-
Story
-
Resolution: Done
-
Undefined
-
None
-
None
-
None
-
False
-
-
False
-
-
Description
As a Parodos user, I want to be able to execute and monitor an escalation workflow so that I can achieve the creation of a namespace in OpenShift.
Scenario: The ticket is closed with no escalation
Acceptance Criteria:
GIVEN
the namespace name AND the escalation manager email
WHEN
I create an escalation use case
AND A Jira ticket is created
AND the ticket is closed before the configured timeout
THEN
the namespace is created with no escalation email
Scenario: The ticket is closed with an escalation
Acceptance Criteria:
GIVEN
the namespace name AND the escalation manager email
WHEN
I create an escalation use case
AND A Jira ticket is created
AND the ticket is not closed before the configured timeout
AND an escalation email is sent to the escalation manager
AND the escalation manager manages ticket closure
THEN
the namespace is created
Acceptance criteria
- The workflow is implemented using a SonataFlow workflow (SWF)
- The Jira server and its credentials are configured in the SWF runtime
- The mail is sent using a public service like MailTrap and the server and credentials are configured in the SWF runtime
- The escalation timeout is configured in the SWF runtime
- The implementation is not adopting a pull-model but an Event Driven Architecture (EDA)
- The workflow is deployed on an OpenShift/Kubernetes cluster and leveraging the Red Hat Serverless/Knative resources to define the eventing infrastructure
- Instructions on how to build, publish and configure the container image(s) are documented
- The workflow is deployed as a serverless service
- An external (serverless) application handles updates from the Jira server and resume the pending workflows
- The Jira server is configured to forward webhook notifications to this application
- The application leverages the Knative infrastructure to resume the workflows
- The output includes:
- The configuration of the created namespace
- A flag specifying whether the escalation was triggered or not
- Error handling: TBD
- Unit tests:
- The notification listener application includes unit test code for the major use cases
- The SWF includes unit test code for the major use cases (TBD is it doable?)
- The complete example is documented and available in the parodos-dev/serverless-workflow-examples repository
- depends on
-
KOGITO-9785 Workflow and State timeout support for Error Handling
-
- Resolved
-
-
FLPATH-627 Document proposed Architecture
-
- Closed
-
-
FLPATH-628 Design Ticker listener
-
- Closed
-
-
FLPATH-629 Design ticket-escalation SWF
-
- Closed
-
-
FLPATH-630 Deploy the example
-
- Closed
-
-
FLPATH-631 (optional) Unit Test
-
- Closed
-
-
FLPATH-632 Extend the SonataFlow timeout specifications
-
- Closed
-
-
FLPATH-633 Integrating with backstage
-
- Closed
-
- is blocked by
-
FLPATH-611 Design escalations and checkers
-
- Closed
-