community 
How to abort a reservation after a critical failure during setup?
Hello folks,
My group has identified a few events that occur during out sandbox setup_script’s execution which we consider to be critical failures. 'Critical Failures' mean something which should not allow the reservation to continue. Specifically, I want to:
- 'Fail' the current reservation
- Set the sandbox state to something other than ‘active’ or ‘activeWithError’
- This is important for informing end-users AND automation
- Set the sandbox state to something other than ‘active’ or ‘activeWithError’
- Somehow report a failure back to the Jenkins job that started the reservation
- I tried using CloudShellAPISession.SetReservationLiveStatus() combined with Sandbox API GET /sandboxes/{sandbox_identifier}, but reservationLiveStatus doesn't seem to be returned by GET /sandboxes/{sandbox_identifier}
- Run the sandbox’s teardown script
- Our teardown script destroys any applications in sandbox that were successfully deployed by setup_script
Has anyone implemented anything like this? Can anyone suggest where I can start?
Best Answer
Since you simply wants to end the sandbox and run teardown in this case, you may do the following in your Setup when 'Critical Failures' events occur:
from cloudshell.workflow.orchestration.sandbox import Sandbox
sandbox = Sandbox()
api_session = sandbox.automation_api
reservation_id = sandbox.id
api_session.WriteMessageToReservationOutput(reservation_id, 'ending sandbox')
api_session.EndReservation(reservation_id)
This will end (i.e. 'fail') the sandbox (state won't be Active or ActiveWithError), and invoke theTeardown script. Hope this helps.
