chore: query-frontend: default to active series sharding (#15091)

Remove experimental settings for enabling/disabling active series
sharding and use of a zero-allocation decoder. These have been enabled
in Grafana Cloud for a over a year, they're not experimental anymore.

Author: 56quarters

CHANGELOG.md

@@ -24,6 +24,7 @@
 * [CHANGE] Query-frontend: Removed support for calculating 'cache-adjusted samples processed' query statistic. The `-query-frontend.cache-samples-processed-stats` CLI flag has been deprecated and will be removed in a future release. Setting it has now no effect. #13582
 * [CHANGE] Querier: Renamed experimental flag `-querier.prefer-availability-zone` to `-querier.prefer-availability-zones` and changed it to accept a comma-separated list of availability zones. All zones in the list are given equal priority when querying ingesters and store-gateways. #13756 #13758
 * [CHANGE] Ingester: Stabilize experimental flag `-ingest-storage.write-logs-fsync-before-kafka-commit-concurrency` to fsync write logs before the offset is committed to Kafka. Remove `-ingest-storage.write-logs-fsync-before-kafka-commit-enabled` since this is always enabled now. #13591
+* [CHANGE] Query-frontend: Remove experimental flags `-query-frontend.shard-active-series-queries`, `-query-frontend.use-active-series-decoder`. These settings are always enabled now. #15091
 * [CHANGE] Ingester: Remove metric `cortex_ingester_owned_target_info_series`; if needed this is better done as a custom active series tracker. #13831
 * [CHANGE] Querier: Remove experimental flag `-querier.response-streaming-enabled`, active series responses are now always streamed to query-frontends. #14095 #14114
 * [CHANGE] Store-gateway: Warn when loading index headers based on TSDB blocks that use v1 of the index file format. #13834

cmd/mimir/config-descriptor.json

@@ -8660,28 +8660,6 @@
           "fieldType": "int",
           "fieldCategory": "advanced"
         },
-        {
-          "kind": "field",
-          "name": "shard_active_series_queries",
-          "required": false,
-          "desc": "True to enable sharding of active series queries.",
-          "fieldValue": null,
-          "fieldDefaultValue": false,
-          "fieldFlag": "query-frontend.shard-active-series-queries",
-          "fieldType": "boolean",
-          "fieldCategory": "experimental"
-        },
-        {
-          "kind": "field",
-          "name": "use_active_series_decoder",
-          "required": false,
-          "desc": "Set to true to use the zero-allocation response decoder for active series queries.",
-          "fieldValue": null,
-          "fieldDefaultValue": false,
-          "fieldFlag": "query-frontend.use-active-series-decoder",
-          "fieldType": "boolean",
-          "fieldCategory": "experimental"
-        },
         {
           "kind": "field",
           "name": "extra_propagated_headers",

cmd/mimir/help-all.txt.tmpl

@@ -2655,14 +2655,10 @@ Usage of ./cmd/mimir/mimir:
     	How often to resolve the scheduler-address, in order to look for new query-scheduler instances. (default 10s)
   -query-frontend.scheduler-worker-concurrency int
     	Number of concurrent workers forwarding queries to single query-scheduler. (default 5)
-  -query-frontend.shard-active-series-queries
-    	[experimental] True to enable sharding of active series queries.
   -query-frontend.split-queries-by-interval duration
     	Split range queries by an interval and execute in parallel. You should use a multiple of 24 hours to optimize querying blocks. 0 to disable it. (default 24h0m0s)
   -query-frontend.subquery-spin-off-enabled
     	[experimental] Enable spinning off subqueries from instant queries as range queries to optimize their performance.
-  -query-frontend.use-active-series-decoder
-    	[experimental] Set to true to use the zero-allocation response decoder for active series queries.
   -query-frontend.use-mimir-query-engine-for-sharding
     	[experimental] Set to true to enable performing query sharding inside the Mimir query engine (MQE). This setting has no effect if sharding is disabled. Requires remote execution and MQE to be enabled.
   -query-scheduler.grpc-client-config.backoff-max-period duration

development/mimir-microservices-mode/config/mimir.yaml

@@ -149,7 +149,6 @@ frontend:
   parallelize_shardable_queries: true
   cache_results: true
   cache_errors: true
-  shard_active_series_queries: true
 
   # Uncomment when using "dns" service discovery mode for query-scheduler.
   # scheduler_address: "query-scheduler:9008"

docs/sources/mimir/configure/configuration-parameters/index.md

@@ -2258,15 +2258,6 @@ results_cache:
 # CLI flag: -query-frontend.query-sharding-target-series-per-shard
 [query_sharding_target_series_per_shard: <int> | default = 0]
 
-# (experimental) True to enable sharding of active series queries.
-# CLI flag: -query-frontend.shard-active-series-queries
-[shard_active_series_queries: <boolean> | default = false]
-
-# (experimental) Set to true to use the zero-allocation response decoder for
-# active series queries.
-# CLI flag: -query-frontend.use-active-series-decoder
-[use_active_series_decoder: <boolean> | default = false]
-
 # (advanced) Comma-separated list of request header names to allow to pass
 # through to the rest of the query path. This is in addition to a list of
 # required headers that the read path needs.

integration/query_frontend_active_series_test.go

@@ -3,7 +3,6 @@
 package integration
 
 import (
-	"fmt"
 	"net/http"
 	"strconv"
 	"testing"
@@ -21,43 +20,32 @@ import (
 )
 
 func TestActiveSeriesWithQuerySharding(t *testing.T) {
-	for _, tc := range []struct {
-		shardingEnabled bool
-	}{
-		{false},
-		{true},
-	} {
-		config := queryFrontendTestConfig{
-			queryStatsEnabled:           true,
-			shardActiveSeriesQueries:    tc.shardingEnabled,
-			querySchedulerDiscoveryMode: "ring",
-			setup: func(t *testing.T, s *e2e.Scenario) (string, map[string]string) {
-				flags := mergeFlags(BlocksStorageFlags(), BlocksStorageS3Flags())
-				minio := e2edb.NewMinio(9000, flags["-blocks-storage.s3.bucket-name"])
-				require.NoError(t, s.StartAndWaitReady(minio))
-
-				return "", flags
-			},
-		}
+	config := queryFrontendTestConfig{
+		queryStatsEnabled:           true,
+		querySchedulerDiscoveryMode: "ring",
+		setup: func(t *testing.T, s *e2e.Scenario) (string, map[string]string) {
+			flags := mergeFlags(BlocksStorageFlags(), BlocksStorageS3Flags())
+			minio := e2edb.NewMinio(9000, flags["-blocks-storage.s3.bucket-name"])
+			require.NoError(t, s.StartAndWaitReady(minio))
+
+			return "", flags
+		},
+	}
 
-		testName := fmt.Sprintf("query query_sharding=%v",
-			tc.shardingEnabled,
-		)
-		t.Run("active series/"+testName, func(t *testing.T) {
-			runTestActiveSeriesWithQueryShardingHTTPTest(t, config)
-		})
+	t.Run("active series", func(t *testing.T) {
+		runTestActiveSeriesHTTPTest(t, config)
+	})
 
-		t.Run("active native histograms/"+testName, func(t *testing.T) {
-			runTestActiveNativeHistogramMetricsWithQueryShardingHTTPTest(t, config)
-		})
-	}
+	t.Run("active native histograms", func(t *testing.T) {
+		runTestActiveNativeHistogramMetricsHTTPTest(t, config)
+	})
 }
 
 const (
 	metricName = "test_metric"
 )
 
-func runTestActiveSeriesWithQueryShardingHTTPTest(t *testing.T, cfg queryFrontendTestConfig) {
+func runTestActiveSeriesHTTPTest(t *testing.T, cfg queryFrontendTestConfig) {
 	numSeries := 100
 	s, c := setupComponentsForActiveSeriesTest(t, cfg, numSeries)
 	defer s.Close()
@@ -76,17 +64,12 @@ func runTestActiveSeriesWithQueryShardingHTTPTest(t *testing.T, cfg queryFronten
 		require.Len(t, response.Data, numSeries)
 	}
 
-	var err error
-	_, err = c.ActiveSeries(metricName, e2emimir.WithQueryShards(512))
-	if cfg.shardActiveSeriesQueries {
-		require.Error(t, err)
-		require.Contains(t, err.Error(), "shard count 512 exceeds allowed maximum (128)")
-	} else {
-		require.NoError(t, err)
-	}
+	_, err := c.ActiveSeries(metricName, e2emimir.WithQueryShards(512))
+	require.Error(t, err)
+	require.Contains(t, err.Error(), "shard count 512 exceeds allowed maximum (128)")
 }
 
-func runTestActiveNativeHistogramMetricsWithQueryShardingHTTPTest(t *testing.T, cfg queryFrontendTestConfig) {
+func runTestActiveNativeHistogramMetricsHTTPTest(t *testing.T, cfg queryFrontendTestConfig) {
 	numSeries := 100
 	s, c := setupComponentsForActiveSeriesTest(t, cfg, numSeries)
 	defer s.Close()
@@ -113,14 +96,9 @@ func runTestActiveNativeHistogramMetricsWithQueryShardingHTTPTest(t *testing.T,
 		})
 	}
 
-	var err error
-	_, err = c.ActiveNativeHistogramMetrics(metricName, e2emimir.WithQueryShards(512))
-	if cfg.shardActiveSeriesQueries {
-		require.Error(t, err)
-		require.Contains(t, err.Error(), "shard count 512 exceeds allowed maximum (128)")
-	} else {
-		require.NoError(t, err)
-	}
+	_, err := c.ActiveNativeHistogramMetrics(metricName, e2emimir.WithQueryShards(512))
+	require.Error(t, err)
+	require.Contains(t, err.Error(), "shard count 512 exceeds allowed maximum (128)")
 }
 
 func setupComponentsForActiveSeriesTest(t *testing.T, cfg queryFrontendTestConfig, numSeries int) (*e2e.Scenario, *e2emimir.Client) {
@@ -160,10 +138,6 @@ func setupComponentsForActiveSeriesTest(t *testing.T, cfg queryFrontendTestConfi
 		flags["-querier.scheduler-address"] = queryScheduler.NetworkGRPCEndpoint()
 	}
 
-	if cfg.shardActiveSeriesQueries {
-		flags["-query-frontend.shard-active-series-queries"] = "true"
-	}
-
 	// Start the query-frontend.
 	queryFrontend := e2emimir.NewQueryFrontend("query-frontend", consul.NetworkHTTPEndpoint(), flags, e2emimir.WithConfigFile(configFile))
 	require.NoError(t, s.Start(queryFrontend))

integration/query_frontend_test.go

@@ -41,7 +41,6 @@ type queryFrontendTestConfig struct {
 	queryStatsEnabled           bool
 	setup                       func(t *testing.T, s *e2e.Scenario) (configFile string, flags map[string]string)
 	withHistograms              bool
-	shardActiveSeriesQueries    bool
 	remoteExecutionEnabled      bool
 }
 

pkg/frontend/querymiddleware/roundtrip.go

@@ -73,8 +73,6 @@ type Config struct {
 	RewriteQueriesHistogram                   bool               `yaml:"rewrite_histogram_queries" category:"experimental"`
 	RewriteQueriesPropagateMatchers           bool               `yaml:"rewrite_propagate_matchers" category:"experimental"`
 	TargetSeriesPerShard                      uint64             `yaml:"query_sharding_target_series_per_shard" category:"advanced"`
-	ShardActiveSeriesQueries                  bool               `yaml:"shard_active_series_queries" category:"experimental"`
-	UseActiveSeriesDecoder                    bool               `yaml:"use_active_series_decoder" category:"experimental"`
 
 	// CacheKeyGenerator allows to inject a CacheKeyGenerator to use for generating cache keys.
 	// If nil, the querymiddleware package uses a DefaultCacheKeyGenerator with SplitQueriesByInterval.
@@ -114,8 +112,6 @@ func (cfg *Config) RegisterFlags(f *flag.FlagSet) {
 	f.Uint64Var(&cfg.TargetSeriesPerShard, "query-frontend.query-sharding-target-series-per-shard", 0, "How many series a single sharded partial query should load at most. This is not a strict requirement guaranteed to be honoured by query sharding, but a hint given to the query sharding when the query execution is initially planned. 0 to disable cardinality-based hints.")
 	f.Var(&cfg.ExtraPropagateHeaders, "query-frontend.extra-propagated-headers", "Comma-separated list of request header names to allow to pass through to the rest of the query path. This is in addition to a list of required headers that the read path needs.")
 	f.StringVar(&cfg.QueryResultResponseFormat, "query-frontend.query-result-response-format", formatProtobuf, fmt.Sprintf("Format to use when retrieving query results from queriers. Supported values: %s", strings.Join(allFormats, ", ")))
-	f.BoolVar(&cfg.ShardActiveSeriesQueries, "query-frontend.shard-active-series-queries", false, "True to enable sharding of active series queries.")
-	f.BoolVar(&cfg.UseActiveSeriesDecoder, "query-frontend.use-active-series-decoder", false, "Set to true to use the zero-allocation response decoder for active series queries.")
 	f.BoolVar(&cfg.CacheSamplesProcessedStats, "query-frontend.cache-samples-processed-stats", false, "Cache statistics of processed samples on results cache. Deprecated: has no effect.")
 	cfg.ResultsCache.RegisterFlags(f)
 
@@ -332,10 +328,9 @@ func newQueryTripperware(
 			activeSeries = newRetryRoundTripper(series, log, cfg.MaxRetries, retryMetrics)
 		}
 
-		if cfg.ShardActiveSeriesQueries {
-			activeSeries = newShardActiveSeriesMiddleware(activeSeries, cfg.UseActiveSeriesDecoder, limits, log)
-			activeNativeHistogramMetrics = newShardActiveNativeHistogramMetricsMiddleware(activeNativeHistogramMetrics, limits, log)
-		}
+		// Shard active series requests using special middleware.
+		activeSeries = newShardActiveSeriesMiddleware(activeSeries, limits, log)
+		activeNativeHistogramMetrics = newShardActiveNativeHistogramMetricsMiddleware(activeNativeHistogramMetrics, limits, log)
 
 		// Enforce read consistency after caching.
 		if len(ingestStorageTopicOffsetsReaders) > 0 {