I would like to start my comment by thanking all of you in your contribution to design/development of Jenkins and Jenkins AccuRev plugin and to your contribution to the OpenSource community overall.
In my organization, we have setup multiple Jenkins jobs and most of them integrate with AccuRev plugin to download the code and subsequently build the project. Some of the jobs, especially the ones that run from Linux Servers consistently fail in the last step with the error reported in this JENKINS-56882 ticket. The job does what is expected. Retrieves the code from AccuRev, and performs all the subsequent tasks. However at the very end, it reports failure. We have been ignoring these failed jobs since the jobs accomplish the expected activities.
With this bug, it will be difficult to know whether the job actually failed, or it is showing a failed status due to this bug. This bug will be an hindrance to take this automated process to upper tiers in controlled environments (TEST / PROD) where the IT will not know the correct status of the job, and unable provide adequate & timely support.
Complete automation is impossible due to this bug as it does not allow for correct exit status tracking. We consider this issue critical and precluding our progress with Jenkins and AccuRev plugin
What is the roadmap to fix this bug ? Your response and attention to fix this issue is much appreciated.
Thanks, Johns
I encountered this issue using Jenkins 2.150.3, Accurev plugin 0.7.20, and Ant plugin 1.9. The issue traced back in my case to RunWithSCM.super.getCulprits returning null when it shouldn't be able to. I was able to roll back to 0.7.13 of the Accurev plugin and the issue went away.