So that's potentially an important thing to fix first.... will it be possible to apply this fix automatically, or will users need to do something by hand (like restart the worker). If the latter is the case, we might need to go in two steps (e.g. skip lock check in a first update, and only apply the full fix in a second step).

I believe they would need to manually restart the worker. My workers are dockerised and I had to hard redeploy them to get them working.

so, that probably means we will have to apply the fix in two steps if that's possible.

I guess I messed up the updating process... I neglected to test it since I thought that it would work in the same way as before. But now I realize that there is a fundamental distinction between the original adhoc locking implementation and the refactored one. The former one was sort of reentrant and so it had no issues with updating.

so, that probably means we will have to apply the fix in two steps if that's possible.

It is possible I guess.

PR1: this PR+comment out locking statement in the worker.
PR2: comment in locking statements.