-
Bug
-
Resolution: Fixed
-
Minor
-
1.625.x / 1.14.2
"Waiting to acquire /.../workspace/... : jenkins.util.Timer [#...]" id=... (0x...) state=WAITING cpu=75% - waiting on <0x...> (a hudson.slaves.WorkspaceList) - locked <0x...> (a hudson.slaves.WorkspaceList) at java.lang.Object.wait(Native Method) at java.lang.Object.wait(Object.java:502) at hudson.slaves.WorkspaceList.acquire(WorkspaceList.java:255) at hudson.slaves.WorkspaceList.acquire(WorkspaceList.java:234) at hudson.slaves.WorkspaceList.acquire(WorkspaceList.java:223) at org.jenkinsci.plugins.workflow.support.pickles.WorkspaceListLeasePickle$1.tryResolve(WorkspaceListLeasePickle.java:67) at org.jenkinsci.plugins.workflow.support.pickles.WorkspaceListLeasePickle$1.tryResolve(WorkspaceListLeasePickle.java:52) at org.jenkinsci.plugins.workflow.support.pickles.TryRepeatedly$1.run(TryRepeatedly.java:62) at ...
Looks like a workspace did not get relocked fast enough to avoid getting grabbed by some other job?
At a minimum, WorkspaceListLeasePickle should printWaitingMessage when acquire blocks, so it is clearer from the build log why the build is still stuck.
Possibly it should fail if it cannot acquire the workspace immediately, since in this case the workspace can be assumed to have already been clobbered by something else. Currently there is no such core API.
- blocks
-
JENKINS-26130 Print progress of pending pickles
-
- Resolved
-
- relates to
-
JENKINS-22767 AbstractLazyLoadRunMap.getById subject to race condition with .load
-
- Resolved
-
-
JENKINS-38994 Diagnose locker of a WorkspaceList.Lease
-
- Open
-
- links to
JENKINS-22767is one conceivable cause of the situation reported here: one copy of a build locked the workspace upon resumption, and soon after another copy was somehow produced and tried to run, but got stuck attempting to relock the same workspace.