Once we installed build-blocker-plugin 1.7.1, workflow jobs hung trying to allocate a node:

Started by user Builder
Running: Allocate node : Start
Still waiting to schedule task
Waiting for next available executor on node1.local

When the workflow job hangs like this, not only is the step hung, but other unrelated jobs will queue up even if they aren't trying to run on the same node, i.e. if my workflow job depends on label 'feature' then even jobs which do not require 'feature' and would not build on a machine my workflow job would build on are also hung up.

The jenkins log contains the following for each hung job:

SEVERE: Timer task hudson.model.Queue$MaintainTask@6bb9a704 failed
java.lang.ClassCastException: org.jenkinsci.plugins.workflow.support.steps.ExecutorStepExecution$PlaceholderTask cannot be cast to hudson.model.AbstractProject
at hudson.plugins.buildblocker.BuildBlockerQueueTaskDispatcher.getBuildBlockerProperty(BuildBlockerQueueTaskDispatcher.java:199)
at hudson.plugins.buildblocker.BuildBlockerQueueTaskDispatcher.canTake(BuildBlockerQueueTaskDispatcher.java:108)
at hudson.model.Queue$JobOffer.canTake(Queue.java:299)
at hudson.model.Queue.maintain(Queue.java:1355)
at hudson.model.Queue$MaintainTask.doRun(Queue.java:2450)
at hudson.triggers.SafeTimerTask.run(SafeTimerTask.java:51)
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471)
at java.util.concurrent.FutureTask.runAndReset(FutureTask.java:304)
at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$301(ScheduledThreadPoolExecutor.java:178)
at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:293)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
at java.lang.Thread.run(Thread.java:745)

Downgrading from build-blocker-plugin 1.7.1 to (in my case) 1.6 solved the problem.