Remove additional-query-queue-dimensions-enabled flag (#9135)

* Remove additional-query-queue-dimensions-enabled flag

* Update CHANGELOG

Author: chencs

CHANGELOG.md

@@ -75,7 +75,7 @@
     * `cortex_ruler_independent_rule_evaluation_concurrency_attempts_completed_total`
 * [ENHANCEMENT] Expose a new `s3.session-token` configuration option to enable using temporary security credentials. #8952
 * [ENHANCEMENT] Add HA deduplication features to the `mimir-microservices-mode` development environment. #9012
-* [ENHANCEMENT] Make `-query-frontend.additional-query-queue-dimensions-enabled` and `-query-scheduler.additional-query-queue-dimensions-enabled` non-operational flags in preparation for removal. #8984
+* [ENHANCEMENT] Remove experimental `-query-frontend.additional-query-queue-dimensions-enabled` and `-query-scheduler.additional-query-queue-dimensions-enabled`. Mimir now always includes "query components" as a queue dimension. #8984 #9135
 * [ENHANCEMENT] Add a new ingester endpoint to prepare instances to downscale. #8956
 * [ENHANCEMENT] Query-scheduler: Add `query-scheduler.prioritize-query-components` which, when enabled, will primarily prioritize dequeuing fairly across queue components, and secondarily prioritize dequeuing fairly across tenants. When disabled, tenant fairness is primarily prioritized. `query-scheduler.use-multi-algorithm-query-queue` must be enabled in order to use this flag. #9016 #9071
 * [ENHANCEMENT] Update runtime configuration to read gzip-compressed files with `.gz` extension. #9074

cmd/mimir/config-descriptor.json

@@ -5754,17 +5754,6 @@
           "fieldType": "int",
           "fieldCategory": "advanced"
         },
-        {
-          "kind": "field",
-          "name": "additional_query_queue_dimensions_enabled",
-          "required": false,
-          "desc": "Non-operational: Enqueue query requests with additional queue dimensions to split tenant request queues into subqueues. This enables separate requests to proceed from a tenant's subqueues even when other subqueues are blocked on slow query requests. Must be set on both query-frontend and scheduler to take effect. (default false)",
-          "fieldValue": null,
-          "fieldDefaultValue": false,
-          "fieldFlag": "query-frontend.additional-query-queue-dimensions-enabled",
-          "fieldType": "boolean",
-          "fieldCategory": "experimental"
-        },
         {
           "kind": "field",
           "name": "split_queries_by_interval",
@@ -16066,17 +16055,6 @@
           "fieldFlag": "query-scheduler.max-outstanding-requests-per-tenant",
           "fieldType": "int"
         },
-        {
-          "kind": "field",
-          "name": "additional_query_queue_dimensions_enabled",
-          "required": false,
-          "desc": "Non-operational: Enqueue query requests with additional queue dimensions to split tenant request queues into subqueues. This enables separate requests to proceed from a tenant's subqueues even when other subqueues are blocked on slow query requests. Must be set on both query-frontend and scheduler to take effect. (default false)",
-          "fieldValue": null,
-          "fieldDefaultValue": false,
-          "fieldFlag": "query-scheduler.additional-query-queue-dimensions-enabled",
-          "fieldType": "boolean",
-          "fieldCategory": "experimental"
-        },
         {
           "kind": "field",
           "name": "use_multi_algorithm_query_queue",

cmd/mimir/help-all.txt.tmpl

@@ -2001,8 +2001,6 @@ Usage of ./cmd/mimir/mimir:
     	The timeout for a query. This config option should be set on query-frontend too when query sharding is enabled. This also applies to queries evaluated by the ruler (internally or remotely). (default 2m0s)
   -query-frontend.active-series-write-timeout duration
     	[experimental] Timeout for writing active series responses. 0 means the value from `-server.http-write-timeout` is used. (default 5m0s)
-  -query-frontend.additional-query-queue-dimensions-enabled
-    	[experimental] Non-operational: Enqueue query requests with additional queue dimensions to split tenant request queues into subqueues. This enables separate requests to proceed from a tenant's subqueues even when other subqueues are blocked on slow query requests. Must be set on both query-frontend and scheduler to take effect. (default false)
   -query-frontend.align-queries-with-step
     	Mutate incoming queries to align their start and end with their step to improve result caching.
   -query-frontend.cache-results
@@ -2215,8 +2213,6 @@ Usage of ./cmd/mimir/mimir:
     	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.use-active-series-decoder
     	[experimental] Set to true to use the zero-allocation response decoder for active series queries.
-  -query-scheduler.additional-query-queue-dimensions-enabled
-    	[experimental] Non-operational: Enqueue query requests with additional queue dimensions to split tenant request queues into subqueues. This enables separate requests to proceed from a tenant's subqueues even when other subqueues are blocked on slow query requests. Must be set on both query-frontend and scheduler to take effect. (default false)
   -query-scheduler.grpc-client-config.backoff-max-period duration
     	Maximum delay when backing off. (default 10s)
   -query-scheduler.grpc-client-config.backoff-min-period duration

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

@@ -1623,14 +1623,6 @@ The `frontend` block configures the query-frontend.
 # CLI flag: -query-frontend.instance-port
 [port: <int> | default = 0]
 
-# (experimental) Non-operational: Enqueue query requests with additional queue
-# dimensions to split tenant request queues into subqueues. This enables
-# separate requests to proceed from a tenant's subqueues even when other
-# subqueues are blocked on slow query requests. Must be set on both
-# query-frontend and scheduler to take effect. (default false)
-# CLI flag: -query-frontend.additional-query-queue-dimensions-enabled
-[additional_query_queue_dimensions_enabled: <boolean> | default = false]
-
 # (advanced) 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.
@@ -1714,14 +1706,6 @@ The `query_scheduler` block configures the query-scheduler.
 # CLI flag: -query-scheduler.max-outstanding-requests-per-tenant
 [max_outstanding_requests_per_tenant: <int> | default = 100]
 
-# (experimental) Non-operational: Enqueue query requests with additional queue
-# dimensions to split tenant request queues into subqueues. This enables
-# separate requests to proceed from a tenant's subqueues even when other
-# subqueues are blocked on slow query requests. Must be set on both
-# query-frontend and scheduler to take effect. (default false)
-# CLI flag: -query-scheduler.additional-query-queue-dimensions-enabled
-[additional_query_queue_dimensions_enabled: <boolean> | default = false]
-
 # (experimental) Use an experimental version of the query queue which has the
 # same behavior as the existing queue, but integrates tenant selection into the
 # tree model.

pkg/frontend/v2/frontend.go

@@ -59,8 +59,6 @@ type Config struct {
 	Addr string `yaml:"address" category:"advanced"`
 	Port int    `category:"advanced"`
 
-	AdditionalQueryQueueDimensionsEnabled bool `yaml:"additional_query_queue_dimensions_enabled" category:"experimental"`
-
 	// These configuration options are injected internally.
 	QuerySchedulerDiscovery schedulerdiscovery.Config `yaml:"-"`
 	LookBackDelta           time.Duration             `yaml:"-"`
@@ -78,8 +76,6 @@ func (cfg *Config) RegisterFlags(f *flag.FlagSet, logger log.Logger) {
 	f.StringVar(&cfg.Addr, "query-frontend.instance-addr", "", "IP address to advertise to the querier (via scheduler) (default is auto-detected from network interfaces).")
 	f.IntVar(&cfg.Port, "query-frontend.instance-port", 0, "Port to advertise to querier (via scheduler) (defaults to server.grpc-listen-port).")
 
-	f.BoolVar(&cfg.AdditionalQueryQueueDimensionsEnabled, "query-frontend.additional-query-queue-dimensions-enabled", false, "Non-operational: Enqueue query requests with additional queue dimensions to split tenant request queues into subqueues. This enables separate requests to proceed from a tenant's subqueues even when other subqueues are blocked on slow query requests. Must be set on both query-frontend and scheduler to take effect. (default false)")
-
 	cfg.GRPCClientConfig.RegisterFlagsWithPrefix("query-frontend.grpc-client-config", f)
 }
 

pkg/scheduler/scheduler.go

@@ -94,19 +94,17 @@ type connectedFrontend struct {
 }
 
 type Config struct {
-	MaxOutstandingPerTenant               int           `yaml:"max_outstanding_requests_per_tenant"`
-	AdditionalQueryQueueDimensionsEnabled bool          `yaml:"additional_query_queue_dimensions_enabled" category:"experimental"`
-	UseMultiAlgorithmQueryQueue           bool          `yaml:"use_multi_algorithm_query_queue" category:"experimental"`
-	PrioritizeQueryComponents             bool          `yaml:"prioritize_query_components" category:"experimental"`
-	QuerierForgetDelay                    time.Duration `yaml:"querier_forget_delay" category:"experimental"`
+	MaxOutstandingPerTenant     int           `yaml:"max_outstanding_requests_per_tenant"`
+	UseMultiAlgorithmQueryQueue bool          `yaml:"use_multi_algorithm_query_queue" category:"experimental"`
+	PrioritizeQueryComponents   bool          `yaml:"prioritize_query_components" category:"experimental"`
+	QuerierForgetDelay          time.Duration `yaml:"querier_forget_delay" category:"experimental"`
 
 	GRPCClientConfig grpcclient.Config         `yaml:"grpc_client_config" doc:"description=This configures the gRPC client used to report errors back to the query-frontend."`
 	ServiceDiscovery schedulerdiscovery.Config `yaml:",inline"`
 }
 
 func (cfg *Config) RegisterFlags(f *flag.FlagSet, logger log.Logger) {
 	f.IntVar(&cfg.MaxOutstandingPerTenant, "query-scheduler.max-outstanding-requests-per-tenant", 100, "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.BoolVar(&cfg.AdditionalQueryQueueDimensionsEnabled, "query-scheduler.additional-query-queue-dimensions-enabled", false, "Non-operational: Enqueue query requests with additional queue dimensions to split tenant request queues into subqueues. This enables separate requests to proceed from a tenant's subqueues even when other subqueues are blocked on slow query requests. Must be set on both query-frontend and scheduler to take effect. (default false)")
 	f.BoolVar(&cfg.UseMultiAlgorithmQueryQueue, "query-scheduler.use-multi-algorithm-query-queue", false, "Use an experimental version of the query queue which has the same behavior as the existing queue, but integrates tenant selection into the tree model.")
 	f.BoolVar(&cfg.PrioritizeQueryComponents, "query-scheduler.prioritize-query-components", false, "When enabled, the query scheduler primarily prioritizes dequeuing fairly from queue components and secondarily prioritizes dequeuing fairly across tenants. When disabled, the query scheduler primarily prioritizes tenant fairness. You must enable the `query-scheduler.use-multi-algorithm-query-queue` setting to use this flag.")
 	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.")