Reduce backoff time + update doc

Author: justinjung04

docs/blocks-storage/querier.md

@@ -238,7 +238,7 @@ querier:
   [store_gateway_consistency_check_max_attempts: <int> | default = 3]
 
   # The maximum number of times we attempt fetching data from ingesters for
-  # retryable errors.
+  # retryable errors (ex. partial data returned).
   # CLI flag: -querier.ingester-query-max-attempts
   [ingester_query_max_attempts: <int> | default = 1]
 

docs/configuration/config-file-reference.md

@@ -4166,7 +4166,7 @@ store_gateway_client:
 [store_gateway_consistency_check_max_attempts: <int> | default = 3]
 
 # The maximum number of times we attempt fetching data from ingesters for
-# retryable errors.
+# retryable errors (ex. partial data returned).
 # CLI flag: -querier.ingester-query-max-attempts
 [ingester_query_max_attempts: <int> | default = 1]
 

pkg/querier/distributor_queryable.go

@@ -25,8 +25,8 @@ import (
 	"github.com/cortexproject/cortex/pkg/util/spanlogger"
 )
 
-const retryMinBackoff = time.Second
-const retryMaxBackoff = 5 * time.Second
+const retryMinBackoff = time.Millisecond
+const retryMaxBackoff = 5 * time.Millisecond
 
 // Distributor is the read interface to the distributor, made an interface here
 // to reduce package coupling.
@@ -203,7 +203,7 @@ func (q *distributorQuerier) streamingSelect(ctx context.Context, sortSeries, pa
 }
 
 func (q *distributorQuerier) queryWithRetry(ctx context.Context, queryFunc func() (*client.QueryStreamResponse, error)) (*client.QueryStreamResponse, error) {
-	if q.ingesterQueryMaxAttempts == 1 {
+	if q.ingesterQueryMaxAttempts <= 1 {
 		return queryFunc()
 	}
 

pkg/querier/querier.go

@@ -128,7 +128,7 @@ func (cfg *Config) RegisterFlags(f *flag.FlagSet) {
 	f.StringVar(&cfg.StoreGatewayAddresses, "querier.store-gateway-addresses", "", "Comma separated list of store-gateway addresses in DNS Service Discovery format. This option should be set when using the blocks storage and the store-gateway sharding is disabled (when enabled, the store-gateway instances form a ring and addresses are picked from the ring).")
 	f.BoolVar(&cfg.StoreGatewayQueryStatsEnabled, "querier.store-gateway-query-stats-enabled", true, "If enabled, store gateway query stats will be logged using `info` log level.")
 	f.IntVar(&cfg.StoreGatewayConsistencyCheckMaxAttempts, "querier.store-gateway-consistency-check-max-attempts", maxFetchSeriesAttempts, "The maximum number of times we attempt fetching missing blocks from different store-gateways. If no more store-gateways are left (ie. due to lower replication factor) than we'll end the retries earlier")
-	f.IntVar(&cfg.IngesterQueryMaxAttempts, "querier.ingester-query-max-attempts", 1, "The maximum number of times we attempt fetching data from ingesters for retryable errors.")
+	f.IntVar(&cfg.IngesterQueryMaxAttempts, "querier.ingester-query-max-attempts", 1, "The maximum number of times we attempt fetching data from ingesters for retryable errors (ex. partial data returned).")
 	f.DurationVar(&cfg.LookbackDelta, "querier.lookback-delta", 5*time.Minute, "Time since the last sample after which a time series is considered stale and ignored by expression evaluations.")
 	f.DurationVar(&cfg.ShuffleShardingIngestersLookbackPeriod, "querier.shuffle-sharding-ingesters-lookback-period", 0, "When distributor's sharding strategy is shuffle-sharding and this setting is > 0, queriers fetch in-memory series from the minimum set of required ingesters, selecting only ingesters which may have received series since 'now - lookback period'. The lookback period should be greater or equal than the configured 'query store after' and 'query ingesters within'. If this setting is 0, queriers always query all ingesters (ingesters shuffle sharding on read path is disabled).")
 	f.BoolVar(&cfg.ThanosEngine, "querier.thanos-engine", false, "Experimental. Use Thanos promql engine https://github.com/thanos-io/promql-engine rather than the Prometheus promql engine.")