Hi Jenkins community,
We have a pipeline that fails almost every time during the finally block when performing archiveArtifacts steps.
We're using Jenkins kubernetes plugin to provision jenkins-slaves automatically on kubernetes cluster.
The issue seems to occur when artifacts are not present, but since we're using "allowEmptyResults: true", we don't expect the job to fail.
There are mainly two exceptions in the log :
We don't know if they are related to each other or not, but they seem to happen together each time.
And then the master jenkins looses the connection to the slave, hence the pipeline is declared as a FAILURE.
We also suspect that this issue is related to the overall memory allocated to the jenkins slave, since we are noticing that there is a peak consumption on the jnlp slave at the very end, and we don't know to what it corresponds, since we're not doing much on the jnlp container (we're using another container for the build process)
Here's the code of the finally block :
and the pod specification :
Any ideas please ?