After the bounded wait (30 seconds), the scheduler is always shut down without waiting more (shutdown(false)). Waiting for the jobs in shutdown(true) has no limit, and a job fired just before the standby may not be listed as running yet, which could block the stop of XWiki.
The wait and the jobs still running after it are logged through the org.xwiki.shutdown logger (INFO while waiting, WARN after the wait), which the default logging configuration shows.
A job still running after the wait is not interrupted, as before: it keeps running while the components are disposed.
The wait delays the other stop listeners and the dispose of the components by up to 30 seconds.
The restore of the jobs at startup checks the stop before each job.
The "Scheduler is shutting down" message, which is now logged at every stop, is an INFO instead of a WARN.
This message was sent by Atlassian Jira (v9.3.0#930000-sha1:287aeb6)
If image attachments aren't displayed, see this article.
