Struct WorkerConfig 

#[non_exhaustive]
pub struct WorkerConfig {

    pub namespace: String,
    pub task_queue: String,
    pub client_identity_override: Option<String>,
    pub max_cached_workflows: usize,
    pub tuner: Option<Arc<dyn WorkerTuner + Sync + Send>>,
    pub workflow_task_poller_behavior: PollerBehavior,
    pub nonsticky_to_sticky_poll_ratio: f32,
    pub activity_task_poller_behavior: PollerBehavior,
    pub nexus_task_poller_behavior: PollerBehavior,
    pub no_remote_activities: bool,
    pub sticky_queue_schedule_to_start_timeout: Duration,
    pub max_heartbeat_throttle_interval: Duration,
    pub default_heartbeat_throttle_interval: Duration,
    pub max_task_queue_activities_per_second: Option<f64>,
    pub max_worker_activities_per_second: Option<f64>,
    pub ignore_evicts_on_shutdown: bool,
    pub fetching_concurrency: usize,
    pub graceful_shutdown_period: Option<Duration>,
    pub local_timeout_buffer_for_activities: Duration,
    pub workflow_failure_errors: HashSet<WorkflowErrorType>,
    pub workflow_types_to_failure_errors: HashMap<String, HashSet<WorkflowErrorType>>,
    pub max_outstanding_workflow_tasks: Option<usize>,
    pub max_outstanding_activities: Option<usize>,
    pub max_outstanding_local_activities: Option<usize>,
    pub max_outstanding_nexus_tasks: Option<usize>,
    pub versioning_strategy: WorkerVersioningStrategy,
    pub heartbeat_interval: Option<Duration>,
}

Defines per-worker configuration options

Fields (Non-exhaustive)

Non-exhaustive structs could have additional fields added in future. Therefore, non-exhaustive structs cannot be constructed in external crates using the traditional Struct { .. } syntax; cannot be matched against without a wildcard ..; and struct update syntax will not work.

namespace: String

The Temporal service namespace this worker is bound to

task_queue: String

What task queue will this worker poll from? This task queue name will be used for both workflow and activity polling.

client_identity_override: Option<String>

A human-readable string that can identify this worker. Using something like sdk version and host name is a good default. If set, overrides the identity set (if any) on the client used by this worker.

max_cached_workflows: usize

If set nonzero, workflows will be cached and sticky task queues will be used, meaning that history updates are applied incrementally to suspended instances of workflow execution. Workflows are evicted according to a least-recently-used policy one the cache maximum is reached. Workflows may also be explicitly evicted at any time, or as a result of errors or failures.

tuner: Option<Arc<dyn WorkerTuner + Sync + Send>>

Set a WorkerTuner for this worker. Either this or at least one of the max_outstanding_* fields must be set.

workflow_task_poller_behavior: PollerBehavior

Maximum number of concurrent poll workflow task requests we will perform at a time on this worker’s task queue. See also WorkerConfig::nonsticky_to_sticky_poll_ratio. If using SimpleMaximum, Must be at least 2 when max_cached_workflows > 0, or is an error.

nonsticky_to_sticky_poll_ratio: f32

Only applies when using PollerBehavior::SimpleMaximum

(max workflow task polls * this number) = the number of max pollers that will be allowed for the nonsticky queue when sticky tasks are enabled. If both defaults are used, the sticky queue will allow 4 max pollers while the nonsticky queue will allow one. The minimum for either poller is 1, so if the maximum allowed is 1 and sticky queues are enabled, there will be 2 concurrent polls.

activity_task_poller_behavior: PollerBehavior

Maximum number of concurrent poll activity task requests we will perform at a time on this worker’s task queue

nexus_task_poller_behavior: PollerBehavior

Maximum number of concurrent poll nexus task requests we will perform at a time on this worker’s task queue

no_remote_activities: bool

If set to true this worker will only handle workflow tasks and local activities, it will not poll for activity tasks.

sticky_queue_schedule_to_start_timeout: Duration

How long a workflow task is allowed to sit on the sticky queue before it is timed out and moved to the non-sticky queue where it may be picked up by any worker.

max_heartbeat_throttle_interval: Duration

Longest interval for throttling activity heartbeats

default_heartbeat_throttle_interval: Duration

Default interval for throttling activity heartbeats in case ActivityOptions.heartbeat_timeout is unset. When the timeout is set in the ActivityOptions, throttling is set to heartbeat_timeout * 0.8.

max_task_queue_activities_per_second: Option<f64>

Sets the maximum number of activities per second the task queue will dispatch, controlled server-side. Note that this only takes effect upon an activity poll request. If multiple workers on the same queue have different values set, they will thrash with the last poller winning.

Setting this to a nonzero value will also disable eager activity execution.

max_worker_activities_per_second: Option<f64>

Limits the number of activities per second that this worker will process. The worker will not poll for new activities if by doing so it might receive and execute an activity which would cause it to exceed this limit. Negative, zero, or NaN values will cause building the options to fail.

ignore_evicts_on_shutdown: bool

If set false (default), shutdown will not finish until all pending evictions have been issued and replied to. If set true shutdown will be considered complete when the only remaining work is pending evictions.

This flag is useful during tests to avoid needing to deal with lots of uninteresting evictions during shutdown. Alternatively, if a lang implementation finds it easy to clean up during shutdown, setting this true saves some back-and-forth.

fetching_concurrency: usize

Maximum number of next page (or initial) history event listing requests we’ll make concurrently. I don’t this it’s worth exposing this to users until we encounter a reason.

graceful_shutdown_period: Option<Duration>

If set, core will issue cancels for all outstanding activities and nexus operations after shutdown has been initiated and this amount of time has elapsed.

local_timeout_buffer_for_activities: Duration

The amount of time core will wait before timing out activities using its own local timers after one of them elapses. This is to avoid racing with server’s own tracking of the timeout.

workflow_failure_errors: HashSet<WorkflowErrorType>

Any error types listed here will cause any workflow being processed by this worker to fail, rather than simply failing the workflow task.

workflow_types_to_failure_errors: HashMap<String, HashSet<WorkflowErrorType>>

Like WorkerConfig::workflow_failure_errors, but specific to certain workflow types (the map key).

max_outstanding_workflow_tasks: Option<usize>

The maximum allowed number of workflow tasks that will ever be given to this worker at one time. Note that one workflow task may require multiple activations - so the WFT counts as “outstanding” until all activations it requires have been completed. Must be at least 2 if max_cached_workflows is > 0, or is an error.

Mutually exclusive with tuner

max_outstanding_activities: Option<usize>

The maximum number of activity tasks that will ever be given to this worker concurrently

Mutually exclusive with tuner

max_outstanding_local_activities: Option<usize>

The maximum number of local activity tasks that will ever be given to this worker concurrently

Mutually exclusive with tuner

max_outstanding_nexus_tasks: Option<usize>

The maximum number of nexus tasks that will ever be given to this worker concurrently

Mutually exclusive with tuner

versioning_strategy: WorkerVersioningStrategy

A versioning strategy for this worker.

heartbeat_interval: Option<Duration>

The interval within which the worker will send a heartbeat. The timer is reset on each existing RPC call that also happens to send this data, like PollWorkflowTaskQueueRequest.

Implementations

impl WorkerConfig

pub fn should_fail_workflow( &self, workflow_type: &str, error_type: &WorkflowErrorType, ) -> bool

Returns true if the configuration specifies we should fail a workflow on a certain error type rather than failing the workflow task.

pub fn computed_deployment_version(&self) -> Option<WorkerDeploymentVersion>

Trait Implementations

impl Clone for WorkerConfig

fn clone(&self) -> WorkerConfig

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.