I have just stumbled onto this same issue, with the following setup:
MultiPhase Job A
Phase 1
Job1
Job2
Job3
Job4
Phase 2
Deploy
MultiPhase Job A runs on master (4 executors), Job1-4 run on
label-restricted slaves.
Repro:
I kick off 4 concurrent builds of Job A. The master node has 4 executors
so all multi-phase instances of job A are kicked off (no queued instances of
A). As 4 instances of multi-phase job A were kicked off, there are now 4
instances of each Job1-4 that need to be built, for this explanation I'll
focus on Job1.
There are only 2 slaves (each with 1 executor) that can build Job1, however
there are 4 instances of Job1 that need to be built. The first two
instances of Job1 are sent to the available 2 slaves. The third instance of
Job1 is queued, waiting for one of the two slaves to be available. THE BUG:
The 4th instance of Job1 does not get queued, it simply gets skipped (gray
circle next to it) with the text "(not built"). Further, "Phase 2" gets
executed on this instance of JobA, which fails because Job1 was never built
and hence the artifact to be deployed simply doesn't exist.
This appears to be an issue using the MultiPhase job plugin + "Allow
Concurrent Builds" + "restrict where this job can be run", and is a blocking
issue for myself and my team. Any insight into a fix for this would be
greatly appreciated.
Thanks!
--
View this message in context:
http://jenkins-ci.361315.n4.nabble.com/Jenkins-silently-skipping-MultiJob-phase-tp4696480p4715982.html
Sent from the Jenkins users mailing list archive at Nabble.com.
--
You received this message because you are subscribed to the Google Groups
"Jenkins Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to jenkinsci-users+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
