-
Bug
-
Resolution: Unresolved
-
Minor
-
None
Sometimes in backup-and-restore scenarios I see in the Jenkins log
WARNING o.c.j.p.k.p.PodTemplateStepExecution#onResume: Unable to inject task listener java.io.IOException: cannot start writing logs to a finished node StepStartNode[id=..., exec=CpsFlowExecution[Owner[...]]] podTemplate in CpsFlowExecution[Owner[...]] at org.jenkinsci.plugins.workflow.support.DefaultStepContext.getListener(DefaultStepContext.java:124) at org.jenkinsci.plugins.workflow.support.DefaultStepContext.get(DefaultStepContext.java:87) at org.csanchez.jenkins.plugins.kubernetes.pipeline.PodTemplateStepExecution.onResume(PodTemplateStepExecution.java:230) at org.jenkinsci.plugins.workflow.flow.FlowExecutionList$ItemListenerImpl$1.onSuccess(FlowExecutionList.java:184) at org.jenkinsci.plugins.workflow.flow.FlowExecutionList$ItemListenerImpl$1.onSuccess(FlowExecutionList.java:179) at ... at org.jenkinsci.plugins.workflow.flow.FlowExecutionList$ItemListenerImpl.onLoaded(FlowExecutionList.java:179) at ...
I think this happens after a broken build of some kind. Could easily be suppressed in the kubernetes plugin, though I wonder if it means something weirder is going on inside Pipeline—why is onResume being called on a step that already completed?