Flows cancelled by monitor
I have some scheduled flows that consistenly get cancelled by monitor because "was hanging in between 2 steps. ". I can't figure out why it breaks, if I run the flow manualy, it gets the job done without problem.
I can't reproduce with a minimal flow, but I can send my actual flows in private, if neeeded
Version: CE v1.555.0
OS: Debian GNU/Linux 11 (bullseye) x86_64
CPU: Intel Xeon (4) @ 2.800GHz
Memory: 12834MiB / 23544MiB
3 Replies
the error
We will need a minimal reproduction to investigate. There are no difference between flows ran manually or not so maybe it's just a question of frequencny
Try to see if you can reproduce consistently with a very frequent schedule and replace step by step what you can with fake logic
and sending the flow and sending the last state before you have this error might be helpful as well as the worker logs of the worker that handle the last correct transition at the time it did the transition
I think that i found the culprit
InternalError: SqlErr: error returned from database: could not extend file "base/16384/16455": No space left on device @jobs.rs:4229:5
one of the flows the got the error throw this error
I'll investigate more, but I think that it
thnks!