Cortex 1.8.0-rc.0

This was a release candidate for 1.8.0.

Cortex 1.7.0

Changelog

Cortex

Note the blocks storage compactor runs a migration task at startup in this version, which can take many minutes and use a lot of RAM.
Turn this off after first run.

• [CHANGE] FramedSnappy encoding support has been removed from Push and Remote Read APIs. This means Prometheus 1.6 support has been removed and the oldest Prometheus version supported in the remote write is 1.7. #3682
• [CHANGE] Ruler: removed the flag -ruler.evaluation-delay-duration-deprecated which was deprecated in 1.4.0. Please use the ruler_evaluation_delay_duration per-tenant limit instead. #3694
• [CHANGE] Removed the flags -<prefix>.grpc-use-gzip-compression which were deprecated in 1.3.0: #3694
  • -query-scheduler.grpc-client-config.grpc-use-gzip-compression: use -query-scheduler.grpc-client-config.grpc-compression instead
  • -frontend.grpc-client-config.grpc-use-gzip-compression: use -frontend.grpc-client-config.grpc-compression instead
  • -ruler.client.grpc-use-gzip-compression: use -ruler.client.grpc-compression instead
  • -bigtable.grpc-use-gzip-compression: use -bigtable.grpc-compression instead
  • -ingester.client.grpc-use-gzip-compression: use -ingester.client.grpc-compression instead
  • -querier.frontend-client.grpc-use-gzip-compression: use -querier.frontend-client.grpc-compression instead
• [CHANGE] Querier: it's not required to set -frontend.query-stats-enabled=true in the querier anymore to enable query statistics logging in the query-frontend. The flag is now required to be configured only in the query-frontend and it will be propagated to the queriers. #3595 #3695
• [CHANGE] Blocks storage: compactor is now required when running a Cortex cluster with the blocks storage, because it also keeps the bucket index updated. #3583
• [CHANGE] Blocks storage: block deletion marks are now stored in a per-tenant global markers/ location too, other than within the block location. The compactor, at startup, will copy deletion marks from the block location to the global location. This migration is required only once, so you can safely disable it via -compactor.block-deletion-marks-migration-enabled=false once new compactor has successfully started once in your cluster. #3583
• [CHANGE] OpenStack Swift: the default value for the -ruler.storage.swift.container-name and -swift.container-name config options has changed from cortex to empty string. If you were relying on the default value, you should set it back to cortex. #3660
• [CHANGE] HA Tracker: configured replica label is now verified against label value length limit (-validation.max-length-label-value). #3668
• [CHANGE] Distributor: extend_writes field in YAML configuration has moved from lifecycler (inside ingester_config) to distributor_config. This doesn't affect command line option -distributor.extend-writes, which stays the same. #3719
• [CHANGE] Alertmanager: Deprecated -cluster. CLI flags in favor of their -alertmanager.cluster. equivalent. The deprecated flags (and their respective YAML config options) are: #3677
  • -cluster.listen-address in favor of -alertmanager.cluster.listen-address
  • -cluster.advertise-address in favor of -alertmanager.cluster.advertise-address
  • -cluster.peer in favor of -alertmanager.cluster.peers
  • -cluster.peer-timeout in favor of -alertmanager.cluster.peer-timeout
• [CHANGE] Blocks storage: the default value of -blocks-storage.bucket-store.sync-interval has been changed from 5m to 15m. #3724
• [FEATURE] Querier: Queries can be federated across multiple tenants. The tenants IDs involved need to be specified separated by a | character in the X-Scope-OrgID request header. This is an experimental feature, which can be enabled by setting -tenant-federation.enabled=true on all Cortex services. #3250
• [FEATURE] Alertmanager: introduced the experimental option -alertmanager.sharding-enabled to shard tenants across multiple Alertmanager instances. This feature is still under heavy development and its usage is discouraged. The following new metrics are exported by the Alertmanager: #3664
  • cortex_alertmanager_ring_check_errors_total
  • cortex_alertmanager_sync_configs_total
  • cortex_alertmanager_sync_configs_failed_total
  • cortex_alertmanager_tenants_discovered
  • cortex_alertmanager_tenants_owned
• [ENHANCEMENT] Allow specifying JAEGER_ENDPOINT instead of sampling server or local agent port. #3682
• [ENHANCEMENT] Blocks storage: introduced a per-tenant bucket index, periodically updated by the compactor, used to avoid full bucket scanning done by queriers, store-gateways and rulers. The bucket index is updated by the compactor during blocks cleanup, on every -compactor.cleanup-interval. #3553 #3555 #3561 #3583 #3625 #3711 #3715
• [ENHANCEMENT] Blocks storage: introduced an option -blocks-storage.bucket-store.bucket-index.enabled to enable the usage of the bucket index in the querier, store-gateway and ruler. When enabled, the querier, store-gateway and ruler will use the bucket index to find a tenant's blocks instead of running the periodic bucket scan. The following new metrics are exported by the querier and ruler: #3614 #3625
  • cortex_bucket_index_loads_total
  • cortex_bucket_index_load_failures_total
  • cortex_bucket_index_load_duration_seconds
  • cortex_bucket_index_loaded
• [ENHANCEMENT] Compactor: exported the following metrics. #3583 #3625
  • cortex_bucket_blocks_count: Total number of blocks per tenant in the bucket. Includes blocks marked for deletion, but not partial blocks.
  • cortex_bucket_blocks_marked_for_deletion_count: Total number of blocks per tenant marked for deletion in the bucket.
  • cortex_bucket_blocks_partials_count: Total number of partial blocks.
  • cortex_bucket_index_last_successful_update_timestamp_seconds: Timestamp of the last successful update of a tenant's bucket index.
• [ENHANCEMENT] Ruler: Add cortex_prometheus_last_evaluation_samples to expose the number of samples generated by a rule group per tenant. #3582
• [ENHANCEMENT] Memberlist: add status page (/memberlist) with available details about memberlist-based KV store and memberlist cluster. It's also possible to view KV values in Go struct or JSON format, or download for inspection. #3575
• [ENHANCEMENT] Memberlist: client can now keep a size-bounded buffer with sent and received messages and display them in the admin UI (/memberlist) for troubleshooting. #3581 #3602
• [ENHANCEMENT] Blocks storage: added block index attributes caching support to metadata cache. The TTL can be configured via -blocks-storage.bucket-store.metadata-cache.block-index-attributes-ttl. #3629
• [ENHANCEMENT] Alertmanager: Add support for Azure blob storage. #3634
• [ENHANCEMENT] Compactor: tenants marked for deletion will now be fully cleaned up after some delay since deletion of last block. Cleanup includes removal of remaining marker files (including tenant deletion mark file) and files under debug/metas. #3613
• [ENHANCEMENT] Compactor: retry compaction of a single tenant on failure instead of re-running compaction for all tenants. #3627
• [ENHANCEMENT] Querier: Implement result caching for tenant query federation. #3640
• [ENHANCEMENT] API: Add a mode query parameter for the config endpoint: #3645
  • /config?mode=diff: Shows the YAML configuration with all values that differ from the defaults.
  • /config?mode=defaults: Shows the YAML configuration with all the default values.
• [ENHANCEMENT] OpenStack Swift: added the following config options to OpenStack Swift backend client: #3660
  • Chunks storage: -swift.auth-version, -swift.max-retries, -swift.connect-timeout, -swift.request-timeout.
  • Blocks storage: -blocks-storage.swift.auth-version, -blocks-storage.swift.max-retries, -blocks-storage.swift.connect-timeout, -blocks-storage.swift.request-timeout.
  • Ruler: -ruler.storage.swift.auth-version, -ruler.storage.swift.max-retries, -ruler.storage.swift.connect-timeout, -ruler.storage.swift.request-timeout.
• [ENHANCEMENT] Disabled in-memory shuffle-sharding subring cache in the store-gateway, ruler and compactor. This should reduce the memory utilisation in these services when shuffle-sharding is enabled, without introducing a significantly increase CPU utilisation. #3601
• [ENHANCEMENT] Shuffle sharding: optimised subring generation used by shuffle sharding. #3601
• [ENHANCEMENT] New /runtime_config endpoint that returns the defined runtime configuration in YAML format. The returned configuration includes overrides. #3639
• [ENHANCEMENT] Query-frontend: included the parameter name failed to validate in HTTP 400 message. #3703
• [ENHANCEMENT] Fail to startup Cortex if provided runtime config is invalid. #3707
• [ENHANCEMENT] Alertmanager: Add flags to customize the cluster configuration: #3667
  • -alertmanager.cluster.gossip-interval: The interval between sending gossip messages. By lowering this value (more frequent) gossip messages are propagated across cluster more quickly at the expense of increased bandwidth usage.
  • -alertmanager.cluster.push-pull-interval: The interval between gossip state syncs. Setting this interval lower (more frequent) will increase convergence speeds across larger clusters at the expense of increased bandwidth usage.
• [ENHANCEMENT] Distributor: change the error message returned when a received series has too many label values. The new message format has the series at the end and this plays better with Prometheus logs truncation. #3718
  • From: sample for '<series>' has <value> label names; limit <value>
  • To: series has too many labels (actual: <value>, limit: <value>) series: '<series>'
• [ENHANCEMENT] Improve bucket index loader to...

Cortex 1.7.0-rc.2

• [BUGFIX] Handle missing samples due to large steps and single point extents. #3818 #3835

Cortex 1.7.0-rc.1

Changelog

• [BUGFIX] Fix ring tokens sorting regression introduced in #3601. #3815

Cortex 1.7.0-rc.0

Changelog

Cortex

• [CHANGE] FramedSnappy encoding support has been removed from Push and Remote Read APIs. This means Prometheus 1.6 support has been removed and the oldest Prometheus version supported in the remote write is 1.7. #3682
• [CHANGE] Ruler: removed the flag -ruler.evaluation-delay-duration-deprecated which was deprecated in 1.4.0. Please use the ruler_evaluation_delay_duration per-tenant limit instead. #3694
• [CHANGE] Removed the flags -<prefix>.grpc-use-gzip-compression which were deprecated in 1.3.0: #3694
  • -query-scheduler.grpc-client-config.grpc-use-gzip-compression: use -query-scheduler.grpc-client-config.grpc-compression instead
  • -frontend.grpc-client-config.grpc-use-gzip-compression: use -frontend.grpc-client-config.grpc-compression instead
  • -ruler.client.grpc-use-gzip-compression: use -ruler.client.grpc-compression instead
  • -bigtable.grpc-use-gzip-compression: use -bigtable.grpc-compression instead
  • -ingester.client.grpc-use-gzip-compression: use -ingester.client.grpc-compression instead
  • -querier.frontend-client.grpc-use-gzip-compression: use -querier.frontend-client.grpc-compression instead
• [CHANGE] Querier: it's not required to set -frontend.query-stats-enabled=true in the querier anymore to enable query statistics logging in the query-frontend. The flag is now required to be configured only in the query-frontend and it will be propagated to the queriers. #3595 #3695
• [CHANGE] Blocks storage: compactor is now required when running a Cortex cluster with the blocks storage, because it also keeps the bucket index updated. #3583
• [CHANGE] Blocks storage: block deletion marks are now stored in a per-tenant global markers/ location too, other than within the block location. The compactor, at startup, will copy deletion marks from the block location to the global location. This migration is required only once, so you can safely disable it via -compactor.block-deletion-marks-migration-enabled=false once new compactor has successfully started once in your cluster. #3583
• [CHANGE] OpenStack Swift: the default value for the -ruler.storage.swift.container-name and -swift.container-name config options has changed from cortex to empty string. If you were relying on the default value, you should set it back to cortex. #3660
• [CHANGE] HA Tracker: configured replica label is now verified against label value length limit (-validation.max-length-label-value). #3668
• [CHANGE] Distributor: extend_writes field in YAML configuration has moved from lifecycler (inside ingester_config) to distributor_config. This doesn't affect command line option -distributor.extend-writes, which stays the same. #3719
• [CHANGE] Alertmanager: Deprecated -cluster. CLI flags in favor of their -alertmanager.cluster. equivalent. The deprecated flags (and their respective YAML config options) are: #3677
  • -cluster.listen-address in favor of -alertmanager.cluster.listen-address
  • -cluster.advertise-address in favor of -alertmanager.cluster.advertise-address
  • -cluster.peer in favor of -alertmanager.cluster.peers
  • -cluster.peer-timeout in favor of -alertmanager.cluster.peer-timeout
• [CHANGE] Blocks storage: the default value of -blocks-storage.bucket-store.sync-interval has been changed from 5m to 15m. #3724
• [FEATURE] Querier: Queries can be federated across multiple tenants. The tenants IDs involved need to be specified separated by a | character in the X-Scope-OrgID request header. This is an experimental feature, which can be enabled by setting -tenant-federation.enabled=true on all Cortex services. #3250
• [FEATURE] Alertmanager: introduced the experimental option -alertmanager.sharding-enabled to shard tenants across multiple Alertmanager instances. This feature is still under heavy development and its usage is discouraged. The following new metrics are exported by the Alertmanager: #3664
  • cortex_alertmanager_ring_check_errors_total
  • cortex_alertmanager_sync_configs_total
  • cortex_alertmanager_sync_configs_failed_total
  • cortex_alertmanager_tenants_discovered
  • cortex_alertmanager_tenants_owned
• [ENHANCEMENT] Allow specifying JAEGER_ENDPOINT instead of sampling server or local agent port. #3682
• [ENHANCEMENT] Blocks storage: introduced a per-tenant bucket index, periodically updated by the compactor, used to avoid full bucket scanning done by queriers, store-gateways and rulers. The bucket index is updated by the compactor during blocks cleanup, on every -compactor.cleanup-interval. #3553 #3555 #3561 #3583 #3625 #3711 #3715
• [ENHANCEMENT] Blocks storage: introduced an option -blocks-storage.bucket-store.bucket-index.enabled to enable the usage of the bucket index in the querier, store-gateway and ruler. When enabled, the querier, store-gateway and ruler will use the bucket index to find a tenant's blocks instead of running the periodic bucket scan. The following new metrics are exported by the querier and ruler: #3614 #3625
  • cortex_bucket_index_loads_total
  • cortex_bucket_index_load_failures_total
  • cortex_bucket_index_load_duration_seconds
  • cortex_bucket_index_loaded
• [ENHANCEMENT] Compactor: exported the following metrics. #3583 #3625
  • cortex_bucket_blocks_count: Total number of blocks per tenant in the bucket. Includes blocks marked for deletion, but not partial blocks.
  • cortex_bucket_blocks_marked_for_deletion_count: Total number of blocks per tenant marked for deletion in the bucket.
  • cortex_bucket_blocks_partials_count: Total number of partial blocks.
  • cortex_bucket_index_last_successful_update_timestamp_seconds: Timestamp of the last successful update of a tenant's bucket index.
• [ENHANCEMENT] Ruler: Add cortex_prometheus_last_evaluation_samples to expose the number of samples generated by a rule group per tenant. #3582
• [ENHANCEMENT] Memberlist: add status page (/memberlist) with available details about memberlist-based KV store and memberlist cluster. It's also possible to view KV values in Go struct or JSON format, or download for inspection. #3575
• [ENHANCEMENT] Memberlist: client can now keep a size-bounded buffer with sent and received messages and display them in the admin UI (/memberlist) for troubleshooting. #3581 #3602
• [ENHANCEMENT] Blocks storage: added block index attributes caching support to metadata cache. The TTL can be configured via -blocks-storage.bucket-store.metadata-cache.block-index-attributes-ttl. #3629
• [ENHANCEMENT] Alertmanager: Add support for Azure blob storage. #3634
• [ENHANCEMENT] Compactor: tenants marked for deletion will now be fully cleaned up after some delay since deletion of last block. Cleanup includes removal of remaining marker files (including tenant deletion mark file) and files under debug/metas. #3613
• [ENHANCEMENT] Compactor: retry compaction of a single tenant on failure instead of re-running compaction for all tenants. #3627
• [ENHANCEMENT] Querier: Implement result caching for tenant query federation. #3640
• [ENHANCEMENT] API: Add a mode query parameter for the config endpoint: #3645
  • /config?mode=diff: Shows the YAML configuration with all values that differ from the defaults.
  • /config?mode=defaults: Shows the YAML configuration with all the default values.
• [ENHANCEMENT] OpenStack Swift: added the following config options to OpenStack Swift backend client: #3660
  • Chunks storage: -swift.auth-version, -swift.max-retries, -swift.connect-timeout, -swift.request-timeout.
  • Blocks storage: -blocks-storage.swift.auth-version, -blocks-storage.swift.max-retries, -blocks-storage.swift.connect-timeout, -blocks-storage.swift.request-timeout.
  • Ruler: -ruler.storage.swift.auth-version, -ruler.storage.swift.max-retries, -ruler.storage.swift.connect-timeout, -ruler.storage.swift.request-timeout.
• [ENHANCEMENT] Disabled in-memory shuffle-sharding subring cache in the store-gateway, ruler and compactor. This should reduce the memory utilisation in these services when shuffle-sharding is enabled, without introducing a significantly increase CPU utilisation. #3601
• [ENHANCEMENT] Shuffle sharding: optimised subring generation used by shuffle sharding. #3601
• [ENHANCEMENT] New /runtime_config endpoint that returns the defined runtime configuration in YAML format. The returned configuration includes overrides. #3639
• [ENHANCEMENT] Query-frontend: included the parameter name failed to validate in HTTP 400 message. #3703
• [ENHANCEMENT] Fail to startup Cortex if provided runtime config is invalid. #3707
• [ENHANCEMENT] Alertmanager: Add flags to customize the cluster configuration: #3667
  • -alertmanager.cluster.gossip-interval: The interval between sending gossip messages. By lowering this value (more frequent) gossip messages are propagated across cluster more quickly at the expense of increased bandwidth usage.
  • -alertmanager.cluster.push-pull-interval: The interval between gossip state syncs. Setting this interval lower (more frequent) will increase convergence speeds across larger clusters at the expense of increased bandwidth usage.
• [ENHANCEMENT] Distributor: change the error message returned when a received series has too many label values. The new message format has the series at the end and this plays better with Prometheus logs truncation. #3718
  • From: sample for '<series>' has <value> label names; limit <value>
  • To: series has too many labels (actual: <value>, limit: <value>) series: '<series>'
• [ENHANCEMENT] Improve bucket index loader to handle edge case where new tenant has not had blocks uploaded to storage yet. #3717
• [BUGFIX] Allow -querier.max-query-lookback use y|w|d suffix like deprecated -store.max-look-back-period. #3598
• [BUGFIX] Memberlist: Entry in the ring should now not appear again after using "Forget" ...

Cortex 1.6.0

Changelog

Cortex

• [CHANGE] Query Frontend: deprecate -querier.compress-http-responses in favour of -api.response-compression-enabled. #3544
• [CHANGE] Querier: deprecated -store.max-look-back-period. You should use -querier.max-query-lookback instead. #3452
• [CHANGE] Blocks storage: increased -blocks-storage.bucket-store.chunks-cache.attributes-ttl default from 24h to 168h (1 week). #3528
• [CHANGE] Blocks storage: the config option -blocks-storage.bucket-store.index-cache.postings-compression-enabled has been deprecated and postings compression is always enabled. #3538
• [CHANGE] Ruler: gRPC message size default limits on the Ruler-client side have changed: #3523
  • limit for outgoing gRPC messages has changed from 2147483647 to 16777216 bytes
  • limit for incoming gRPC messages has changed from 4194304 to 104857600 bytes
• [FEATURE] Distributor/Ingester: Provide ability to not overflow writes in the presence of a leaving or unhealthy ingester. This allows for more efficient ingester rolling restarts. #3305
• [FEATURE] Query-frontend: introduced query statistics logged in the query-frontend when enabled via -frontend.query-stats-enabled=true. When enabled, the metric cortex_query_seconds_total is tracked, counting the sum of the wall time spent across all queriers while running queries (on a per-tenant basis). The metrics cortex_request_duration_seconds and cortex_query_seconds_total are different: the first one tracks the request duration (eg. HTTP request from the client), while the latter tracks the sum of the wall time on all queriers involved executing the query. #3539
• [ENHANCEMENT] API: Add GZIP HTTP compression to the API responses. Compression can be enabled via -api.response-compression-enabled. #3536
• [ENHANCEMENT] Added zone-awareness support on queries. When zone-awareness is enabled, queries will still succeed if all ingesters in a single zone will fail. #3414
• [ENHANCEMENT] Blocks storage ingester: exported more TSDB-related metrics. #3412
  • cortex_ingester_tsdb_wal_corruptions_total
  • cortex_ingester_tsdb_head_truncations_failed_total
  • cortex_ingester_tsdb_head_truncations_total
  • cortex_ingester_tsdb_head_gc_duration_seconds
• [ENHANCEMENT] Enforced keepalive on all gRPC clients used for inter-service communication. #3431
• [ENHANCEMENT] Added cortex_alertmanager_config_hash metric to expose hash of Alertmanager Config loaded per user. #3388
• [ENHANCEMENT] Query-Frontend / Query-Scheduler: New component called "Query-Scheduler" has been introduced. Query-Scheduler is simply a queue of requests, moved outside of Query-Frontend. This allows Query-Frontend to be scaled separately from number of queues. To make Query-Frontend and Querier use Query-Scheduler, they need to be started with -frontend.scheduler-address and -querier.scheduler-address options respectively. #3374 #3471
• [ENHANCEMENT] Query-frontend / Querier / Ruler: added -querier.max-query-lookback to limit how long back data (series and metadata) can be queried. This setting can be overridden on a per-tenant basis and is enforced in the query-frontend, querier and ruler. #3452 #3458
• [ENHANCEMENT] Querier: added -querier.query-store-for-labels-enabled to query store for label names, label values and series APIs. Only works with blocks storage engine. #3461 #3520
• [ENHANCEMENT] Ingester: exposed -blocks-storage.tsdb.wal-segment-size-bytes config option to customise the TSDB WAL segment max size. #3476
• [ENHANCEMENT] Compactor: concurrently run blocks cleaner for multiple tenants. Concurrency can be configured via -compactor.cleanup-concurrency. #3483
• [ENHANCEMENT] Compactor: shuffle tenants before running compaction. #3483
• [ENHANCEMENT] Compactor: wait for a stable ring at startup, when sharding is enabled. #3484
• [ENHANCEMENT] Store-gateway: added -blocks-storage.bucket-store.index-header-lazy-loading-enabled to enable index-header lazy loading (experimental). When enabled, index-headers will be mmap-ed only once required by a query and will be automatically released after -blocks-storage.bucket-store.index-header-lazy-loading-idle-timeout time of inactivity. #3498
• [ENHANCEMENT] Alertmanager: added metrics cortex_alertmanager_notification_requests_total and cortex_alertmanager_notification_requests_failed_total. #3518
• [ENHANCEMENT] Ingester: added -blocks-storage.tsdb.head-chunks-write-buffer-size-bytes to fine-tune the TSDB head chunks write buffer size when running Cortex blocks storage. #3518
• [ENHANCEMENT] /metrics now supports OpenMetrics output. HTTP and gRPC servers metrics can now include exemplars. #3524
• [ENHANCEMENT] Expose gRPC keepalive policy options by gRPC server. #3524
• [ENHANCEMENT] Blocks storage: enabled caching of meta.json attributes, configurable via -blocks-storage.bucket-store.metadata-cache.metafile-attributes-ttl. #3528
• [ENHANCEMENT] Compactor: added a config validation check to fail fast if the compactor has been configured invalid block range periods (each period is expected to be a multiple of the previous one). #3534
• [ENHANCEMENT] Blocks storage: concurrently fetch deletion marks from object storage. #3538
• [ENHANCEMENT] Blocks storage ingester: ingester can now close idle TSDB and delete local data. #3491 #3552
• [ENHANCEMENT] Blocks storage: add option to use V2 signatures for S3 authentication. #3540
• [ENHANCEMENT] Exported process metrics to monitor the number of memory map areas allocated. #3537
  • • process_memory_map_areas
  • • process_memory_map_areas_limit
• [ENHANCEMENT] Ruler: Expose gRPC client options. #3523
• [ENHANCEMENT] Compactor: added metrics to track on-going compaction. #3535
  • cortex_compactor_tenants_discovered
  • cortex_compactor_tenants_skipped
  • cortex_compactor_tenants_processing_succeeded
  • cortex_compactor_tenants_processing_failed
• [ENHANCEMENT] Added new experimental API endpoints: POST /purger/delete_tenant and GET /purger/delete_tenant_status for deleting all tenant data. Only works with blocks storage. Compactor removes blocks that belong to user marked for deletion. #3549 #3558
• [ENHANCEMENT] Chunks storage: add option to use V2 signatures for S3 authentication. #3560
• [BUGFIX] Query-Frontend: cortex_query_seconds_total now return seconds not nanoseconds. #3589
• [BUGFIX] Blocks storage ingester: fixed some cases leading to a TSDB WAL corruption after a partial write to disk. #3423
• [BUGFIX] Blocks storage: Fix the race between ingestion and /flush call resulting in overlapping blocks. #3422
• [BUGFIX] Querier: fixed -querier.max-query-into-future which wasn't correctly enforced on range queries. #3452
• [BUGFIX] Fixed float64 precision stability when aggregating metrics before exposing them. This could have lead to false counters resets when querying some metrics exposed by Cortex. #3506
• [BUGFIX] Querier: the meta.json sync concurrency done when running Cortex with the blocks storage is now controlled by -blocks-storage.bucket-store.meta-sync-concurrency instead of the incorrect -blocks-storage.bucket-store.block-sync-concurrency (default values are the same). #3531
• [BUGFIX] Querier: fixed initialization order of querier module when using blocks storage. It now (again) waits until blocks have been synchronized. #3551

Blocksconvert

• [ENHANCEMENT] Scheduler: ability to ignore users based on regexp, using -scheduler.ignore-users-regex flag. #3477
• [ENHANCEMENT] Builder: Parallelize reading chunks in the final stage of building block. #3470
• [ENHANCEMENT] Builder: remove duplicate label names from chunk. #3547

Cortex 1.6.0-rc.1

Changelog

Cortex

• [BUGFIX] Query-Frontend: cortex_query_seconds_total now return seconds not nanoseconds. #3589

Cortex 1.6.0-rc.0

Changelog

Cortex

• [CHANGE] Query Frontend: deprecate -querier.compress-http-responses in favour of -api.response-compression-enabled. #3544
• [CHANGE] Querier: deprecated -store.max-look-back-period. You should use -querier.max-query-lookback instead. #3452
• [CHANGE] Blocks storage: increased -blocks-storage.bucket-store.chunks-cache.attributes-ttl default from 24h to 168h (1 week). #3528
• [CHANGE] Blocks storage: the config option -blocks-storage.bucket-store.index-cache.postings-compression-enabled has been deprecated and postings compression is always enabled. #3538
• [CHANGE] Ruler: gRPC message size default limits on the Ruler-client side have changed: #3523
  • limit for outgoing gRPC messages has changed from 2147483647 to 16777216 bytes
  • limit for incoming gRPC messages has changed from 4194304 to 104857600 bytes
• [FEATURE] Distributor/Ingester: Provide ability to not overflow writes in the presence of a leaving or unhealthy ingester. This allows for more efficient ingester rolling restarts. #3305
• [FEATURE] Query-frontend: introduced query statistics logged in the query-frontend when enabled via -frontend.query-stats-enabled=true. When enabled, the metric cortex_query_seconds_total is tracked, counting the sum of the wall time spent across all queriers while running queries (on a per-tenant basis). The metrics cortex_request_duration_seconds and cortex_query_seconds_total are different: the first one tracks the request duration (eg. HTTP request from the client), while the latter tracks the sum of the wall time on all queriers involved executing the query. #3539
• [ENHANCEMENT] API: Add GZIP HTTP compression to the API responses. Compression can be enabled via -api.response-compression-enabled. #3536
• [ENHANCEMENT] Added zone-awareness support on queries. When zone-awareness is enabled, queries will still succeed if all ingesters in a single zone will fail. #3414
• [ENHANCEMENT] Blocks storage ingester: exported more TSDB-related metrics. #3412
  • cortex_ingester_tsdb_wal_corruptions_total
  • cortex_ingester_tsdb_head_truncations_failed_total
  • cortex_ingester_tsdb_head_truncations_total
  • cortex_ingester_tsdb_head_gc_duration_seconds
• [ENHANCEMENT] Enforced keepalive on all gRPC clients used for inter-service communication. #3431
• [ENHANCEMENT] Added cortex_alertmanager_config_hash metric to expose hash of Alertmanager Config loaded per user. #3388
• [ENHANCEMENT] Query-Frontend / Query-Scheduler: New component called "Query-Scheduler" has been introduced. Query-Scheduler is simply a queue of requests, moved outside of Query-Frontend. This allows Query-Frontend to be scaled separately from number of queues. To make Query-Frontend and Querier use Query-Scheduler, they need to be started with -frontend.scheduler-address and -querier.scheduler-address options respectively. #3374 #3471
• [ENHANCEMENT] Query-frontend / Querier / Ruler: added -querier.max-query-lookback to limit how long back data (series and metadata) can be queried. This setting can be overridden on a per-tenant basis and is enforced in the query-frontend, querier and ruler. #3452 #3458
• [ENHANCEMENT] Querier: added -querier.query-store-for-labels-enabled to query store for label names, label values and series APIs. Only works with blocks storage engine. #3461 #3520
• [ENHANCEMENT] Ingester: exposed -blocks-storage.tsdb.wal-segment-size-bytes config option to customise the TSDB WAL segment max size. #3476
• [ENHANCEMENT] Compactor: concurrently run blocks cleaner for multiple tenants. Concurrency can be configured via -compactor.cleanup-concurrency. #3483
• [ENHANCEMENT] Compactor: shuffle tenants before running compaction. #3483
• [ENHANCEMENT] Compactor: wait for a stable ring at startup, when sharding is enabled. #3484
• [ENHANCEMENT] Store-gateway: added -blocks-storage.bucket-store.index-header-lazy-loading-enabled to enable index-header lazy loading (experimental). When enabled, index-headers will be mmap-ed only once required by a query and will be automatically released after -blocks-storage.bucket-store.index-header-lazy-loading-idle-timeout time of inactivity. #3498
• [ENHANCEMENT] Alertmanager: added metrics cortex_alertmanager_notification_requests_total and cortex_alertmanager_notification_requests_failed_total. #3518
• [ENHANCEMENT] Ingester: added -blocks-storage.tsdb.head-chunks-write-buffer-size-bytes to fine-tune the TSDB head chunks write buffer size when running Cortex blocks storage. #3518
• [ENHANCEMENT] /metrics now supports OpenMetrics output. HTTP and gRPC servers metrics can now include exemplars. #3524
• [ENHANCEMENT] Expose gRPC keepalive policy options by gRPC server. #3524
• [ENHANCEMENT] Blocks storage: enabled caching of meta.json attributes, configurable via -blocks-storage.bucket-store.metadata-cache.metafile-attributes-ttl. #3528
• [ENHANCEMENT] Compactor: added a config validation check to fail fast if the compactor has been configured invalid block range periods (each period is expected to be a multiple of the previous one). #3534
• [ENHANCEMENT] Blocks storage: concurrently fetch deletion marks from object storage. #3538
• [ENHANCEMENT] Blocks storage ingester: ingester can now close idle TSDB and delete local data. #3491 #3552
• [ENHANCEMENT] Blocks storage: add option to use V2 signatures for S3 authentication. #3540
• [ENHANCEMENT] Exported process metrics to monitor the number of memory map areas allocated. #3537
  • • process_memory_map_areas
  • • process_memory_map_areas_limit
• [ENHANCEMENT] Ruler: Expose gRPC client options. #3523
• [ENHANCEMENT] Compactor: added metrics to track on-going compaction. #3535
  • cortex_compactor_tenants_discovered
  • cortex_compactor_tenants_skipped
  • cortex_compactor_tenants_processing_succeeded
  • cortex_compactor_tenants_processing_failed
• [ENHANCEMENT] Added new experimental API endpoints: POST /purger/delete_tenant and GET /purger/delete_tenant_status for deleting all tenant data. Only works with blocks storage. Compactor removes blocks that belong to user marked for deletion. #3549 #3558
• [ENHANCEMENT] Chunks storage: add option to use V2 signatures for S3 authentication. #3560
• [BUGFIX] Blocks storage ingester: fixed some cases leading to a TSDB WAL corruption after a partial write to disk. #3423
• [BUGFIX] Blocks storage: Fix the race between ingestion and /flush call resulting in overlapping blocks. #3422
• [BUGFIX] Querier: fixed -querier.max-query-into-future which wasn't correctly enforced on range queries. #3452
• [BUGFIX] Fixed float64 precision stability when aggregating metrics before exposing them. This could have lead to false counters resets when querying some metrics exposed by Cortex. #3506
• [BUGFIX] Querier: the meta.json sync concurrency done when running Cortex with the blocks storage is now controlled by -blocks-storage.bucket-store.meta-sync-concurrency instead of the incorrect -blocks-storage.bucket-store.block-sync-concurrency (default values are the same). #3531
• [BUGFIX] Querier: fixed initialization order of querier module when using blocks storage. It now (again) waits until blocks have been synchronized. #3551

Blocksconvert

• [ENHANCEMENT] Scheduler: ability to ignore users based on regexp, using -scheduler.ignore-users-regex flag. #3477
• [ENHANCEMENT] Builder: Parallelize reading chunks in the final stage of building block. #3470
• [ENHANCEMENT] Builder: remove duplicate label names from chunk. #3547

Cortex 1.5.0

Changelog

Cortex

• [CHANGE] Blocks storage: update the default HTTP configuration values for the S3 client to the upstream Thanos default values. #3244
  • -blocks-storage.s3.http.idle-conn-timeout is set 90 seconds.
  • -blocks-storage.s3.http.response-header-timeout is set to 2 minutes.
• [CHANGE] Improved shuffle sharding support in the write path. This work introduced some config changes: #3090
  • Introduced -distributor.sharding-strategy CLI flag (and its respective sharding_strategy YAML config option) to explicitly specify which sharding strategy should be used in the write path
  • -experimental.distributor.user-subring-size flag renamed to -distributor.ingestion-tenant-shard-size
  • user_subring_size limit YAML config option renamed to ingestion_tenant_shard_size
• [CHANGE] Dropped "blank Alertmanager configuration; using fallback" message from Info to Debug level. #3205
• [CHANGE] Zone-awareness replication for time-series now should be explicitly enabled in the distributor via the -distributor.zone-awareness-enabled CLI flag (or its respective YAML config option). Before, zone-aware replication was implicitly enabled if a zone was set on ingesters. #3200
• [CHANGE] Removed the deprecated CLI flag -config-yaml. You should use -schema-config-file instead. #3225
• [CHANGE] Enforced the HTTP method required by some API endpoints which did (incorrectly) allow any method before that. #3228
  • GET /
  • GET /config
  • GET /debug/fgprof
  • GET /distributor/all_user_stats
  • GET /distributor/ha_tracker
  • GET /all_user_stats
  • GET /ha-tracker
  • GET /api/v1/user_stats
  • GET /api/v1/chunks
  • GET <legacy-http-prefix>/user_stats
  • GET <legacy-http-prefix>/chunks
  • GET /services
  • GET /multitenant_alertmanager/status
  • GET /status (alertmanager microservice)
  • GET|POST /ingester/ring
  • GET|POST /ring
  • GET|POST /store-gateway/ring
  • GET|POST /compactor/ring
  • GET|POST /ingester/flush
  • GET|POST /ingester/shutdown
  • GET|POST /flush
  • GET|POST /shutdown
  • GET|POST /ruler/ring
  • POST /api/v1/push
  • POST <legacy-http-prefix>/push
  • POST /push
  • POST /ingester/push
• [CHANGE] Renamed CLI flags to configure the network interface names from which automatically detect the instance IP. #3295
  • -compactor.ring.instance-interface renamed to -compactor.ring.instance-interface-names
  • -store-gateway.sharding-ring.instance-interface renamed to -store-gateway.sharding-ring.instance-interface-names
  • -distributor.ring.instance-interface renamed to -distributor.ring.instance-interface-names
  • -ruler.ring.instance-interface renamed to -ruler.ring.instance-interface-names
• [CHANGE] Renamed -<prefix>.redis.enable-tls CLI flag to -<prefix>.redis.tls-enabled, and its respective YAML config option from enable_tls to tls_enabled. #3298
• [CHANGE] Increased default -<prefix>.redis.timeout from 100ms to 500ms. #3301
• [CHANGE] cortex_alertmanager_config_invalid has been removed in favor of cortex_alertmanager_config_last_reload_successful. #3289
• [CHANGE] Query-frontend: POST requests whose body size exceeds 10MiB will be rejected. The max body size can be customised via -frontend.max-body-size. #3276
• [FEATURE] Shuffle sharding: added support for shuffle-sharding queriers in the query-frontend. When configured (-frontend.max-queriers-per-tenant globally, or using per-tenant limit max_queriers_per_tenant), each tenants's requests will be handled by different set of queriers. #3113 #3257
• [FEATURE] Shuffle sharding: added support for shuffle-sharding ingesters on the read path. When ingesters shuffle-sharding is enabled and -querier.shuffle-sharding-ingesters-lookback-period is set, queriers will fetch in-memory series from the minimum set of required ingesters, selecting only ingesters which may have received series since 'now - lookback period'. #3252
• [FEATURE] Query-frontend: added compression config to support results cache with compression. #3217
• [FEATURE] Add OpenStack Swift support to blocks storage. #3303
• [FEATURE] Added support for applying Prometheus relabel configs on series received by the distributor. A metric_relabel_configs field has been added to the per-tenant limits configuration. #3329
• [FEATURE] Support for Cassandra client SSL certificates. #3384
• [ENHANCEMENT] Ruler: Introduces two new limits -ruler.max-rules-per-rule-group and -ruler.max-rule-groups-per-tenant to control the number of rules per rule group and the total number of rule groups for a given user. They are disabled by default. #3366
• [ENHANCEMENT] Allow to specify multiple comma-separated Cortex services to -target CLI option (or its respective YAML config option). For example, -target=all,compactor can be used to start Cortex single-binary with compactor as well. #3275
• [ENHANCEMENT] Expose additional HTTP configs for the S3 backend client. New flag are listed below: #3244
  • -blocks-storage.s3.http.idle-conn-timeout
  • -blocks-storage.s3.http.response-header-timeout
  • -blocks-storage.s3.http.insecure-skip-verify
• [ENHANCEMENT] Added cortex_query_frontend_connected_clients metric to show the number of workers currently connected to the frontend. #3207
• [ENHANCEMENT] Shuffle sharding: improved shuffle sharding in the write path. Shuffle sharding now should be explicitly enabled via -distributor.sharding-strategy CLI flag (or its respective YAML config option) and guarantees stability, consistency, shuffling and balanced zone-awareness properties. #3090 #3214
• [ENHANCEMENT] Ingester: added new metric cortex_ingester_active_series to track active series more accurately. Also added options to control whether active series tracking is enabled (-ingester.active-series-enabled, defaults to false), and how often this metric is updated (-ingester.active-series-update-period) and max idle time for series to be considered inactive (-ingester.active-series-idle-timeout). #3153
• [ENHANCEMENT] Store-gateway: added zone-aware replication support to blocks replication in the store-gateway. #3200
• [ENHANCEMENT] Store-gateway: exported new metrics. #3231
  • cortex_bucket_store_cached_series_fetch_duration_seconds
  • cortex_bucket_store_cached_postings_fetch_duration_seconds
  • cortex_bucket_stores_gate_queries_max
• [ENHANCEMENT] Added -version flag to Cortex. #3233
• [ENHANCEMENT] Hash ring: added instance registered timestamp to the ring. #3248
• [ENHANCEMENT] Reduce tail latency by smoothing out spikes in rate of chunk flush operations. #3191
• [ENHANCEMENT] User Cortex as User Agent in http requests issued by Configs DB client. #3264
• [ENHANCEMENT] Experimental Ruler API: Fetch rule groups from object storage in parallel. #3218
• [ENHANCEMENT] Chunks GCS object storage client uses the fields selector to limit the payload size when listing objects in the bucket. #3218 #3292
• [ENHANCEMENT] Added shuffle sharding support to ruler. Added new metric cortex_ruler_sync_rules_total. #3235
• [ENHANCEMENT] Return an explicit error when the store-gateway is explicitly requested without a blocks storage engine. #3287
• [ENHANCEMENT] Ruler: only load rules that belong to the ruler. Improves rules synching performances when ruler sharding is enabled. #3269
• [ENHANCEMENT] Added -<prefix>.redis.tls-insecure-skip-verify flag. #3298
• [ENHANCEMENT] Added cortex_alertmanager_config_last_reload_successful_seconds metric to show timestamp of last successful AM config reload. #3289
• [ENHANCEMENT] Blocks storage: reduced number of bucket listing operations to list block content (applies to newly created blocks only). #3363
• [ENHANCEMENT] Ruler: Include the tenant ID on the notifier logs. #3372
• [ENHANCEMENT] Blocks storage Compactor: Added -compactor.enabled-tenants and -compactor.disabled-tenants to explicitly enable or disable compaction of specific tenants. #3385
• [ENHANCEMENT] Blocks storage ingester: Creating checkpoint only once even when there are multiple Head compactions in a single Compact() call. #3373
• [BUGFIX] Blocks storage ingester: Read repair memory-mapped chunks file which can end up being empty on abrupt shutdowns combined with faulty disks. #3373
• [BUGFIX] Blocks storage ingester: Close TSDB resources on failed startup preventing ingester OOMing. #3373
• [BUGFIX] No-longer-needed ingester operations for queries triggered by queriers and rulers are now canceled. #3178
• [BUGFIX] Ruler: directories in the configured rules-path will be removed on startup and shutdown in order to ensure they don't persist between runs. #3195
• [BUGFIX] Handle hash-collisions in the query path. #3192
• [BUGFIX] Check for postgres rows errors. #3197
• [BUGFIX] Ruler Experimental API: Don't allow rule groups without names or empty rule groups. #3210
• [BUGFIX] Experimental Alertmanager API: Do not allow empty Alertmanager configurations or bad template filenames to be submitted through the configuration API. #3185
• [BUGFIX] Reduce failures to update heartbeat when using Consul. #3259
• [BUGFIX] When using ruler sharding, moving all user rule groups from ruler to a different one and then back could end up with some user groups not being evaluated at all. #3235
• [BUGFIX] Fixed shuffle sharding consistency when zone-awareness is enabled and the shard size is increased or instances in a new zone are added. #3299
• [BUGFIX] Use a valid grpc header when logging IP addresses. #3307
• [BUGFIX] Fixed the metric cortex_prometheus_rule_group_duration_seconds in the Ruler, it wouldn't report any values. #3310
• [BUGFIX] Fixed gRPC connections leaking in rulers when rulers sharding is enabled and APIs called. #3314
• [BUGFIX] Fixed shuffle sharding consistency when zone-awareness is enabled and the shard size is increased or instances...

Cortex 1.5.0-rc.1

Changelog

• [BUGFIX] Fixed the route label on cortex_querier_.* metrics. #3436
• [BUGFIX] Fixed query-frontend and querier API authentication bypass when Cortex is running with authentication disabled. #3439