add changelog

Signed-off-by: Ben Ye <benye@amazon.com>

Author: yeya24

CHANGELOG.md

@@ -3,6 +3,7 @@
 ## master / unreleased
 * [ENHANCEMENT] Update Go version to 1.19.3. #4988
 * [ENHANCEMENT] Querier: limit series query to only ingesters if `start` param is not specified. #4976
+* [ENHANCEMENT] Query-frontend/scheduler: add a new limit `frontend.max-outstanding-requests-per-tenant` for configuring queue size per tenant. Started deprecating two flags `-query-scheduler.max-outstanding-requests-per-tenant` and `-querier.max-outstanding-requests-per-tenant`, and change their value default to 0. Now if both the old flag and new flag are specified, the old flag's queue size will be picked. #5005
 * [FEATURE] Querier/Query Frontend: support Prometheus /api/v1/status/buildinfo API. #4978
 * [FEATURE] Ingester: Add active series to all_user_stats page. #4972
 * [FEATURE] Query Frontend: Log query params in query frontend even if error happens. #5005

docs/configuration/config-file-reference.md

@@ -172,7 +172,7 @@ runtime_config:
 [memberlist: <memberlist_config>]
 
 query_scheduler:
-  # Deprecated (user frontend.max-outstanding-requests-per-tenant instead):
+  # Deprecated (use frontend.max-outstanding-requests-per-tenant instead):
   # Maximum number of outstanding requests per tenant per query-scheduler.
   # In-flight requests above this limit will fail with HTTP response status code
   # 429.
@@ -917,9 +917,9 @@ The `query_frontend_config` configures the Cortex query-frontend.
 # CLI flag: -frontend.query-stats-enabled
 [query_stats_enabled: <boolean> | default = false]
 
-# Deprecated (user frontend.max-outstanding-requests-per-tenant instead):
-# Maximum number of outstanding requests per tenant per frontend; requests
-# beyond this error with HTTP 429.
+# Deprecated (use frontend.max-outstanding-requests-per-tenant instead): Maximum
+# number of outstanding requests per tenant per frontend; requests beyond this
+# error with HTTP 429.
 # CLI flag: -querier.max-outstanding-requests-per-tenant
 [max_outstanding_per_tenant: <int> | default = 0]
 

pkg/frontend/v1/frontend.go

@@ -37,7 +37,7 @@ type Config struct {
 
 // RegisterFlags adds the flags required to config this to the given FlagSet.
 func (cfg *Config) RegisterFlags(f *flag.FlagSet) {
-	f.IntVar(&cfg.MaxOutstandingPerTenant, "querier.max-outstanding-requests-per-tenant", 0, "Deprecated (user frontend.max-outstanding-requests-per-tenant instead): Maximum number of outstanding requests per tenant per frontend; requests beyond this error with HTTP 429.")
+	f.IntVar(&cfg.MaxOutstandingPerTenant, "querier.max-outstanding-requests-per-tenant", 0, "Deprecated (use frontend.max-outstanding-requests-per-tenant instead) and will be removed in v1.17.0: Maximum number of outstanding requests per tenant per frontend; requests beyond this error with HTTP 429.")
 	f.DurationVar(&cfg.QuerierForgetDelay, "query-frontend.querier-forget-delay", 0, "If a querier disconnects without sending notification about graceful shutdown, the query-frontend will keep the querier in the tenant's shard until the forget delay has passed. This feature is useful to reduce the blast radius when shuffle-sharding is enabled.")
 }
 

pkg/scheduler/queue/user_queues.go

@@ -117,8 +117,8 @@ func (q *queues) getOrAddQueue(userID string, maxQueriers int) chan Request {
 
 	if uq == nil {
 		queueSize := q.limits.MaxOutstandingPerTenant(userID)
-		// 0 is the default value. If the flag is set then we
-		// use the old flag for compatibility reason.
+		// 0 is the default value of the flag. If the old flag is set
+		// then we use its value for compatibility reason.
 		if q.maxUserQueueSize != 0 {
 			queueSize = q.maxUserQueueSize
 		}

pkg/scheduler/scheduler.go

@@ -86,7 +86,7 @@ type Config struct {
 }
 
 func (cfg *Config) RegisterFlags(f *flag.FlagSet) {
-	f.IntVar(&cfg.MaxOutstandingPerTenant, "query-scheduler.max-outstanding-requests-per-tenant", 0, "Deprecated (user frontend.max-outstanding-requests-per-tenant instead): Maximum number of outstanding requests per tenant per query-scheduler. In-flight requests above this limit will fail with HTTP response status code 429.")
+	f.IntVar(&cfg.MaxOutstandingPerTenant, "query-scheduler.max-outstanding-requests-per-tenant", 0, "Deprecated (use frontend.max-outstanding-requests-per-tenant instead) and will be removed in v1.17.0: Maximum number of outstanding requests per tenant per query-scheduler. In-flight requests above this limit will fail with HTTP response status code 429.")
 	f.DurationVar(&cfg.QuerierForgetDelay, "query-scheduler.querier-forget-delay", 0, "If a querier disconnects without sending notification about graceful shutdown, the query-scheduler will keep the querier in the tenant's shard until the forget delay has passed. This feature is useful to reduce the blast radius when shuffle-sharding is enabled.")
 	cfg.GRPCClientConfig.RegisterFlagsWithPrefix("query-scheduler.grpc-client-config", f)
 }