RuntimeIOException from Jira has an amusing (and incorrect, IMO) message:
https://docs.atlassian.com/software/jira/docs/api/7.6.1/com/...
> An IOException was encountered and the stupid programmer didn't know
> how to recover, so this got thrown instead.
It's a misguided doc comment because "recovering" from error is usually the wrong thing to do - usually the right thing is to abort whatever action is taking place, whether it's a request handler, event loop or standalone program. Situations like low disk space, incorrect file permissions, missing files and so on usually can't be recovered deep in the stack or without manual intervention.