I'm running a Kubernetes/Docker job on Rundeck 2.11.5-1. My job looks like:
Job
Sub Job 1
Remote Command (kubectl run command)
On error
Sub Job
Sub Job 2
Remote Command (kubectl run command)
On error
Sub Job
The problem I'm having is that if Sub Job 1 fails, its "On Error" does not trigger and Sub Job 2 runs as if all was well.
Is there something that kubectl needs to return to indicate there was an error?
What are some things I should look for/do to cause my job to stop on error.
I found the solution. Add --restart=Never
to the kubectl statement and the Rundeck job now correctly terminates on an apps failure.