If an OutOfMemoryError or some other error of similar severity occurs, currently the queue threads will just abort, and I think an error will display in the server log.
As always, when OOMs start happening there's no guarantee that Connect (or any application) is going to be able to gracefully recover, especially if it causes RAM corruption.
However we can make a best-effort attempt by catching any Throwable and continuing to let the queue thread run, making sure to still send an error to the server log and possibly also an alertable error event.
If an OutOfMemoryError or some other error of similar severity occurs, currently the queue threads will just abort, and I think an error will display in the server log.
As always, when OOMs start happening there's no guarantee that Connect (or any application) is going to be able to gracefully recover, especially if it causes RAM corruption.
However we can make a best-effort attempt by catching any Throwable and continuing to let the queue thread run, making sure to still send an error to the server log and possibly also an alertable error event.
Imported Issue. Original Details: Jira Issue Key: MIRTH-4335 Reporter: narupley Created: 2018-10-23T08:08:28.000-0700