Details
-
Bug
-
Status: Open (View Workflow)
-
Major
-
Resolution: Unresolved
-
None
-
Windows 10
Jenkins 2.73.1
Description
After Jenkins is restarted for any reason, ongoing jobs go into a stale state.
Job will try to resume, but get stuck displaying:
{{Resuming build at Mon Nov 27 17:25:51 IST 2017 after Jenkins restart }}
Ready to run at Mon Nov 27 17:26:00 IST 2017
Nothing further will happen.
The stage where this happens (python pytest) reads:
lock('server-gpu') {
bat '''
call activate py35
call conda env list
call python.exe -m pytest --junitxml=test_results_nightly.xml
'''
}
Not sure if this has to do with the lock. The job acquired the lock before the restart and after the restart, other jobs I start wait for the lock so it seems to have gotten it.
Any idea for a solution or workaround (which is not completely manual) would be appreciated.
Attachments
Activity
shachar ilan
created issue -
shachar ilan
made changes -
| Field | Original Value | New Value |
|---|---|---|
| Description |
After Jenkins is restarted for any reason, ongoing jobs go into a stale state. Job will try to resume, but get stuck displaying: {{Resuming build at Mon Nov 27 17:25:51 IST 2017 after Jenkins restart }} {{Ready to run at Mon Nov 27 17:26:00 IST 2017}} Nothing further will happen. The stage where this happens (python pytest) reads: {code:java} lock('server-gpu') { bat ''' call activate py35 call conda env list call python.exe -m pytest --junitxml=test_results_nightly.xml ''' }{code} Not sure if this has to do with the lock. The job acquired the lock before the restart and after the restart, other jobs I start wait for the lock so it seems to have gotten it. Any idea for a solution or workaround (which is not completely manual) would be appreciated. |
After Jenkins is restarted for any reason, ongoing jobs go into a stale state. Job will try to resume, but get stuck displaying: \{\{Resuming build at Mon Nov 27 17:25:51 IST 2017 after Jenkins restart }} {{Ready to run at Mon Nov 27 17:26:00 IST 2017}} Nothing further will happen. The stage where this happens (python pytest) reads: {code:java} lock('server-gpu') { bat ''' call activate py35 call conda env list call python.exe -m pytest --junitxml=test_results_nightly.xml ''' }{code} Not sure if this has to do with the lock. The job acquired the lock before the restart and after the restart, other jobs I start wait for the lock so it seems to have gotten it. Any idea for a solution or workaround (which is not completely manual) would be appreciated. |
| Component/s | lockable-resources-plugin [ 18222 ] | |
| Component/s | pipeline [ 21692 ] |
I'm almost 100% sure that this is related to the lockable resources plugin. I created a small sample pipeline:
timestamps { echo 'Starting' lock('lockable-resources-test') { echo 'Got lock' } echo 'Done' }When reserving the lock from some other job before starting this job - neither this nor the other job would continue after the reboot. Even after killing the other job freeing up the lock this job would not continue.