Encountered a few times 1-16-23. Recovery went fine (and is faster with version 3.5.0 than with version 3.3.1, as expected) but this should still be avoided! If there's really no way to prevent the user from closing the transcript (shell) window, then maybe it's better to not use a shell window and instead figure out a different quick-and-easy way to make the transcript (including tracebacks) accessible.
Usually, the operator doesn't need to open the transcript window except to view tracebacks, so this isn't high priority, and arguably not a bug.
Encountered a few times 1-16-23. Recovery went fine (and is faster with version 3.5.0 than with version 3.3.1, as expected) but this should still be avoided! If there's really no way to prevent the user from closing the transcript (shell) window, then maybe it's better to not use a shell window and instead figure out a different quick-and-easy way to make the transcript (including tracebacks) accessible.
Usually, the operator doesn't need to open the transcript window except to view tracebacks, so this isn't high priority, and arguably not a bug.