[Yandex Cloud documentation](../../../index.md) > [Yandex Serverless Integrations](../../index.md) > [Concepts](../index.md) > Workflows > Starting a workflow execution
# Workflow execution
An execution contains all information about a particular workflow run. Possible execution statuses: `In queue`, `In progress`, `Paused`, `Canceled`, `Error`, and `Executed`. Learn more about [possible error codes](#errors).
You can start a workflow in one of the following ways:
* Via the management console, CLI, or API. Read more in [Starting a workflow execution](../../operations/workflows/execution/start.md).
* On a timer. The schedule is set in [cron expression](../cron.md) format. Read more in [Scheduling a workflow to start on a timer](../../operations/workflows/execution/start-by-timer.md).
* Via a link over HTTPS. Read more in [Executing a workflow via a link](../../operations/workflows/execution/start-by-link.md).
* Using an API gateway. Read more in [Adding the x-yc-apigateway-integration:http extension for integration with Yandex Workflows](../../../api-gateway/operations/spec-constructor/workflows.md).
* Using Yandex EventRouter. To do this, you need to specify the workflow as a target in a [rule](../eventrouter/rule.md). For more on how to create a rule, see [Creating a rule with a Yandex Workflows target](../../operations/eventrouter/rule/create-workflows.md). The rule defines the events that should trigger a workflow. The event body is provided to the execution as an input parameter.
If a workflow is:
* Private, to run it, the user needs workflow execution permissions, e.g., the [serverless.workflows.executor](../../security/workflows.md#serverless-workflows-executor) role.
* Public, any user can run it. Read more in [Making a workflow public](../../operations/workflows/workflow/make-public.md).
## Possible error codes {#errors}
Error | Description
--- | ---
`ALL` | Error you can specify in a retry policy for the policy to apply to any error type other than `STEP_INTERNAL`.
`STEP_DATA_LIMIT_EXCEEDED` | Input or output data limit exceeded. Read more in [Quotas and limits in Serverless Integrations](../limits.md).
`STEP_NO_CHOICE_MATCHED` | There is no suitable execution path in `choices`. Read more in [Switch](yawl/management/switch.md).
`STEP_PERMISSION_DENIED` | No access to resource.
`STEP_TIMEOUT` | Step timeout exceeded. Read more in [Quotas and limits in Serverless Integrations](../limits.md).
`STEP_INVALID_OUTPUT` | Invalid output data.
`STEP_INTERNAL` | Internal error. If a step terminates with this error, you cannot apply the retry policy, and the execution immediately gets the `Error` status.
`STEP_INVALID_TEMPLATE_EXPRESSION` | Invalid jq template.
`STEP_FAIL` | The execution ended with an error at the `Fail` step. Read more in [Fail](yawl/management/fail.md).
`STEP_FAILED_PRECONDITION` | The resource state is invalid to complete the step, e.g., an email address is unverified or blocked.
`STEP_INVALID_ARGUMENT` | Invalid step parameters.
`STEP_QUOTA_EXCEEDED` | The resource request limit is reached.
`HTTP_CALL_400`
`HTTP_CALL_401`
`HTTP_CALL_402`
`HTTP_CALL_403`
`HTTP_CALL_404`
`HTTP_CALL_405`
`HTTP_CALL_406`
`HTTP_CALL_407`
`HTTP_CALL_408`
`HTTP_CALL_409`
`HTTP_CALL_410`
`HTTP_CALL_411`
`HTTP_CALL_412`
`HTTP_CALL_413`
`HTTP_CALL_414`
`HTTP_CALL_415`
`HTTP_CALL_416`
`HTTP_CALL_417`
`HTTP_CALL_418`
`HTTP_CALL_419`
`HTTP_CALL_420`
`HTTP_CALL_421`
`HTTP_CALL_422`
`HTTP_CALL_423`
`HTTP_CALL_424`
`HTTP_CALL_425`
`HTTP_CALL_426`
`HTTP_CALL_427`
`HTTP_CALL_428`
`HTTP_CALL_429`
`HTTP_CALL_431`
`HTTP_CALL_449`
`HTTP_CALL_451`
`HTTP_CALL_499`
`HTTP_CALL_500`
`HTTP_CALL_501`
`HTTP_CALL_505`
`HTTP_CALL_502`
`HTTP_CALL_503`
`HTTP_CALL_504`
`HTTP_CALL_506`
`HTTP_CALL_507`
`HTTP_CALL_508`
`HTTP_CALL_509`
`HTTP_CALL_510`
`HTTP_CALL_511`
`HTTP_CALL_520`
`HTTP_CALL_521`
`HTTP_CALL_522`
`HTTP_CALL_523`
`HTTP_CALL_524`
`HTTP_CALL_525`
`HTTP_CALL_526` | HTTP response state codes. For more information, see [MDN Web Docs](https://developer.mozilla.org/en-US/docs/Web/HTTP/Status#client_error_responses).
`GRPC_CALL_CANCELLED`
`GRPC_CALL_DEADLINE_EXCEEDED`
`GRPC_CALL_UNIMPLEMENTED`
`GRPC_CALL_UNAVAILABLE`
`GRPC_CALL_UNKNOWN`
`GRPC_CALL_INTERNAL`
`GRPC_CALL_RESOURCE_EXHAUSTED`
`GRPC_CALL_UNAUTHENTICATED` | gRPC statuses. For more information, see [this gRPC guide](https://grpc.github.io/grpc/core/md_doc_statuscodes.html).
`GRPC_CALL_INVALID_OPTIONS` | Invalid gRPC call parameters.
`GRPC_CALL_INVALID_REFLECTION_SERVER_RESPONSE` | Unexpected response from gRPC server.
`FUNCTION_CALL_INVALID_RESPONSE` | Invalid function code or format of returned JSON response. Read more in [Possible Cloud Functions response codes](../../../functions/concepts/function-invoke.md#http-state).
`CONTAINER_CALL_400`
`CONTAINER_CALL_401`
`CONTAINER_CALL_402`
`CONTAINER_CALL_403`
`CONTAINER_CALL_404`
`CONTAINER_CALL_405`
`CONTAINER_CALL_406`
`CONTAINER_CALL_407`
`CONTAINER_CALL_408`
`CONTAINER_CALL_409`
`CONTAINER_CALL_410`
`CONTAINER_CALL_411`
`CONTAINER_CALL_412`
`CONTAINER_CALL_413`
`CONTAINER_CALL_414`
`CONTAINER_CALL_415`
`CONTAINER_CALL_416`
`CONTAINER_CALL_417`
`CONTAINER_CALL_418`
`CONTAINER_CALL_419`
`CONTAINER_CALL_420`
`CONTAINER_CALL_421`
`CONTAINER_CALL_422`
`CONTAINER_CALL_423`
`CONTAINER_CALL_424`
`CONTAINER_CALL_425`
`CONTAINER_CALL_426`
`CONTAINER_CALL_427`
`CONTAINER_CALL_428`
`CONTAINER_CALL_429`
`CONTAINER_CALL_431`
`CONTAINER_CALL_449`
`CONTAINER_CALL_451`
`CONTAINER_CALL_499`
`CONTAINER_CALL_500`
`CONTAINER_CALL_501`
`CONTAINER_CALL_505`
`CONTAINER_CALL_502`
`CONTAINER_CALL_503`
`CONTAINER_CALL_504`
`CONTAINER_CALL_506`
`CONTAINER_CALL_507`
`CONTAINER_CALL_508`
`CONTAINER_CALL_509`
`CONTAINER_CALL_510`
`CONTAINER_CALL_511`
`CONTAINER_CALL_520`
`CONTAINER_CALL_521`
`CONTAINER_CALL_522`
`CONTAINER_CALL_523`
`CONTAINER_CALL_524`
`CONTAINER_CALL_525`
`CONTAINER_CALL_526` | HTTP response state codes. For more information, see [MDN Web Docs](https://developer.mozilla.org/en-US/docs/Web/HTTP/Status#client_error_responses).
`YDB_CALL_SERVICE_UNAVAILABLE` | Temporary server-side failure.