordBatch, RecordBatchStreamWrapper, SendableRecordBatchStream};
|
||||
use datafusion::physical_plan::metrics::ExecutionPlanMetricsSet;
|
||||
use datafusion::physical_plan::{DisplayAs, DisplayFormatType};
|
||||
use datatypes::compute::concat_batches;
|
||||
use datatypes::schema::SchemaRef;
|
||||
use smallvec::{smallvec, SmallVec};
|
||||
use snafu::{ensure, OptionExt, ResultExt};
|
||||
use store_api::metadata::RegionMetadataRef;
|
||||
use store_api::region_engine::{PartitionRange, PrepareRequest, RegionScanner, ScannerProperties};
|
||||
use tokio::sync::mpsc::error::{SendTimeoutError, TrySendError};
|
||||
use tokio::sync::mpsc::{self, Receiver, Sender};
|
||||
use tokio::sync::Semaphore;
|
||||
|
||||
use crate::error::{
|
||||
ComputeArrowSnafu, Error, InvalidSenderSnafu, PartitionOutOfRangeSnafu, Result,
|
||||
ScanMultiTimesSnafu, ScanSeriesSnafu,
|
||||
};
|
||||
use crate::read::range::RangeBuilderList;
|
||||
use crate::read::scan_region::{ScanInput, StreamContext};
|
||||
use crate::read::scan_util::{PartitionMetrics, PartitionMetricsList};
|
||||
use crate::read::seq_scan::{build_sources, SeqScan};
|
||||
use crate::read::{Batch, ScannerMetrics};
|
||||
|
||||
/// Timeout to send a batch to a sender.
|
||||
const SEND_TIMEOUT: Duration = Duration::from_millis(10);
|
||||
|
||||
/// List of receivers.
|
||||
type ReceiverList = Vec<Option<Receiver<Result<SeriesBatch>>>>;
|
||||
|
||||
/// Scans a region and returns sorted rows of a series in the same partition.
|
||||
///
|
||||
/// The output order is always order by `(primary key, time index)` inside every
|
||||
/// partition.
|
||||
/// Always returns the same series (primary key) to the same partition.
|
||||
pub struct SeriesScan {
|
||||
/// Properties of the scanner.
|
||||
properties: ScannerProperties,
|
||||
/// Context of streams.
|
||||
stream_ctx: Arc<StreamContext>,
|
||||
/// Receivers of each partition.
|
||||
receivers: Mutex<ReceiverList>,
|
||||
/// Metrics for each partition.
|
||||
/// The scanner only sets in query and keeps it empty during compaction.
|
||||
metrics_list: Arc<PartitionMetricsList>,
|
||||
}
|
||||
|
||||
impl SeriesScan {
|
||||
/// Creates a new [SeriesScan].
|
||||
pub(crate) fn new(input: ScanInput) -> Self {
|
||||
let mut properties = ScannerProperties::default()
|
||||
.with_append_mode(input.append_mode)
|
||||
.with_total_rows(input.total_rows());
|
||||
let stream_ctx = Arc::new(StreamContext::seq_scan_ctx(input, false));
|
||||
properties.partitions = vec![stream_ctx.partition_ranges()];
|
||||
|
||||
Self {
|
||||
properties,
|
||||
stream_ctx,
|
||||
receivers: Mutex::new(Vec::new()),
|
||||
metrics_list: Arc::new(PartitionMetricsList::default()),
|
||||
}
|
||||
}
|
||||
|
||||
fn scan_partition_impl(
|
||||
&self,
|
||||
metrics_set: &ExecutionPlanMetricsSet,
|
||||
partition: usize,
|
||||
) -> Result<SendableRecordBatchStream, BoxedError> {
|
||||
if partition >= self.properties.num_partitions() {
|
||||
return Err(BoxedError::new(
|
||||
PartitionOutOfRangeSnafu {
|
||||
given: partition,
|
||||
all: self.properties.num_partitions(),
|
||||
}
|
||||
.build(),
|
||||
));
|
||||
}
|
||||
|
||||
self.maybe_start_distributor(metrics_set, &self.metrics_list);
|
||||
|
||||
let part_metrics =
|
||||
new_partition_metrics(&self.stream_ctx, metrics_set, partition, &self.metrics_list);
|
||||
let mut receiver = self.take_receiver(partition).map_err(BoxedError::new)?;
|
||||
let stream_ctx = self.stream_ctx.clone();
|
||||
|
||||
let stream = try_stream! {
|
||||
part_metrics.on_first_poll();
|
||||
|
||||
let cache = &stream_ctx.input.cache_strategy;
|
||||
let mut df_record_batches = Vec::new();
|
||||
let mut fetch_start = Instant::now();
|
||||
while let Some(result) = receiver.recv().await {
|
||||
let mut metrics = ScannerMetrics::default();
|
||||
let series = result.map_err(BoxedError::new).context(ExternalSnafu)?;
|
||||
metrics.scan_cost += fetch_start.elapsed();
|
||||
fetch_start = Instant::now();
|
||||
|
||||
let convert_start = Instant::now();
|
||||
df_record_batches.reserve(series.batches.len());
|
||||
for batch in series.batches {
|
||||
metrics.num_batches += 1;
|
||||
metrics.num_rows += batch.num_rows();
|
||||
|
||||
let record_batch = stream_ctx.input.mapper.convert(&batch, cache)?;
|
||||
df_record_batches.push(record_batch.into_df_record_batch());
|
||||
}
|
||||
|
||||
let output_schema = stream_ctx.input.mapper.output_schema();
|
||||
let df_record_batch =
|
||||
concat_batches(output_schema.arrow_schema(), &df_record_batches)
|
||||
.context(ComputeArrowSnafu)
|
||||
.map_err(BoxedError::new)
|
||||
.context(ExternalSnafu)?;
|
||||
df_record_batches.clear();
|
||||
let record_batch =
|
||||
RecordBatch::try_from_df_record_batch(output_schema, df_record_batch)?;
|
||||
metrics.convert_cost += convert_start.elapsed();
|
||||
|
||||
let yield_start = Instant::now();
|
||||
yield record_batch;
|
||||
metrics.yield_cost += yield_start.elapsed();
|
||||
|
||||
part_metrics.merge_metrics(&metrics);
|
||||
}
|
||||
};
|
||||
|
||||
let stream = Box::pin(RecordBatchStreamWrapper::new(
|
||||
self.stream_ctx.input.mapper.output_schema(),
|
||||
Box::pin(stream),
|
||||
));
|
||||
|
||||
Ok(stream)
|
||||
}
|
||||
|
||||
/// Takes the receiver for the partition.
|
||||
fn take_receiver(&self, partition: usize) -> Result<Receiver<Result<SeriesBatch>>> {
|
||||
let mut rx_list = self.receivers.lock().unwrap();
|
||||
rx_list[partition]
|
||||
.take()
|
||||
.context(ScanMultiTimesSnafu { partition })
|
||||
}
|
||||
|
||||
/// Starts the distributor if the receiver list is empty.
|
||||
fn maybe_start_distributor(
|
||||
&self,
|
||||
metrics_set: &ExecutionPlanMetricsSet,
|
||||
metrics_list: &Arc<PartitionMetricsList>,
|
||||
) {
|
||||
let mut rx_list = self.receivers.lock().unwrap();
|
||||
if !rx_list.is_empty() {
|
||||
return;
|
||||
}
|
||||
|
||||
let (senders, receivers) = new_channel_list(self.properties.num_partitions());
|
||||
let mut distributor = SeriesDistributor {
|
||||
stream_ctx: self.stream_ctx.clone(),
|
||||
semaphore: Some(Arc::new(Semaphore::new(self.properties.num_partitions()))),
|
||||
partitions: self.properties.partitions.clone(),
|
||||
senders,
|
||||
metrics_set: metrics_set.clone(),
|
||||
metrics_list: metrics_list.clone(),
|
||||
};
|
||||
common_runtime::spawn_global(async move {
|
||||
distributor.execute().await;
|
||||
});
|
||||
|
||||
*rx_list = receivers;
|
||||
}
|
||||
|
||||
/// Scans the region and returns a stream.
|
||||
pub(crate) async fn build_stream(&self) -> Result<SendableRecordBatchStream, BoxedError> {
|
||||
let part_num = self.properties.num_partitions();
|
||||
let metrics_set = ExecutionPlanMetricsSet::default();
|
||||
let streams = (0..part_num)
|
||||
.map(|i| self.scan_partition(&metrics_set, i))
|
||||
.collect::<Result<Vec<_>, BoxedError>>()?;
|
||||
let chained_stream = ChainedRecordBatchStream::new(streams).map_err(BoxedError::new)?;
|
||||
Ok(Box::pin(chained_stream))
|
||||
}
|
||||
}
|
||||
|
||||
fn new_channel_list(num_partitions: usize) -> (SenderList, ReceiverList) {
|
||||
let (senders, receivers): (Vec<_>, Vec<_>) = (0..num_partitions)
|
||||
.map(|_| {
|
||||
let (sender, receiver) = mpsc::channel(1);
|
||||
(Some(sender), Some(receiver))
|
||||
})
|
||||
.unzip();
|
||||
(SenderList::new(senders), receivers)
|
||||
}
|
||||
|
||||
impl RegionScanner for SeriesScan {
|
||||
fn properties(&self) -> &ScannerProperties {
|
||||
&self.properties
|
||||
}
|
||||
|
||||
fn schema(&self) -> SchemaRef {
|
||||
self.stream_ctx.input.mapper.output_schema()
|
||||
}
|
||||
|
||||
fn metadata(&self) -> RegionMetadataRef {
|
||||
self.stream_ctx.input.mapper.metadata().clone()
|
||||
}
|
||||
|
||||
fn scan_partition(
|
||||
&self,
|
||||
metrics_set: &ExecutionPlanMetricsSet,
|
||||
partition: usize,
|
||||
) -> Result<SendableRecordBatchStream, BoxedError> {
|
||||
self.scan_partition_impl(metrics_set, partition)
|
||||
}
|
||||
|
||||
fn prepare(&mut self, request: PrepareRequest) -> Result<(), BoxedError> {
|
||||
self.properties.prepare(request);
|
||||
Ok(())
|
||||
}
|
||||
|
||||
fn has_predicate(&self) -> bool {
|
||||
let predicate = self.stream_ctx.input.predicate();
|
||||
predicate.map(|p| !p.exprs().is_empty()).unwrap_or(false)
|
||||
}
|
||||
|
||||
fn set_logical_region(&mut self, logical_region: bool) {
|
||||
self.properties.set_logical_region(logical_region);
|
||||
}
|
||||
}
|
||||
|
||||
impl DisplayAs for SeriesScan {
|
||||
fn fmt_as(&self, t: DisplayFormatType, f: &mut fmt::Formatter) -> fmt::Result {
|
||||
write!(
|
||||
f,
|
||||
"SeriesScan: region={}, ",
|
||||
self.stream_ctx.input.mapper.metadata().region_id
|
||||
)?;
|
||||
match t {
|
||||
DisplayFormatType::Default => self.stream_ctx.format_for_explain(false, f),
|
||||
DisplayFormatType::Verbose => {
|
||||
self.stream_ctx.format_for_explain(true, f)?;
|
||||
self.metrics_list.format_verbose_metrics(f)
|
||||
}
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
impl fmt::Debug for SeriesScan {
|
||||
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
|
||||
f.debug_struct("SeriesScan")
|
||||
.field("num_ranges", &self.stream_ctx.ranges.len())
|
||||
.finish()
|
||||
}
|
||||
}
|
||||
|
||||
#[cfg(test)]
|
||||
impl SeriesScan {
|
||||
/// Returns the input.
|
||||
pub(crate) fn input(&self) -> &ScanInput {
|
||||
&self.stream_ctx.input
|
||||
}
|
||||
}
|
||||
|
||||
/// The distributor scans series and distributes them to different partitions.
|
||||
struct SeriesDistributor {
|
||||
/// Context for the scan stream.
|
||||
stream_ctx: Arc<StreamContext>,
|
||||
/// Optional semaphore for limiting the number of concurrent scans.
|
||||
semaphore: Option<Arc<Semaphore>>,
|
||||
/// Partition ranges to scan.
|
||||
partitions: Vec<Vec<PartitionRange>>,
|
||||
/// Senders of all partitions.
|
||||
senders: SenderList,
|
||||
/// Metrics set to report.
|
||||
/// The distributor report the metrics as an additional partition.
|
||||
/// This may double the scan cost of the [SeriesScan] metrics. We can
|
||||
/// get per-partition metrics in verbose mode to see the metrics of the
|
||||
/// distributor.
|
||||
metrics_set: ExecutionPlanMetricsSet,
|
||||
metrics_list: Arc<PartitionMetricsList>,
|
||||
}
|
||||
|
||||
impl SeriesDistributor {
|
||||
/// Executes the distributor.
|
||||
async fn execute(&mut self) {
|
||||
if let Err(e) = self.scan_partitions().await {
|
||||
self.senders.send_error(e).await;
|
||||
}
|
||||
}
|
||||
|
||||
/// Scans all parts.
|
||||
async fn scan_partitions(&mut self) -> Result<()> {
|
||||
let part_metrics = new_partition_metrics(
|
||||
&self.stream_ctx,
|
||||
&self.metrics_set,
|
||||
self.partitions.len(),
|
||||
&self.metrics_list,
|
||||
);
|
||||
part_metrics.on_first_poll();
|
||||
|
||||
let range_builder_list = Arc::new(RangeBuilderList::new(
|
||||
self.stream_ctx.input.num_memtables(),
|
||||
self.stream_ctx.input.num_files(),
|
||||
));
|
||||
// Scans all parts.
|
||||
let mut sources = Vec::with_capacity(self.partitions.len());
|
||||
for partition in &self.partitions {
|
||||
sources.reserve(partition.len());
|
||||
for part_range in partition {
|
||||
build_sources(
|
||||
&self.stream_ctx,
|
||||
part_range,
|
||||
false,
|
||||
&part_metrics,
|
||||
range_builder_list.clone(),
|
||||
&mut sources,
|
||||
);
|
||||
}
|
||||
}
|
||||
|
||||
// Builds a reader that merge sources from all parts.
|
||||
let mut reader =
|
||||
SeqScan::build_reader_from_sources(&self.stream_ctx, sources, self.semaphore.clone())
|
||||
.await?;
|
||||
let mut metrics = ScannerMetrics::default();
|
||||
let mut fetch_start = Instant::now();
|
||||
|
||||
let mut current_series = SeriesBatch::default();
|
||||
while let Some(batch) = reader.next_batch().await? {
|
||||
metrics.scan_cost += fetch_start.elapsed();
|
||||
fetch_start = Instant::now();
|
||||
metrics.num_batches += 1;
|
||||
metrics.num_rows += batch.num_rows();
|
||||
|
||||
debug_assert!(!batch.is_empty());
|
||||
if batch.is_empty() {
|
||||
continue;
|
||||
}
|
||||
|
||||
let Some(last_key) = current_series.current_key() else {
|
||||
current_series.push(batch);
|
||||
continue;
|
||||
};
|
||||
|
||||
if last_key == batch.primary_key() {
|
||||
current_series.push(batch);
|
||||
continue;
|
||||
}
|
||||
|
||||
// We find a new series, send the current one.
|
||||
let to_send = std::mem::replace(&mut current_series, SeriesBatch::single(batch));
|
||||
let yield_start = Instant::now();
|
||||
self.senders.send_batch(to_send).await?;
|
||||
metrics.yield_cost += yield_start.elapsed();
|
||||
}
|
||||
|
||||
if !current_series.is_empty() {
|
||||
let yield_start = Instant::now();
|
||||
self.senders.send_batch(current_series).await?;
|
||||
metrics.yield_cost += yield_start.elapsed();
|
||||
}
|
||||
|
||||
metrics.scan_cost += fetch_start.elapsed();
|
||||
part_metrics.merge_metrics(&metrics);
|
||||
part_metrics.set_num_series_send_timeout(self.senders.num_timeout);
|
||||
|
||||
part_metrics.on_finish();
|
||||
|
||||
Ok(())
|
||||
}
|
||||
}
|
||||
|
||||
/// Batches of the same series.
|
||||
#[derive(Default)]
|
||||
struct SeriesBatch {
|
||||
batches: SmallVec<[Batch; 4]>,
|
||||
}
|
||||
|
||||
impl SeriesBatch {
|
||||
/// Creates a new [SeriesBatch] from a single [Batch].
|
||||
fn single(batch: Batch) -> Self {
|
||||
Self {
|
||||
batches: smallvec![batch],
|
||||
}
|
||||
}
|
||||
|
||||
fn current_key(&self) -> Option<&[u8]> {
|
||||
self.batches.first().map(|batch| batch.primary_key())
|
||||
}
|
||||
|
||||
fn push(&mut self, batch: Batch) {
|
||||
self.batches.push(batch);
|
||||
}
|
||||
|
||||
/// Returns true if there is no batch.
|
||||
fn is_empty(&self) -> bool {
|
||||
self.batches.is_empty()
|
||||
}
|
||||
}
|
||||
|
||||
/// List of senders.
|
||||
struct SenderList {
|
||||
senders: Vec<Option<Sender<Result<SeriesBatch>>>>,
|
||||
/// Number of None senders.
|
||||
num_nones: usize,
|
||||
/// Index of the current partition to send.
|
||||
sender_idx: usize,
|
||||
/// Number of timeout.
|
||||
num_timeout: usize,
|
||||
}
|
||||
|
||||
impl SenderList {
|
||||
fn new(senders: Vec<Option<Sender<Result<SeriesBatch>>>>) -> Self {
|
||||
let num_nones = senders.iter().filter(|sender| sender.is_none()).count();
|
||||
Self {
|
||||
senders,
|
||||
num_nones,
|
||||
sender_idx: 0,
|
||||
num_timeout: 0,
|
||||
}
|
||||
}
|
||||
|
||||
/// Finds a partition and tries to send the batch to the partition.
|
||||
/// Returns None if it sends successfully.
|
||||
fn try_send_batch(&mut self, mut batch: SeriesBatch) -> Result<Option<SeriesBatch>> {
|
||||
for _ in 0..self.senders.len() {
|
||||
ensure!(self.num_nones < self.senders.len(), InvalidSenderSnafu);
|
||||
|
||||
let sender_idx = self.fetch_add_sender_idx();
|
||||
let Some(sender) = &self.senders[sender_idx] else {
|
||||
continue;
|
||||
};
|
||||
|
||||
match sender.try_send(Ok(batch)) {
|
||||
Ok(()) => return Ok(None),
|
||||
Err(TrySendError::Full(res)) => {
|
||||
// Safety: we send Ok.
|
||||
batch = res.unwrap();
|
||||
}
|
||||
Err(TrySendError::Closed(res)) => {
|
||||
self.senders[sender_idx] = None;
|
||||
self.num_nones += 1;
|
||||
// Safety: we send Ok.
|
||||
batch = res.unwrap();
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
Ok(Some(batch))
|
||||
}
|
||||
|
||||
/// Finds a partition and sends the batch to the partition.
|
||||
async fn send_batch(&mut self, mut batch: SeriesBatch) -> Result<()> {
|
||||
// Sends the batch without blocking first.
|
||||
match self.try_send_batch(batch)? {
|
||||
Some(b) => {
|
||||
// Unable to send batch to partition.
|
||||
batch = b;
|
||||
}
|
||||
None => {
|
||||
return Ok(());
|
||||
}
|
||||
}
|
||||
|
||||
loop {
|
||||
ensure!(self.num_nones < self.senders.len(), InvalidSenderSnafu);
|
||||
|
||||
let sender_idx = self.fetch_add_sender_idx();
|
||||
let Some(sender) = &self.senders[sender_idx] else {
|
||||
continue;
|
||||
};
|
||||
// Adds a timeout to avoid blocking indefinitely and sending
|
||||
// the batch in a round-robin fashion when some partitions
|
||||
// don't poll their inputs. This may happen if we have a
|
||||
// node like sort merging. But it is rare when we are using SeriesScan.
|
||||
match sender.send_timeout(Ok(batch), SEND_TIMEOUT).await {
|
||||
Ok(()) => break,
|
||||
Err(SendTimeoutError::Timeout(res)) => {
|
||||
self.num_timeout += 1;
|
||||
// Safety: we send Ok.
|
||||
batch = res.unwrap();
|
||||
}
|
||||
Err(SendTimeoutError::Closed(res)) => {
|
||||
self.senders[sender_idx] = None;
|
||||
self.num_nones += 1;
|
||||
// Safety: we send Ok.
|
||||
batch = res.unwrap();
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
Ok(())
|
||||
}
|
||||
|
||||
async fn send_error(&self, error: Error) {
|
||||
let error = Arc::new(error);
|
||||
for sender in self.senders.iter().flatten() {
|
||||
let result = Err(error.clone()).context(ScanSeriesSnafu);
|
||||
let _ = sender.send(result).await;
|
||||
}
|
||||
}
|
||||
|
||||
fn fetch_add_sender_idx(&mut self) -> usize {
|
||||
let sender_idx = self.sender_idx;
|
||||
self.sender_idx = (self.sender_idx + 1) % self.senders.len();
|
||||
sender_idx
|
||||
}
|
||||
}
|
||||
|
||||
fn new_partition_metrics(
|
||||
stream_ctx: &StreamContext,
|
||||
metrics_set: &ExecutionPlanMetricsSet,
|
||||
partition: usize,
|
||||
metrics_list: &PartitionMetricsList,
|
||||
) -> PartitionMetrics {
|
||||
let metrics = PartitionMetrics::new(
|
||||
stream_ctx.input.mapper.metadata().region_id,
|
||||
partition,
|
||||
"SeriesScan",
|
||||
stream_ctx.query_start,
|
||||
metrics_set,
|
||||
);
|
||||
|
||||
metrics_list.set(partition, metrics.clone());
|
||||
metrics
|
||||
}
|
||||
@@ -346,6 +346,7 @@ impl BloomFilterIndexer {
|
||||
|
||||
#[cfg(test)]
|
||||
pub(crate) mod tests {
|
||||
use std::iter;
|
||||
|
||||
use api::v1::SemanticType;
|
||||
use datatypes::data_type::ConcreteDataType;
|
||||
@@ -460,15 +461,15 @@ pub(crate) mod tests {
|
||||
|
||||
Batch::new(
|
||||
primary_key,
|
||||
Arc::new(UInt64Vector::from_iter_values(std::iter::repeat_n(
|
||||
0, num_rows,
|
||||
))),
|
||||
Arc::new(UInt64Vector::from_iter_values(std::iter::repeat_n(
|
||||
0, num_rows,
|
||||
))),
|
||||
Arc::new(UInt8Vector::from_iter_values(std::iter::repeat_n(
|
||||
1, num_rows,
|
||||
))),
|
||||
Arc::new(UInt64Vector::from_iter_values(
|
||||
iter::repeat(0).take(num_rows),
|
||||
)),
|
||||
Arc::new(UInt64Vector::from_iter_values(
|
||||
iter::repeat(0).take(num_rows),
|
||||
)),
|
||||
Arc::new(UInt8Vector::from_iter_values(
|
||||
iter::repeat(1).take(num_rows),
|
||||
)),
|
||||
vec![u64_field],
|
||||
)
|
||||
.unwrap()
|
||||
|
||||
@@ -489,12 +489,12 @@ mod tests {
|
||||
Arc::new(UInt64Vector::from_iter_values(
|
||||
(0..num_rows).map(|n| n as u64),
|
||||
)),
|
||||
Arc::new(UInt64Vector::from_iter_values(std::iter::repeat_n(
|
||||
0, num_rows,
|
||||
))),
|
||||
Arc::new(UInt8Vector::from_iter_values(std::iter::repeat_n(
|
||||
1, num_rows,
|
||||
))),
|
||||
Arc::new(UInt64Vector::from_iter_values(
|
||||
std::iter::repeat(0).take(num_rows),
|
||||
)),
|
||||
Arc::new(UInt8Vector::from_iter_values(
|
||||
std::iter::repeat(1).take(num_rows),
|
||||
)),
|
||||
vec![
|
||||
BatchColumn {
|
||||
column_id: 1,
|
||||
|
||||
@@ -326,6 +326,7 @@ impl InvertedIndexer {
|
||||
#[cfg(test)]
|
||||
mod tests {
|
||||
use std::collections::BTreeSet;
|
||||
use std::iter;
|
||||
|
||||
use api::v1::SemanticType;
|
||||
use datafusion_expr::{binary_expr, col, lit, Expr as DfExpr, Operator};
|
||||
@@ -423,15 +424,15 @@ mod tests {
|
||||
|
||||
Batch::new(
|
||||
primary_key,
|
||||
Arc::new(UInt64Vector::from_iter_values(std::iter::repeat_n(
|
||||
0, num_rows,
|
||||
))),
|
||||
Arc::new(UInt64Vector::from_iter_values(std::iter::repeat_n(
|
||||
0, num_rows,
|
||||
))),
|
||||
Arc::new(UInt8Vector::from_iter_values(std::iter::repeat_n(
|
||||
1, num_rows,
|
||||
))),
|
||||
Arc::new(UInt64Vector::from_iter_values(
|
||||
iter::repeat(0).take(num_rows),
|
||||
)),
|
||||
Arc::new(UInt64Vector::from_iter_values(
|
||||
iter::repeat(0).take(num_rows),
|
||||
)),
|
||||
Arc::new(UInt8Vector::from_iter_values(
|
||||
iter::repeat(1).take(num_rows),
|
||||
)),
|
||||
vec![u64_field],
|
||||
)
|
||||
.unwrap()
|
||||
|
||||
@@ -27,7 +27,7 @@ use snafu::{OptionExt, ResultExt};
|
||||
use store_api::storage::TimeSeriesRowSelector;
|
||||
|
||||
use crate::error::{
|
||||
DecodeStatsSnafu, FieldTypeMismatchSnafu, RecordBatchSnafu, Result, StatsNotPresentSnafu,
|
||||
DecodeStatsSnafu, FieldTypeMismatchSnafu, FilterRecordBatchSnafu, Result, StatsNotPresentSnafu,
|
||||
};
|
||||
use crate::read::compat::CompatBatch;
|
||||
use crate::read::last_row::RowGroupLastRowCachedReader;
|
||||
@@ -294,7 +294,7 @@ impl RangeBase {
|
||||
};
|
||||
if filter
|
||||
.evaluate_scalar(&pk_value)
|
||||
.context(RecordBatchSnafu)?
|
||||
.context(FilterRecordBatchSnafu)?
|
||||
{
|
||||
continue;
|
||||
} else {
|
||||
@@ -311,11 +311,11 @@ impl RangeBase {
|
||||
let field_col = &input.fields()[field_index].data;
|
||||
filter
|
||||
.evaluate_vector(field_col)
|
||||
.context(RecordBatchSnafu)?
|
||||
.context(FilterRecordBatchSnafu)?
|
||||
}
|
||||
SemanticType::Timestamp => filter
|
||||
.evaluate_vector(input.timestamps())
|
||||
.context(RecordBatchSnafu)?,
|
||||
.context(FilterRecordBatchSnafu)?,
|
||||
};
|
||||
|
||||
mask = mask.bitand(&result);
|
||||
|
||||
Some files were not shown because too many files have changed in this diff Show More
Reference in New Issue
Block a user