(Description copied from JENKINS-5406. Comments there indicate that maybe the Perforce plugin is at fault here in not allowing the polling to happen while a build is in progress.)

Just ran into a problem with downstream-ext 1.3, Hudson 1.342, Perforce plugin 1.0.17.

I have a build X which I have set to trigger builds A, B, and C on success, but only if there are SCM changes. A, B, and C is the order that the downstream jobs are tested for SCM changes when I look at the logs.

The problem is that when build B is already running, the polling appears to hang for job B. After many minutes, I aborted the job, which generated in the logs:

FATAL: SCM polling aborted
java.lang.InterruptedException
	at java.lang.Object.wait(Native Method)
	at java.lang.Object.wait(Object.java:485)
	at hudson.slaves.WorkspaceList.acquire(WorkspaceList.java:173)
	at hudson.model.AbstractProject.pollSCMChanges(AbstractProject.java:1062)
	at hudson.plugins.downstream_ext.DownstreamDependency.shouldTriggerBuild(DownstreamDependency.java:38)
	at hudson.tasks.BuildTrigger.execute(BuildTrigger.java:187)
	at hudson.model.Build$RunnerImpl.cleanUp(Build.java:160)
	at hudson.model.Run.run(Run.java:1240)
	at hudson.model.FreeStyleBuild.run(FreeStyleBuild.java:46)
	at hudson.model.ResourceController.execute(ResourceController.java:88)
	at hudson.model.Executor.run(Executor.java:122)
Job-B has no SCM changes. Triggering skipped.

The polling for C then proceeded without incident. This happened on two consecutive builds, so that's enough for me to not call it a random failure.

Build B is a very long-running build... I haven't yet run into the situation where A or C were running at the time of the completion of X.