I attempted to duplicate the problem using a multibranch pipeline and was unsuccessful. My guess is that the issue is either somehow specific to the multibranch pipeline configuration you are using or is connected to an existing workspace from a prior build of the same job which is being reused by the pipeline job. If you run the same multibranch pipeline job in parallel, does the failure happen on all the jobs running in parallel, or only on the first job that was started? When multiple jobs run in parallel, Jenkins allocates a new workspace for the parallel jobs so that they will not collide with running jobs,. |