Understanding Workflow Services Status Conditions

This document describes the status and conditions of the KogitoServerlessWorkflow object deployed by the Kogito Serverless Workflow Operator.

Kubernetes Status is an important property to observe in order to understand what is currently happening with the object. It can also help you troubleshoot or integrate with other objects in the cluster.

General Status

The table below lists the general structure of a workflow status:

Table 1. Description of KogitoServerlessWorkflow status object
Status	Description
`ObservedGeneration`	Last object generation observed by the status
`Conditions`	Current observed workflow conditions
`Address`	External or internal workflow access endpoint. Implements the Knative Addressable Interface
`RecoverFailureAttempts`	How many attempts the operator tried to recover from a failure

The Conditions property might vary depending on the workflow profile. The next sections describe the current implementation.

Development Profile Conditions

When you deploy a workflow with the development profile, the operator deploys a ready-to-use container with a running workflow instance.

The following table lists the possible conditions.

Table 2. Conditions Scenarios in Development
Condition	Status	Reason	Description
Built	Unknown		In development profile there’s no build process, so this state remains in `Unknown` status
Running	True		The workflow is running and in healthy state
Running	False	WaitingForDeployment	The workflow is waiting for the underlying deployment object to have a minimum availability
Running	False	DeploymentFailure	There was a problem with the underlying deployment object. Check the message in this condition and the workflow pod logs for more info
Running	False	DeploymentIsUnavailable	The underlying deployment object doesn’t have the minimum availability for this workflow. Check the message in this condition and the workflow pod logs for more info
Running	False	AttemptToRedeployFailed	If the workflow deployment is not available, the operator will try to rollout the deployment three times before entering in this stage. Check the message in this condition and the workflow pod logs for more info

In normal conditions, the workflow will transition from Running, WaitingForDeployment condition to Running. In case something wrong happens, consult the section Workflow Troubleshooting in Development.

Production Profile Conditions

Deploying the workflow in Production profile makes the operator to build an immutable image for the workflow service. The build step can be followed by observing the workflow conditions.

Table 3. Condition Scenarios in Production
Condition	Status	Reason	Description
Built	False	BuildIsRunningReason	The build is currently running
Built	False	BuildFailedReason	The build has failed and any attempts to have it run failed
Built	True		Last build has finished successfully
Running	True		The workflow is running and in healthy state
Running	False	WaitingForPlatform	The workflow can’t have a pod running until a `KogitoServerlessPlatform` is ready
Running	False	WaitingForBuildReason	The workflow is waiting for the build to finish to start running
Running	False	WaitingForDeployment	The workflow is waiting for the underlying deployment object to have a minimum availability
Running	False	DeploymentFailure	There was a problem with the underlying deployment object. Check the message in this condition and the workflow pod logs for more info
Running	False	DeploymentIsUnavailable	The underlying deployment object doesn’t have the minimum availability for this workflow. Check the message in this condition and the workflow pod logs for more info
Running	False	AttemptToRedeployFailed	If the workflow deployment is not available, the operator will try to roll out the deployment three times before entering this stage. Check the message in this condition and the workflow pod logs for more info

The normal conditions for the workflow resource are to place a KogitoServerlessBuild to run and wait for it to finish. As soon as the image is ready, the workflow transitions to the deployment phase, which is to provision a new workflow service pod to run with the built image.

Additional resources

Found an issue?

If you find an issue or any misleading information, please feel free to report it here. We really appreciate it!