Unfortunately, the issue is not 100% reproducible but here is a scenario in which I've seen it happening.

1. Create a long running (up to an hour) pipeline job.
2. Set disableResume on the job.
3. Initiate a build.
4. Restart the Jenkins container (high memory/CPU pressure might be a factor here as well).
5. There is a chance the build is trying to resume with the following indication:

Failing build: shutting down master and build is marked to not resume 
Resuming build at Wed Sep 23 19:14:28 UTC 2020 after Jenkins restart

Such builds are also stuck. They do not produce any more output and cannot be killed via UI. They can only be aborted via Script Console.