I’ve encountered an issue with the scheduling trigger. When I set it to run every 30 seconds, it works fine. However, if I set it to anything above 60 seconds, it still triggers every minute. For instance, setting it to 120 seconds will still cause it to trigger every 60 seconds, always at the 00 second mark (e.g., 15:21:00, 15:22:00).
This issue occurs only on the self-hosted version 1.51.2. The cloud version 1.50.1 does not have this problem.
P.S. I just downgraded to version 1.50.1, and the issue is not occurring. It seems like a bug in version 1.51.2.
This is interesting. There were no inherent changes made to the schedule trigger node between these versions. I’m now updating one of my instances to test this out.
Meantime, could you share how you’re hosting your instance? Just to help reduce the number of variables
And also, does this happen only on actie executions or test ones as well?
So, we updated our setup to use Cron consistently everywhere. Which is an issue now because Cron format does not support “every X seconds/minutes” for a value of X above 59.
We need to either
figure out a way to normalize the cron expression, and possibly register a group of crons to make it act like a duration > 59
revert the code for seconds duration to use a timer instead of a cron expression, or
start enforcing a max value of seconds/minutes duration to 59 in the UI as well.
The last one sounds semantically correct, but it’ll also be a breaking change.
It’s possible that we won’t get this fixed tomorrow, but I’ll bring this up with the team to see if we can find a quick solution for this.
until then, maybe you could try changing the duration to 59 seconds or 2 minutes
No updates yet, Typically once fixed we post an update on the thread to make you aware. In this case have you not been able to work around it by setting it to run every 2 minutes instead which would do the same thing?
