[Issue #513] Script freezes unexpectedly with no error written to log. Manual pause/resume brings it back

Hi, a potential fixed has been found :smiley:

=> Please test the latest UI.Vision V5.0.5.

Technically V5.0.5 is exactly the same as V5.0.3 but with the issue fixed.

@admin For firefox this bug was solved too ?

I do not find any news version of ui.vision oin firefox addons page

Thanks

Thanks @admin. We do several runs in a day where we encounter the freezing so we’ll test with V5.0.5 BETA and provide feedback.

@newuserkantu The Firefox version is now available, too.

1 Like

Hi, I hope this will definitely fix the problem, thank you very much for your reactivity and for the way you take care of the users !
Louis

Thank you! Initial test runs show improvement. I’ve been able to run my macro longer than before, though I did have an error output that may point to an additional issue. I think a problem has been solved, but perhaps not all problems that contribute to this behavior. Will continue to monitor…

Seems like this bug actually mutated into [Fixed] ipcPromise: onAsk timeout 3000 for cmd "RUN_COMMAND"

@commensal No, that is something different (I replied in the ipcPromise thread). The error this thread is/was about is if the macro suddenly stops with no error message.

Hi @admin. Thank you once again for the fix. Just want to report that we’ve had ZERO issues related to this particular bug (freezing for no reason) since the release of v5.0.5 last June 19th.

We do several runs a day as explained in this earlier post. So it’s a pretty good test case for this particular scenario.

Seeing that no one else has posted an issue since the release of v5.0.5, it does look like this is finally solved! Thank you again.

I’m kinda curious what the issue was, if you’re willing to share. Keep up the great development work.

Thanks for the great feedback, and thanks again to everyone for their patience while we where hunting down this issue. Overall, it was a collection of several issues which is one of the reasons that made this bug so tricky. And in fact, we are still aware of an rare issue with the OPEN command, and will be working on this, too.

But the key issue was that a random ID was not really random and unique, and so we had sometimes the same inter-process communication ID even when they should have been different.

2 Likes