# Workflow Task errors reference

> For the complete documentation index, see [llms.txt](https://docs.temporal.io/llms.txt).
> Any documentation page is available as raw Markdown by appending `.md` to its URL.

> Workflow Task failure causes and Resource Exhausted causes in Temporal, with what triggers each one and how to resolve it in your Workflow or Worker.

> **ℹ️ Info:**
>
> For the Failure types that Workflows, Activities, and Nexus Operations return, see the [Temporal Failures reference](/references/failures).
> For symptom-based guidance, see [Troubleshooting Workflow Execution failures](/troubleshooting/execution-failures).
>

When a [Workflow Task](/tasks#workflow-task) fails, the Temporal Service records a [WorkflowTaskFailed](/references/events#workflowtaskfailed) Event with a cause in `workflow_task_failed_event_attributes`.
The Temporal Service retries the failed Workflow Task, and the Workflow Execution stays open unless the entry says otherwise.
Many of these failures come from a [Command](/references/commands) the Worker sent with unset or invalid attributes.

Workflow Task failure causes are defined in the [`WorkflowTaskFailedCause` enum](https://github.com/temporalio/api/blob/master/temporal/api/enums/v1/failed_cause.proto) of the Temporal gRPC API.
The [Resource Exhausted causes](#resource-exhausted-causes) on this page come from the `ResourceExhaustedCause` enum in the same file.
They're carried on a gRPC `RESOURCE_EXHAUSTED` response, not on a Workflow Task failure Event.

Each error on this page lists its enum value as **Cause**.

## Command attribute errors

The Temporal Service validates the attributes of each Command.
If a required attribute is unset or a value exceeds a size limit, the Workflow Task fails.
The SDK sets most Command attributes, so an unset attribute can indicate an outdated SDK.

### Bad Cancel Timer Attributes 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_BAD_CANCEL_TIMER_ATTRIBUTES`

The [CancelTimer](/references/commands#canceltimer) Command is missing a Timer ID.

Set a valid Timer ID and deploy the change to your Workers.

### Bad Cancel Workflow Execution Attributes 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_BAD_CANCEL_WORKFLOW_EXECUTION_ATTRIBUTES`

The [CancelWorkflowExecution](/references/commands#cancelworkflowexecution) Command has unset attributes.

Set the missing attributes and deploy the change to your Workers.

### Bad Complete Workflow Execution Attributes 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_BAD_COMPLETE_WORKFLOW_EXECUTION_ATTRIBUTES`

The [CompleteWorkflowExecution](/references/commands#completeworkflowexecution) Command has unset attributes, or its result exceeds the Payload size limit.

Set the missing attributes.
If the Workflow result is too large, store the data in external storage and return a reference to it.
See [Payload size limit](/troubleshooting/blob-size-limit-error).

### Bad Continue as New Attributes 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_BAD_CONTINUE_AS_NEW_ATTRIBUTES`

The [ContinueAsNewWorkflowExecution](/references/commands#continueasnewworkflowexecution) Command has an unset or invalid attribute.

Set the missing attributes.
If the input or Memo exceeds the size limit, reduce its size.
Check that every [Search Attribute](/search-attribute) the new run sets is registered on the Namespace.

### Bad Fail Workflow Execution Attributes 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_BAD_FAIL_WORKFLOW_EXECUTION_ATTRIBUTES`

The [FailWorkflowExecution](/references/commands#failworkflowexecution) Command has unset attributes, such as a missing Failure, or its Failure exceeds the size limit.

Reduce the size of the error details the Workflow returns.
If the attributes are unset, upgrade to the latest release of your [Temporal SDK](/encyclopedia/architecture/temporal-sdks).

### Bad Modify Workflow Properties Attributes 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_BAD_MODIFY_WORKFLOW_PROPERTIES_ATTRIBUTES`

The [ModifyWorkflowProperties](/references/commands#modifyworkflowproperties) Command has unset attributes, or the [Memo](/workflow-execution#memo) update exceeds the size limit.

Set the missing attributes and reduce the size of the Memo.

### Bad Record Marker Attributes 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_BAD_RECORD_MARKER_ATTRIBUTES`

The [RecordMarker](/references/commands#recordmarker) Command has an unset or invalid [Marker](/references/events#markerrecorded) name.

The SDK issues RecordMarker Commands, not your Workflow code.
Upgrade to the latest release of your Temporal SDK.

### Bad Request Cancel Activity Attributes 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_BAD_REQUEST_CANCEL_ACTIVITY_ATTRIBUTES`

The [RequestCancelActivityTask](/references/commands#requestcancelactivitytask) Command has unset attributes, or the SDK's History Builder is in an invalid state.

Upgrade to the latest release of your Temporal SDK.
If the error continues, check your Workflow code for [non-deterministic changes](/workflow-definition#non-deterministic-change).

### Bad Request Cancel External Workflow Execution Attributes 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_BAD_REQUEST_CANCEL_EXTERNAL_WORKFLOW_EXECUTION_ATTRIBUTES`

The [RequestCancelExternalWorkflowExecution](/references/commands#requestcancelexternalworkflowexecution) Command has unset or invalid attributes.

Set any missing attributes, such as the Workflow ID or Run ID, and shorten any fields that exceed length limits.

A Workflow Task can't both start a [Child Workflow](/child-workflows) and request its cancellation.
If your Workflow does both in the same Workflow Task, request the cancellation in a later Workflow Task.

### Bad Request Cancel Nexus Operation Attributes 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_BAD_REQUEST_CANCEL_NEXUS_OPERATION_ATTRIBUTES`

The [RequestCancelNexusOperation](/references/commands#requestcancelnexusoperation) Command has unset or invalid attributes.

Read the reason in the error message for the attribute that failed validation.

### Bad Schedule Activity Attributes 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_BAD_SCHEDULE_ACTIVITY_ATTRIBUTES`

The [ScheduleActivityTask](/references/commands#scheduleactivitytask) Command has unset or invalid attributes, or the Activity input exceeds the Payload size limit.

Set a Start-To-Close Timeout or a Schedule-To-Close Timeout on the [Activity](/activities).
If the input is too large, store the data in external storage and pass a reference to the Activity.

### Bad Schedule Nexus Operation Attributes 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_BAD_SCHEDULE_NEXUS_OPERATION_ATTRIBUTES`

The [ScheduleNexusOperation](/references/commands#schedulenexusoperation) Command has unset or invalid attributes.
For example, the Nexus Endpoint name in the caller Workflow doesn't exist.

Read the reason in the error message for the attribute that failed validation.

### Bad Search Attributes 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_BAD_SEARCH_ATTRIBUTES`

The Workflow set unset or invalid [Search Attributes](/search-attribute).
The Workflow Task keeps failing on retry until you fix the Search Attributes.

Register each Search Attribute on the Namespace with the type your Workflow uses, and keep values within the size limits.

### Bad Signal Input Size 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_BAD_SIGNAL_INPUT_SIZE`

The input of a [Signal](/sending-messages#sending-signals) sent to another Workflow Execution exceeds the size limit.

Reduce the size of the Signal input and deploy the change to your Workers.

### Bad Signal Workflow Execution Attributes 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_BAD_SIGNAL_WORKFLOW_EXECUTION_ATTRIBUTES`

The [SignalExternalWorkflowExecution](/references/commands#signalexternalworkflowexecution) Command has unset or invalid attributes, or its input exceeds the size limit.

Set the missing attributes and reduce the size of the input.

### Bad Start Child Execution Attributes 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_BAD_START_CHILD_EXECUTION_ATTRIBUTES`

The [StartChildWorkflowExecution](/references/commands#startchildworkflowexecution) Command has unset or invalid attributes, or its input exceeds the size limit.

Reduce the size of the Child Workflow input.
Check that every Search Attribute the Child Workflow sets is registered on the Namespace.

### Bad Start Timer Attributes 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_BAD_START_TIMER_ATTRIBUTES`

The [StartTimer](/references/commands#starttimer) Command is missing a Timer ID.

Set a valid Timer ID and deploy the change to your Workers.

### Cause Bad Update 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_BAD_UPDATE_WORKFLOW_EXECUTION_MESSAGE`

A [Workflow Update](/sending-messages#sending-updates) message from the [Worker](/workers#worker) (Acceptance, Rejection, or Response) has an invalid format or is missing required fields.

This error can indicate an unsupported SDK.
Use a [supported SDK](/encyclopedia/architecture/temporal-sdks) and upgrade to its latest release.

## Duplicate ID errors

### Schedule Activity Duplicate ID 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_SCHEDULE_ACTIVITY_DUPLICATE_ID`

The Workflow scheduled an [Activity](/activities) with an Activity ID that's already in use in the Workflow Execution.

Give each Activity in the Workflow a unique Activity ID.

### Start Timer Duplicate ID 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_START_TIMER_DUPLICATE_ID`

The Workflow started a Timer with a Timer ID that's already in use in the Workflow Execution.

Give each Timer in the Workflow a unique Timer ID.

## Pending operation limit errors

A Workflow Execution has a maximum number of pending operations of each type.
When a Command would exceed a limit, the Workflow Task fails and is retried.
See [Workflow Execution limits](/workflow-execution/limits#workflow-execution-limits).

On a self-hosted Temporal Service, you can change these limits with [dynamic configuration](/references/dynamic-configuration#size-limit-settings).

### Pending Activities Limit Exceeded 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_PENDING_ACTIVITIES_LIMIT_EXCEEDED`
- Default limit: 2,000 pending Activities

Scheduling another [Activity](/activities) would exceed the Workflow Execution's limit of pending Activities.

Reduce the number of Activities the Workflow runs at the same time, for example by processing work in batches.

### Pending Child Workflows Limit Exceeded 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_PENDING_CHILD_WORKFLOWS_LIMIT_EXCEEDED`
- Default limit: 2,000 pending Child Workflows

Starting another [Child Workflow](/child-workflows) would exceed the Workflow Execution's limit of pending Child Workflows.

Reduce the number of Child Workflows the Workflow runs at the same time.

### Pending Nexus Operations Limit Exceeded 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_PENDING_NEXUS_OPERATIONS_LIMIT_EXCEEDED`
- Default limit: 30 pending Nexus Operations

Scheduling another Nexus Operation would exceed the Workflow Execution's limit of pending Nexus Operations.

Reduce the number of Nexus Operations the Workflow runs at the same time.
See [Workflow Execution Nexus Operation limits](/workflow-execution/limits#workflow-execution-nexus-operation-limits) and, for Temporal Cloud, [Per Workflow Nexus Operation limits](/evaluate/cloud/limits#per-workflow-nexus-operation-limits).

### Pending Request Cancel Limit Exceeded 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_PENDING_REQUEST_CANCEL_LIMIT_EXCEEDED`
- Default limit: 2,000 pending Cancellation Requests to other Workflows

Requesting cancellation of another Workflow Execution would exceed the limit of pending Cancellation Requests.

Reduce the number of Cancellation Requests the Workflow sends at the same time.

### Pending Signals Limit Exceeded 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_PENDING_SIGNALS_LIMIT_EXCEEDED`
- Default limit: 2,000 pending Signals to other Workflows

Sending another [Signal](/sending-messages#sending-signals) to an external Workflow Execution would exceed the limit of pending outgoing Signals.

Reduce the number of Signals the Workflow sends to other Workflow Executions at the same time.

## Size limit errors

### gRPC Message Too Large 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_GRPC_MESSAGE_TOO_LARGE`

The Workflow Task completion exceeds the 4 MB gRPC message size limit.
Replay produces the same oversized request, so the Temporal Service terminates the Workflow Execution.

This error usually means the Workflow schedules too many Activities, Child Workflows, or Commands in a single Workflow Task, or returns a large result.

Fix your Workflow code and start a new Workflow Execution.
Split work into smaller batches, return smaller results, use [Continue-As-New](/workflow-execution/continue-as-new) for long-running Workflows, or compress large Payloads with a custom [Payload Codec](/payload-codec).
See [gRPC message size limit](/troubleshooting/blob-size-limit-error).

### Payloads Too Large 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_PAYLOADS_TOO_LARGE`

A Payload in the Workflow Task completion exceeds the Payload size limit.
SDKs that report this cause, such as Python SDK 1.23.0 and later and Go SDK 1.43.0 and later, leave the Workflow Execution open.

Store large data in external storage, pass a reference through the Workflow, and deploy the fix to your Workers.
See [Payload size limit](/troubleshooting/blob-size-limit-error).

## Workflow code and Worker errors

### Nondeterminism Error 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_NON_DETERMINISTIC_ERROR`

During Replay, the Workflow code produced Commands that don't match the Event History.
This happens when Workflow code changes in a [non-deterministic way](/workflow-definition#non-deterministic-change) while Workflow Executions are open.

See [Non-determinism error](/troubleshooting/execution-failures#non-determinism-error) for how to find and fix the change.

### Workflow Worker Unhandled Failure 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_WORKFLOW_WORKER_UNHANDLED_FAILURE`

The [Workflow Definition](/workflow-definition) raised an error or panic that the SDK doesn't treat as a Workflow Execution failure.

Read the stack trace in the WorkflowTaskFailed Event or the Worker logs, fix the code, and deploy the change to your Workers.

### Unhandled Command 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_UNHANDLED_COMMAND`

New [Events](/references/events), such as Signals, arrived while the Workflow Task was running, and the Workflow tried to close without handling them.
The Temporal Service fails the Workflow Task and schedules a new one that includes the new Events.

This error is common when a Workflow receives many [Signals](/sending-messages#sending-signals).
Before the Workflow completes, process every pending Signal.
In the Go SDK, drain the Signal Channel with `ReceiveAsync`.

### Cause Unhandled Update 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_UNHANDLED_UPDATE`

The Temporal Service received a Workflow Update while the Worker processed a Workflow Task that would close the Workflow Execution.

The Temporal Service rejects the Workflow Task completion and rewinds the Workflow so that Workflow code handles the pending Update.
This error is common when a Workflow receives frequent Updates.

### Cause Bad Binary 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_BAD_BINARY`

The [Worker](/workers) that completed the Workflow Task has a binary checksum that's marked as bad on the Namespace.

Deploy a Worker build that isn't marked as bad.

### Cause Unspecified 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_UNSPECIFIED`

The Workflow Task failed without a specific cause.

Read the failure message in the WorkflowTaskFailed Event and check your Workflow Definition.

## Temporal Service errors

The Temporal Service fails these Workflow Tasks because of an operation outside your Workflow code.

### Cause Reset Workflow 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_RESET_WORKFLOW`

The Workflow Task failed because the Workflow Execution was [reset](/workflow-execution/event#reset).

The reset starts a new run of the Workflow Execution.
If the new run didn't start, reset the Workflow Execution again.

### Reset Sticky Task Queue 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_RESET_STICKY_TASK_QUEUE`

The Worker failed the Workflow Task so that the next Workflow Task goes to the normal [Task Queue](/task-queue) instead of the Sticky Task Queue.

No action is needed.
The Temporal Service schedules the next Workflow Task automatically.

### Failover Close Command 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_FAILOVER_CLOSE_COMMAND`

A [Namespace](/namespaces) failover forced the Workflow Task to close.

No action is needed.
The Temporal Service schedules a retry automatically.

### Force Close Command 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_FORCE_CLOSE_COMMAND`

The Temporal Service forced the Workflow Task to close.
If the error is recoverable, the Temporal Service schedules a retry.

### Feature Disabled 

- Cause: `WORKFLOW_TASK_FAILED_CAUSE_FEATURE_DISABLED`

The Workflow Task requested a feature that's disabled on the Temporal Service, either for the whole Temporal Service or, more often, for the Workflow's Namespace.

Read the failure message in the WorkflowTaskFailed Event for the feature name, then turn on the feature for the Namespace or remove its use from the Workflow.

## Resource Exhausted causes 

The Temporal Service returns a gRPC `RESOURCE_EXHAUSTED` error with one of these causes when it throttles a request.
SDKs retry these requests automatically.
For triage steps, see [RESOURCE_EXHAUSTED on user-facing operations](/troubleshooting/request-failures#resource_exhausted-on-user-facing-operations).

### Resource Exhausted Cause APS Limit 

- Cause: `RESOURCE_EXHAUSTED_CAUSE_APS_LIMIT`

The caller exceeded the Namespace's actions per second limit.

Reduce your request rate, or on Temporal Cloud, compare your throughput with the Namespace [limits](/evaluate/cloud/limits) and request an increase.

### Resource Exhausted Cause Busy Workflow 

- Cause: `RESOURCE_EXHAUSTED_CAUSE_BUSY_WORKFLOW`

The target Workflow Execution is busy.

Reduce the rate of requests, such as Signals and Updates, that you send to a single Workflow Execution.

### Resource Exhausted Cause Circuit Breaker Open 

- Cause: `RESOURCE_EXHAUSTED_CAUSE_CIRCUIT_BREAKER_OPEN`

A circuit breaker in the Temporal Service is open or half-open, and the Temporal Service is shedding load.

Treat this cause as a capacity problem.
Raising a rate limit doesn't help.
Add capacity to the Temporal Service or slow down your workload.

### Resource Exhausted Cause Concurrent Limit 

- Cause: `RESOURCE_EXHAUSTED_CAUSE_CONCURRENT_LIMIT`

The caller exceeded the maximum number of concurrent requests, such as long-poll requests from Workers.

Reduce the [poller count](/develop/worker-performance/task-queues#poller-count) per [Worker](/workers) or the number of Workers polling the Namespace.

### Resource Exhausted Cause Ops Limit 

- Cause: `RESOURCE_EXHAUSTED_CAUSE_OPS_LIMIT`

The Namespace exceeded its operations rate limit.

Reduce your request rate, or on Temporal Cloud, compare your throughput with the Namespace [limits](/evaluate/cloud/limits) and request an increase.

### Resource Exhausted Cause Persistence Limit 

- Cause: `RESOURCE_EXHAUSTED_CAUSE_PERSISTENCE_LIMIT`

The Namespace exceeded the persistence rate limit.

On a self-hosted Temporal Service, check persistence latency and the [persistence QPS limits](/references/dynamic-configuration#qps-limits-for-persistence-store).

### Resource Exhausted Cause Persistence Storage Limit 

- Cause: `RESOURCE_EXHAUSTED_CAUSE_PERSISTENCE_STORAGE_LIMIT`

The persistence storage limit was exceeded.

### Resource Exhausted Cause RPS Limit 

- Cause: `RESOURCE_EXHAUSTED_CAUSE_RPS_LIMIT`

The caller exceeded the requests per second limit.

Reduce your request rate.
On Temporal Cloud, compare your throughput with the Namespace [limits](/evaluate/cloud/limits) and request an increase.
On a self-hosted Temporal Service, check the [Service-level RPS limits](/references/dynamic-configuration#service-level-rps-limits) after you confirm that persistence is healthy.

### Resource Exhausted Cause System Overload 

- Cause: `RESOURCE_EXHAUSTED_CAUSE_SYSTEM_OVERLOADED`

The Temporal Service is overloaded and is shedding load to protect itself.

Treat this cause as a capacity problem.
Raising a rate limit doesn't help.
Add capacity to the Temporal Service or slow down your workload.

### Resource Exhausted Cause Worker Deployment Limits 

- Cause: `RESOURCE_EXHAUSTED_CAUSE_WORKER_DEPLOYMENT_LIMITS`

The request reached a [Worker Versioning limit](/evaluate/cloud/limits#worker-versioning-level), such as the maximum number of Worker Deployments in a Namespace or Versions in a Worker Deployment.

To free space for a new Version, stop polling from an old, drained Version so the Temporal Service can delete it.
See [Garbage collection](/production-deployment/worker-deployments/worker-versioning/sunset-and-gc#garbage-collection).

### Resource Exhausted Cause Unspecified 

- Cause: `RESOURCE_EXHAUSTED_CAUSE_UNSPECIFIED`

The Temporal Service throttled the request without a specific cause.