v1.17.4

Release Highlights

This release includes bug fixes for Archival and Retry logic, and removes 30 days max Retention Period limit.

IMPORTANT NOTE FOR ARCHIVAL USERS
If you are using the Archival feature and if v1.17.4 of the History Service runs with an older version of the Worker Service during the upgrade, it is possible to have a "dangling Workflow mutable state" where the Event History has been deleted.
To avoid that when upgrading to this release, perform the following steps:

1. Stop the old version of the Worker Service.
2. Deploy this patched release.
3. Start the Worker Service after the upgrade.

All changes

2022-08-16 - 7f72106 - Prepare 1.17.3 patch
2022-08-16 - 06d641f - Fix formating
2022-08-16 - 4426702 - Fix failover queue shutdown (#3232)
2022-08-15 - 0bd7435 - Fix archival activities error handling (#3227)
2022-08-12 - f8a59f7 - Use detached context for shard context operations (#3194)
2022-08-12 - dae8a51 - Fix queue processor throttling logic (#3195)
2022-08-12 - cea0709 - Remove namespace max retention (#3148)
2022-08-12 - cdb9ae0 - Add version check for delete workflow transfer task (#3159)
2022-08-12 - c61450f - Fix inline history archival (#3216)
2022-08-12 - ba452a2 - additional history scavenger heartbeat (#3186)
2022-08-12 - 5b05741 - Simplify system retry logic: Part 1 (#3172)
2022-08-12 - 51817aa - Use a prefix for scheduler workflow ids (#3201)
2022-08-12 - 4b7507d - Improve task processing retry behavior (#3218)
2022-08-12 - 23e40a9 - Fix shard close error type (#3215)
2022-08-12 - bd22047 - Run one per-ns worker per namespace instead of namespace × component (#3116)
2022-08-12 - 6aaddf3 - Run task rescheduler in background (#3038)

Details about v1.17.0 release can be found here.

Helpful links to get you started with Temporal

Temporal Docs
Server
Docker Compose
Helm Chart

Docker images for this release (use tag 1.17.4)

Server
Server With Auto Setup (what is Auto-Setup?)
Admin-Tools

v1.17.2

Release Highlights

This release includes fixes for metrics, rate limiters and Elasticsearch based visibility performance optimizations.

All changes

2022-08-02 - a39ba31 - Prepare 1.17.2 patch
2022-08-02 - 3b8f5c9 - Fix tally timer metric (#3173)
2022-08-02 - 26af874 - Cherry Pick: Adds ability to specify sanitizeOptions for prometheus metrics (#3170) (#3174)
2022-07-29 - 7d9b262 - Core persistence priority rate limiting (#3139)
2022-07-29 - d186259 - Don't dispatch expired tasks from taskReader buffer (#3161)
2022-07-29 - 28cd79b - Remove RunId from visibility query sort for ES 7 (#3140)
2022-07-29 - 0b28e9a - Consolidate rpc client metrics logic (#3147)
2022-07-29 - 478582c - Generate most of the gRPC wrappers in client (#3001)
2022-07-29 - efc3203 - Add config to disable ORDER BY clause (#3128)
2022-07-29 - 4a5f6ed - Use default consistency level for visibility DeleteWorkflowExecution (#3132)
2022-07-29 - b48cf2a - Fix timer metric bucket (#3136)
2022-07-29 - 84149d3 - Fix task rate limiting (#3137)
2022-07-29 - 4d7772d - Fix queue rate limiting busy loop (#3111)
2022-07-28 - a1d1650 - Config worker use public frontend endpoint (#3162)
2022-07-28 - 8dea880 - Update shard replication timestamp with max generated task (#3158)

Details about v1.17.0 release can be found here.

Helpful links to get you started with Temporal

Temporal Docs
Server
Docker Compose
Helm Chart

Docker images for this release (use tag 1.17.2)

Server
Server With Auto Setup (what is Auto-Setup?)
Admin-Tools

v1.17.1

Release Highlights

This release includes fixes for replication stack, delete namespace functionality, and potential deadlock.

All changes

2022-07-11 - e88aa24 - Prepare 1.17.1 patch
2022-07-11 - 13cd5a5 - Reorder frontend rate limiter (#3080)
2022-07-08 - 0b9f15c - Resource aware operation retry (#3069)
2022-07-08 - 56c5295 - Do not double execute close execution task if visibility archival is disabled (#3072)
2022-07-08 - 646f12d - Fix namespace failover callback (#3066)
2022-07-08 - 6737693 - Handle workflow already completed when verifying first workflow task scheduled (#3061)
2022-07-08 - 12c173c - Fix schedule jitter calculation bug (#3059)
2022-07-08 - a95c2fc - Fix error shadowing in standby executors (#3053)
2022-07-08 - 0b01a72 - Improve error handling for standby transfer task verification (#3050)
2022-07-08 - 3278111 - Clean up history branch ancestor operation (#3046)
2022-07-08 - fb208d5 - Cancel activity when eager execution and request cancellation are in the same WFT (#3029)
2022-07-08 - 6513af8 - Release shard lock earlier during delete workflow execution (#3028)
2022-07-08 - 4657f0f - Don't check delete workflow execution transfer task version (#3021)
2022-07-06 - 972af2b - Do not hold lock when running namespace change callback (#3067)
2022-06-30 - 0f13b23 - add support for multiple pragma statements and driver syntax (#3031)
2022-06-30 - c6810d5 - Update backfill history in sync workflow state (#3042)

Details about v1.17.0 release can be found here.

Helpful links to get you started with Temporal

Temporal Docs
Server
Docker Compose
Helm Chart

Docker images for this release (use tag 1.17.1)

Server
Server With Auto Setup (what is Auto-Setup?)
Admin-Tools

v1.16.3

Release Highlights

This release includes fixes for replication stack as well as for shard lock scope.

All changes

2022-07-11 - 83c2b6a - Prepare 1.16.3 patch
2022-07-11 - afd374d - Cherry pick #3080 for 1.16 (#3081)
2022-07-08 - 082d7c9 - Release shard lock earlier during delete workflow execution (#3028)
2022-06-01 - 05a90f8 - Move cluster ack level on ack level update (#2927)
2022-06-01 - 021d501 - Fix dlq retry policy
2022-06-01 - 360062b - Fix failover queue creation after shard reload (#2862)
2022-06-01 - 79102bf - Handle NDC active -> passive transition with transient workflow & no buffered event
2022-06-01 - bd87aa3 - Minor tweak to migration workflow wait replication check (#2839)
2022-06-01 - aad0f85 - Misc changes to migration workflow (#2838)
2022-06-01 - b7562a4 - Check workflow task after reapply events (#2840)
2022-05-20 - 3fff3f0 - Fix record child completion mutable state stale check (#2885)

Details about v1.16.0 release can be found here.

Helpful links to get you started with Temporal

Temporal Docs
Server
Docker Compose
Helm Chart

Docker images for this release (use tag 1.16.3)

Server
Server With Auto Setup (what is Auto-Setup?)
Admin-Tools

v1.17.0

IMPORTANT

This release contains a potential shard lock issue. Do not use this release. Use v1.17.4 instead.

Release highlights

Elasticsearch schema changes

Elasticsearch schema v2 has been added.

The Elasticsearch schema v2 supports new built-in Search Attributes. This schema only adds new mappings, it is not a full reindex.

Use the v2 schema upgrade script schema/elasticsearch/visibility/versioned/v2/upgrade.sh. This can be done before or after upgrading to this release, but it must be done before using the experimental Scheduled Workflow feature.

Metrics

Multiple interfaces in the metrics package are deprecated. The following interfaces are scheduled to be removed in v1.18.0:

• common/metrics/interfaces.go
  • UserScope
  • Scope
  • Client
  • Reporter

Initial implementation to replace these interfaces is now:

• common/metrics/metrics.go
  • MetricsHandler

Histogram metrics use per unit bucket instead of latency unit bucket to emit metrics. In previous releases, the histogram metrics have to x1000 as they are using milliseconds unit. With the per unit bucket update, there is no need to have the x1000 compensation for Dimensionless unit and Bytes unit metrics.

• common/metrics/config.go
  • defaultPerUnitHistogramBoundaries

New metrics service_error_with_type is added for all service errors. It included error_type as tag. The existing metrics service_errors_<FOR_EACH_SPECIFIC_ERROR_TYPE> are deprecated and will be removed from future release. Example of such existing metrics.
New metrics persistence_error_with_type is added for all persistence errors. It included error_type as tag. The existing metrics persistence_errors_<FOR_EACH_SPECIFIC_ERROR_TYPE> are deprecated and will be removed from future release. Example of such existing metrics.

Workflow Execution deletion

Operator Service now exposes the new DeleteWorkflowExecution API, which accepts the name of a Namespace and a Workflow Execution. The Workflow Execution can be in any state. If it is Running, it will be terminated first. The API is exposed via tctl (version >=1.17.0-alpha.2):

$ tctl config set version next
$ tctl workflow delete --workflow-id my_workflow
$ tctl workflow delete --workflow-id my_workflow --run-id my_run_id

If --run-id is not specified, the latest Workflow Execution will be deleted.

Namespace deletion

Operator Service now exposes the new DeleteNamespace API, which accepts the name of a Namespace to delete. Be very careful with this API: it can’t be undone. The API is exposed via tctl:

$ tctl config set version next
$ tctl namespace delete --name my_namespace_name

Namespace deletion is implemented using a system Workflow that is run by a system Worker. It is an asynchronous process, but the Namespace name can be reused as soon as API returns:

1. The Namespace is marked as deleted and renamed to a temporary name. After the DeleteNamespace API returns, the name can be reused (that is, a Namespace with the same name can be created).
2. All Workflow Executions in the deleted Namespace are deleted asynchronously from the database using the DeleteWorkflowExecution API described earlier.
3. The Namespace is deleted from the database.

Please note: this feature is still under active testing.

Host-level priority Task processing

• New Task processing framework in History service for multi-tenancy. This feature is disabled by default in v1.17 and will be enabled in a future release.
• Improved Task retry mechanism to avoid Tasks occupying a Worker goroutine for an extended period of time and blocking the entire queue.
• Configuration for enabling host-level Task Worker pool. Example of dynamic config to enable them:

history.timerProcessorEnablePriorityTaskScheduler:
- constraints: {}
  value: true
history.transferProcessorEnablePriorityTaskScheduler:
- constraints: {}
  value: true
history.visibilityProcessorEnablePriorityTaskScheduler:
- constraints: {}
  value: true

• Configuration for Worker pool size (history.timer/transfer/visibilityProcessorSchedulerWorkerCount) for controlling total traffic sent downstream.
• Configuration for adjusting high-priority Task quota allowed for each Namespace (history.timer/transfer/visibilityTaskHighPriorityRPS).
• Configuration for Task processing round-robin weights (history.timer/transfer/visibilityProcessorSchedulerRoundRobinWeights) among various priorities.
• New task_priority tag for Task processing metrics.

Scheduled Workflows (experimental)

Scheduled Workflows are a new feature that's more flexible than the existing CronSchedule. They're disabled by default in this release, but you can enable them with dynamic config settings. See more docs [here].

SQLite persistence

The SQLite module was replaced with a purego implementation. CGO is no longer required to use SQLite persistence. This means that you can use binaries attached to this release to run local server for development. Just download archive corresponding to your OS and architecture (Mac M1 works!), unzip it, and run server locally without docker-compose and any other dependencies with simple command:

$ ./temporal-server --env development-sqlite start

Helpful links to get you started with Temporal

Temporal Docs
Server
Docker Compose
Helm Chart

Docker images for this release (use tag v1.17.0)

Server
Server With Auto-Setup
what is Auto-Setup?
Admin-Tools

What's Changed

2022-06-17 - 79844fd - use correct scope constructor (#3005)
2022-06-17 - 0b0fea0 - Correct tag exclusion logic and add tests. (#3003)
2022-06-16 - b7262fe - Metrics perf investigation (#3002)
2022-06-15 - 9eb72ea - Rename dt and di variables to workflowTask/wt (#2999)
2022-06-15 - ba0c063 - Update docker-compose.cdc.yml (#2998)
2022-06-15 - 461005f - Remove unneeded duration/seconds conversion (#2994)
2022-06-15 - 49cb949 - Rename scheduleID to scheduledEventID and startedID to startedEventID (#2992)
2022-06-14 - 1dade46 - improve error message (#2937)
2022-06-14 - 2453946 - add default metricoptions where nil (#2993)
2022-06-14 - b3c2810 - Add equality check for task predicate (#2988)
2022-06-14 - 5cfbdc1 - Use metricProvider in task scheduler related components (#2989)
2022-06-13 - c7e9e93 - Update docker-compose.yml (#2984)
2022-06-13 - d4e7632 - mitigate performance issue with userscope (#2983)
2022-06-13 - a09f7ce - Do not retry while holding the shard lock (#2990)
2022-06-13 - d250e0f - Use metricProvider in task executable and rescheduler (#2930)
2022-06-13 - bd33793 - Error with tag metrics (#2987)
2022-06-13 - 946b1ce - Move task range definition to queues package (#2986)
2022-06-12 - 3c5b09f - Add statsd tag separator capability (#2763)
2022-06-10 - b571edf - Multi-cursor components: queue iterator (#2974)
2022-06-09 - 3f302aa - Remove extra namespace checks (#2975)
2022-06-09 - 1e4a76d - Multi-cursor components: queue scope (#2973)
2022-06-08 - 5135a68 - check for nil config and return noop (#2979)
2022-06-08 - 8114a4f - Remove unindexed fields from index schema template (#2976)
2022-06-08 - d16cc46 - Adding serveroption for metrics.MetricHandler (#2978)
2022-06-08 - 49b57b7 - Don't index search attribute used only for passing data through visibility (#2972)
2022-06-08 - f0f1793 - Fix error details bug in deletenamespace workflow (#2954)
2022-06-08 - 9de6b45 - Fix getRemoteClusterInfo race condition (#2971)
2022-06-07 - 8610945 - Update SDK to v1.15.0 (#2964)
2022-06-07 - d3dc17e - Remove caching objects to prevent leak (#2966)
2022-06-06 - e4918d9 - Adding reporter to shard context for access to UserScope (#2961)
2022-06-06 - 81e7a13 - moving handler instantiation to allow noop reporter for autosetup (#2960)
2022-06-06 - b4f2ff5 - also lookup scopedef by common if not found by service (#2959)
2022-06-06 - 48cc8d1 - Revert "Refactor authorization interceptor by extracting utilities (#2945)" (#2958)
2022-06-06 - 14556dd - Fill in fields in ListSchedulesResponse (#2946)
2022-06-06 - ba8ce65 - Multi-cursor queue components: task predicates (#2957)
2022-06-06 - e16e6e8 - Multi-cursor queue components: task key and range (#2956)
2022-06-06 - bef2790 - Add schedule-related search attributes (#2953)
2022-06-06 - 96a0bcd - Tune host level task processing performance (#2955)
2022-06-06 - e5b4e14 - Update docker-compose.yml (#2952)
2022-06-04 - 732d43a - Fix task parallel processor test (#2951)
2022-06-04 - fa2e06c - Update UI readme to v2 (#2936)
2022-06-04 - 8444848 - Fix ...

v1.16.2

Release Highlights

This release includes fixes for:

• Workflow stuck issues during namespace failover
• Missing first workflow task when starting an abandoned child workflow
• SignalWorkflowExecution waking up a workflow in retry backoff period

The release also improves the configuration around namespace cache refresh.

All changes

2022-05-05 - e12a51c - Prepare 1.16.2 patch
2022-05-05 - 705cbe9 - Do not wake up workflow in retry backoff upon signal (#2771)
2022-05-05 - ddb29dc - Expose namespace data (#2804)
2022-05-05 - ae84776 - Always schedule first workflow task for started abandoned child (#2414)
2022-05-05 - 91b099b - Ensure queue processor cluster ack level is below failover ack level (#2794)
2022-05-05 - f62095d - Remove task version and visibility timestamp override (#2789)
2022-05-05 - aa566f7 - Add dynamic config for namespace refresh interval (#2766)

Details about v1.16.0 release can be found here.

Helpful links to get you started with Temporal

Temporal Docs
Server
Docker Compose
Helm Chart

Docker images for this release (use tag 1.16.2)

Server
Server With Auto Setup (what is Auto-Setup?)
Admin-Tools

v1.16.1

Release Highlights

This release includes fixes for duplicated callback listener ID on namespace registry and cluster metadata. It also improves the logic for dropping failed workflow task and logging for history task processing logic.

All changes

2022-04-19 - 1c77018 - Prepare 1.16.1 patch
2022-04-19 - 254a381 - Use caller object as callback listener ID (#2734)
2022-04-19 - 0b27212 - Use uuid as namespace notification ID (#2733)
2022-04-15 - ca6b7ac - Do not drop workflow task if it failed due to unhandled events (#2720)
2022-04-15 - 0d0fc87 - Log NotFound error for transfer task processing (#2723)

Details about v1.16.0 release can be found here.

Helpful links to get you started with Temporal

Temporal Docs
Server
Docker Compose
Helm Chart

Docker images for this release (use tag 1.16.1)

Server
Server With Auto Setup (what is Auto-Setup?)
Admin-Tools

v1.16.0

Release Highlights

Schema changes

Before upgrading to your Temporal Cluster to this release 1.16, you must upgrade your storage schema version to the following:

• MySQL schema version 1.8
• PostgreSQL schema version 1.8
• Cassandra schema version 1.7

Use the Schema Upgrade Tool to upgrade your schema version.
How to use the Schema Upgrade Tool: Upgrade Server version.

Note that schema version is NOT the same as the storage type (database) version, see “Latest SQL database support” below.

Docker Version

• This version (1.16) will be the last version that we build our images based on Alpine 3.13. In the next release (1.17), we are going to upgrade our base image to use Alpine 3.15, which may not be able to run on docker versions lower than 20.10.0. Please check more details here.

Golang version

• Upgraded golang to 1.18

Tctl

• tctl implementation has been moved to a new repository

API

• New Operator service for managing namespace and search attribute
• New DeleteNamespace API in OperatorService. Please note that the delete namespace feature is still under active development and not production ready
• New TERMINATE_IF_RUNNING workflowIdReusePolicy. If a workflow is running using the same workflow ID, this policy will terminate it and start a new one; otherwise, the behavior is the same as ALLOW_DUPLICATE
• RequestCancelWorkflowExecution and RequestCancelExternalWorkflowExecutionCommand now supports reason field. The reason will also be recorded in WorkflowExecutionCancelRequestedEvent and RequestCancelExternalWorkflowExecutionInitiatedEvent
• NonRetryableErrorTypes now supports TemporalTimeout:StartToClose and TemporalTimeout:Heartbeat for customizing retry behavior upon timeout failure
• New last_heartbeat_details field in RespondActivityTaskFailedRequest and RespondActivityTaskFailedByIdRequest
• New activity_failure_include_heartbeat system capability indicating if RespondActivityTaskFailed API supports including heartbeat details
• New ListHistoryTasks API in admin service for listing transfer, timer, replication and visibility tasks
• New RebuildMutableState API in admin service

Configuration

• New frontend.namespaceRPS.visibility and frontend.namespaceBurst.visibility dynamic config for per instance visibility API rate limit
• Removed per service Metrics config
• New RemoteClusters field for controlling TLS settings for takling to remote cluster
• New SKIP_DB_CREATE environment variable for skipping DB creation in auto-setup
• If you use SQL persistence layer, please add the following configuration to your dynamicconfig file so that the ID length limit matches the actual limit used in sql schema. A sample configuration file can be found here

  limit.maxIDLength:
  - value: 255
      constraints: {}
  

Metric

• New matching service task_lag_per_tl metric for estimated task lag
• New service_name metrics tag
• New matching task_write_throttle_count and task_write_latency metrics
• New worrkflow_task_attempt metric
• New task_skipped metrics for number of skipped history tasks

Tool

• New disable-initial-host-lookup flag in temporal-cassandra-tool for disabling initial host lookup

Namespace deletion

• System workflows for marking namespace as deleted and reclaiming resources
• Exclude deleted namespaces from ListNamespace response
• Allow delete namespace by NamesapceId
• Delete namespace feature is still under active development and not production ready

Activity local dispatch

• An optimization for skipping matching service and task polling for running activity tasks
• This optimization requires temporal SDK support and it's still under development

Improvements and fixes

• Buffer event as local namespace after promotion
• Use type from type map before type from Metadata
• Add sqlite persistence tests
• Fix history client NPE for invalid shardID
• Fix timer queue cluster ack level update
• Return ErrExecutionsStillExist if executions are still exist
• Fix record child workflow complete mutable state stale check
• Fix namespace not active error metric
• Unify error handling of CRUD operation of workflow within shard context
• Ensure minimal context timeout for shard context APIs
• Handle workflow not found in replication
• Handle activity retry timer in passive
• Allow delete open workflows from passive sides
• De-duplicate reapply event
• Fix replication error handling for reseting workflow with pending child
• Change operator service error codes if Search attributes are not found or already exist
• Limit timer look ahead
• Use ListWorkflow instead of ScanWorkflow for batch operation
• Remove replication start event sanity check
• Copy version history before returning
• Fix shard context error state check
• Add SetWorkflowExecution persistence API
• Pass useTLS in GetFrontendClientConfig's callback
• Handle delete history branch when mutable state does not exist
• Rework health check
• Support pagination for admin get task queue tasks
• Fix unloading shard after task id sanity check
• Fix get history scheduled tasks query
• Do not generate sync activity task for deleted activity
• Support pagination for SQL GetTransfer/VisibilityTasks
• Fix potential shard deadlock issue
• Update error code and message for missing activity task
• Return NOT_FOUND for invalid workflow ID
• Fixing syntax in first migration to the Cassandra schema
• Backoff failed workflow task
• Add task queue to SQL visibility
• Check if global ns enabled in handler
• Preserve the info when updating is_global_namepsace_enabled
• Add dynamic config change diff logging
• Force metrics.Reporter as metrics extension point
• Improve extensibility of optentelemetry metrics
• Fix DeleteWorkflowExecution API when delete non current execution
• Generate namespace replication only with 2+ clusters
• Add tag support to sdkLogger
• Expose metrics reporters constructors to end-user
• Treat enablement of TLS separately for server and client config
• Add support for GetWorkflowHistory in reverse order
• Fix timerMaxReadLevel map init and update
• Do not blindly retry timer task read
• Exit retry for long poll if context is near deadline
• Convert panic when it is not error

Refactoring

• Delete bootstrap params
• Add context parameter to persistence interface and implementations
• Update iterator to use go generics
• Remove custom build info script
• Replacing SdkClient injection with SdkClientFactory injection
• Make matching persistence range queries [inclusive, exclusive)
• Remove deprecated admin list tasks APIs
• Refactor task filter function and interface
• Unify history task persistence range query behavior
• Refactor persistence task serializer and factory
• Unify history queue processor and task generation implementation
• Provide serverOptions as fx.Out struct
• Make queue processor depend on history engine interface
• Refactor handleWorkflowTaskCompleted retry loop

Others

• Upgrade web images to the latest versions
• Remove docker files
• Removing vim from docker images due to CVE
• Added mockgen to base-ci-builder Docker image
• [docs] improve documentation for building images
• Add Temoporal Web 2.0 to development docker-compose
• Split goreleaser archive in two:cgo/nocgo

All changes

2022-04-11 - bcf9e57 - Client error log level (#2701)
2022-04-06 - 659c052 - Change Invalid search attribute value message (#2705)
2022-04-01 - c1d63b7 - buffer event as local namespace after promotion (#2699)
2022-04-01 - 232ea21 - Use type from type map before type from Metadata (#2697)
2022-03-30 - 306eda8 - Add sqlite persistence tests (#2681)
2022-03-30 - ad56ef2 - Upgrade web images to the latest versions (#2684)
2022-03-30 - 2ebc73e - Delete bootstrap params (#2652)
2022-03-29 - b440f92 - Fix history client NPE for invalid shardID (#2679)
2022-03-29 - ccba09b - Fix timer queue cluster ack level update (#2678)
2022-03-29 - 6b15af3 - Return ErrExecutionsStillExist if executions are still exist (#2676)
2022-03-29 - c6243eb - Set WorkflowID for child workflows (#2675)
2022-03-29 - a274580 - Fix record child workflow complete mutable state stale check (#2673)
2022-03-29 - 3af96c6 - Persistence context part 6: visibility store interfaces (#2674)
2022-03-29 - 1632340 - Update iterator to use go generics (#2672)
2022-03-29 - 22f58a9 - Add DeleteNamespace API unit tests (#2670)
2022-03-29 - 7bff5ee - Improve deletenamespace workflow logging (#2667)
2022-03-28 - 9469618 - Store namespaces in DELETED state in registry (#2665)
2022-03-28 - b87a3ea - Fix parent close policy processor for deleted namespace (#2664)
2022-03-28 - c6a0cfc - Move namespace cache delay to reclaim resources workflow (#2658)
2022-03-28 - 30c2867 - Fix namespace not active error metric (#2663)
2022-03-28 - bfaee75 - Unify error handling of CRUD operation of workflow within shard context (#2662)
2022-03-28 - f066d97 - Ensure minimal context timeout for shard context APIs (#2660)
2022-03-28 - d6f85df - Persistence context part 5: persistence store interfaces (#2659)
2022-03-28 - 642d64a - Merge SQL schema changes for 1.16 release (#2661)
2022-03-25 - e660403 - Handle workflow not found in replication (#2633)
2022-03-25 - 6b8986c - Add reason field for workflow cancellation (#2653)
2022-03-25 - cc51711 - Add DeleteNamespace API (#2645)
2022-03-25 - 752ff50 - Exclude deleted namespaces...

v1.15.2

Release Highlights

This release includes a fix for a potential shard deadlock.

All changes

2022-03-04 - 6d90f65- Fix potential shard deadlock issue(#2570)

Details about v1.15.0 release can be found here.

Helpful links to get you started with Temporal

Temporal Docs
Server
Docker Compose
Helm Chart

Docker images for this release (use tag 1.15.2)

Server
Server With Auto Setup (what is Auto-Setup?)
Admin-Tools

v1.14.6

Release Highlights

This release includes a fix for a potential shard deadlock.

All changes

2022-03-04 - f911933 - Fix potential shard deadlock issue(#2570)

Details about v1.14.0 release can be found here.

Helpful links to get you started with Temporal

Temporal Docs
Server
Docker Compose
Helm Chart

Docker images for this release (use tag 1.14.6)

Server
Server With Auto Setup (what is Auto-Setup?)
Admin-Tools