Running parallel steps in a pipeline, with failfast = true. If one branch fails with an error, or if the user aborts the task, the job hangs indefinitely and can only be removed by this sequence of actions:
Abort the job
Copy the job with another name
Delete the job (causes an error, but clearly has an effect)
Stop Jenkins
Delete the Job's workspace folder
Restart Jenkins.
this is very similar to issue 35213 which claims to have been resolved prior to Jenkins 2.13 and Pipeline 2.1, but clearly hasn't been resolved in all situations.

The following is a pipeline script that reproduces the bug 9 times out of 10. 1 time out of 10 it exits without a problem.

Attached is a screenshot of the console output of the hanging job.

The second attachment is the screenshot after the user attempts to abort the job. Clicking on the link "Click here to forcibly terminate running steps" has no effect, simply nothing happens.

stage concurrency: 5, name: 'Lock_Workflow'
node
{

parallel(
Branch1:

{ bat "ping localhost -n 11" }

,

Branch2:

{ bat "ping localhost -n 2" throw new Exception() }

,

failFast: true

)
}