Mike,
Hi. I'm seeing a behavior in DAGMan in 6.8.0 that I never saw in 6.6.10
(our previously installed version). Every once in a while, a DAG will
get put on hold automatically, for no apparent reason. After some
digging in the logs, I see this in the DAG's dagman.out log:
9/6 12:37:41 BAD EVENT: job (4952.0.0) ended, submit count < 1 (0)
9/6 12:37:41 BAD EVENT is warning only
9/6 12:37:41 ERROR "Assertion ERROR on (job->_queuedNodeJobProcs >=
0)" at line 608 in file dag.C
And this in the submit machine's SchedLog:
9/6 12:37:41 (pid:14159) (4561.0) Problem parsing user policy for
job: The UNKNOWN (never set) OnExitRemove _expression_ '' evaluated to
UNDEFINED. Putting job on hold.
9/6 12:37:41 (pid:14159) Job 4561.0 put on hold: The UNKNOWN (never
set) OnExitRemove _expression_ '' evaluated to UNDEFINED
When the DAG job is released, it seems to continue on just fine. Is
this a bug in 6.8.0? I can send complete logs for the DAG and the
submit machine to a developer (~400k) if that'd be helpful.
I don't think we've seen this before.
Could you please send me the complete dagman.out file, and the log file(s)
for all of the node jobs?
Kent Wenger
Condor Team
_______________________________________________
Condor-users mailing list
To unsubscribe, send a message to condor-users-request@xxxxxxxxxxx with a
subject: Unsubscribe
You can also unsubscribe by visiting
https://lists.cs.wisc.edu/mailman/listinfo/condor-users
The archives can be found at either
https://lists.cs.wisc.edu/archive/condor-users/
http://www.opencondor.org/spaces/viewmailarchive.action?key=CONDOR