Store: Support disable cache index header file. (#5773)

* Store: Support disable cache index header file. Signed-off-by: wanjunlei <[email protected]> * Store: add a seprate flag to disable caching index header file Signed-off-by: wanjunlei <[email protected]> * Tools: add cleanup API for bucket web Signed-off-by: wanjunlei <[email protected]> * resolve conversation Signed-off-by: wanjunlei <[email protected]> * resolve confilcts Signed-off-by: wanjunlei <[email protected]> * change the flag to `--cache-index-header` Signed-off-by: wanjunlei <[email protected]> * Wrap mem writer in file writer Signed-off-by: wanjunlei <[email protected]> * update CHANGELOG Signed-off-by: wanjunlei <[email protected]> * update CHANGELOG Signed-off-by: wanjunlei <[email protected]> * fix bug Signed-off-by: wanjunlei <[email protected]> --------- Signed-off-by: wanjunlei <[email protected]> Co-authored-by: wanjunlei <[email protected]>
thanos-io · Jan 28, 2023 · 1b12ab9 · 1b12ab9
1 parent 431a6cf
commit 1b12ab9
Show file tree

Hide file tree

Showing 8 changed files with 546 additions and 171 deletions.
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -14,6 +14,7 @@ We use *breaking :warning:* to mark changes that are not backward compatible (re
 
 - [#5990](https://github.com/thanos-io/thanos/pull/5990) Cache/Redis: add support for Redis Sentinel via new option `master_name`.
 - [#6008](https://github.com/thanos-io/thanos/pull/6008) *: Add counter metric `gate_queries_total` to gate.
+- [#5773](https://github.com/thanos-io/thanos/pull/5773) Store: Support disable cache index header file.
 
 ### Changed
 

diff --git a/cmd/thanos/store.go b/cmd/thanos/store.go
@@ -58,6 +58,7 @@ type storeConfig struct {
 	indexCacheConfigs           extflag.PathOrContent
 	objStoreConfig              extflag.PathOrContent
 	dataDir                     string
+	cacheIndexHeader            bool
 	grpcConfig                  grpcConfig
 	httpConfig                  httpConfig
 	indexCacheSizeBytes         units.Base2Bytes
@@ -90,9 +91,12 @@ func (sc *storeConfig) registerFlag(cmd extkingpin.FlagClause) {
 	sc.httpConfig = *sc.httpConfig.registerFlag(cmd)
 	sc.grpcConfig = *sc.grpcConfig.registerFlag(cmd)
 
-	cmd.Flag("data-dir", "Local data directory used for caching purposes (index-header, in-mem cache items and meta.jsons). If removed, no data will be lost, just store will have to rebuild the cache. NOTE: Putting raw blocks here will not cause the store to read them. For such use cases use Prometheus + sidecar.").
+	cmd.Flag("data-dir", "Local data directory used for caching purposes (index-header, in-mem cache items and meta.jsons). If removed, no data will be lost, just store will have to rebuild the cache. NOTE: Putting raw blocks here will not cause the store to read them. For such use cases use Prometheus + sidecar. Ignored if -no-cache-index-header option is specified.").
 		Default("./data").StringVar(&sc.dataDir)
 
+	cmd.Flag("cache-index-header", "Cache TSDB index-headers on disk to reduce startup time. When set to true, Thanos Store will download index headers from remote object storage on startup and create a header file on disk. Use --data-dir to set the directory in which index headers will be downloaded.").
+		Default("true").BoolVar(&sc.cacheIndexHeader)
+
 	cmd.Flag("index-cache-size", "Maximum size of items held in the in-memory index cache. Ignored if --index-cache.config or --index-cache.config-file option is specified.").
 		Default("250MB").BytesVar(&sc.indexCacheSizeBytes)
 
@@ -237,6 +241,11 @@ func runStore(
 	conf storeConfig,
 	flagsMap map[string]string,
 ) error {
+	dataDir := conf.dataDir
+	if !conf.cacheIndexHeader {
+		dataDir = ""
+	}
+
 	grpcProbe := prober.NewGRPC()
 	httpProbe := prober.NewHTTP()
 	statusProber := prober.Combine(
@@ -318,7 +327,7 @@ func runStore(
 	}
 
 	ignoreDeletionMarkFilter := block.NewIgnoreDeletionMarkFilter(logger, bkt, time.Duration(conf.ignoreDeletionMarksDelay), conf.blockMetaFetchConcurrency)
-	metaFetcher, err := block.NewMetaFetcher(logger, conf.blockMetaFetchConcurrency, bkt, conf.dataDir, extprom.WrapRegistererWithPrefix("thanos_", reg),
+	metaFetcher, err := block.NewMetaFetcher(logger, conf.blockMetaFetchConcurrency, bkt, dataDir, extprom.WrapRegistererWithPrefix("thanos_", reg),
 		[]block.MetadataFilter{
 			block.NewTimePartitionMetaFilter(conf.filterConf.MinTime, conf.filterConf.MaxTime),
 			block.NewLabelShardedMetaFilter(relabelConfig),
@@ -360,7 +369,7 @@ func runStore(
 	bs, err := store.NewBucketStore(
 		bkt,
 		metaFetcher,
-		conf.dataDir,
+		dataDir,
 		store.NewChunksLimiterFactory(conf.maxSampleCount/store.MaxSamplesPerChunk), // The samples limit is an approximation based on the max number of samples per chunk.
 		store.NewSeriesLimiterFactory(conf.maxTouchedSeriesCount),
 		store.NewBytesLimiterFactory(conf.maxDownloadedBytes),

diff --git a/docs/components/store.md b/docs/components/store.md
@@ -36,6 +36,12 @@ Flags:
                                  Number of goroutines to use when constructing
                                  index-cache.json blocks from object storage.
                                  Must be equal or greater than 1.
+      --cache-index-header       Cache TSDB index-headers on disk to reduce
+                                 startup time. When set to true, Thanos Store
+                                 will download index headers from remote object
+                                 storage on startup and create a header file on
+                                 disk. Use --data-dir to set the directory in
+                                 which index headers will be downloaded.
       --chunk-pool-size=2GB      Maximum size of concurrently allocatable
                                  bytes reserved strictly to reuse for chunks in
                                  memory.
@@ -47,9 +53,10 @@ Flags:
                                  purposes (index-header, in-mem cache items and
                                  meta.jsons). If removed, no data will be lost,
                                  just store will have to rebuild the cache.
-                                 NOTE: Putting raw blocks here will not cause
-                                 the store to read them. For such use cases use
-                                 Prometheus + sidecar.
+                                 NOTE: Putting raw blocks here will not
+                                 cause the store to read them. For such use
+                                 cases use Prometheus + sidecar. Ignored if
+                                 -no-cache-index-header option is specified.
       --grpc-address="0.0.0.0:10901"
                                  Listen ip:port address for gRPC endpoints
                                  (StoreAPI). Make sure this address is routable