Timeouts are back today, with multiple flows returning 524. Flows also getting stuck in a queued state
1 Like
Hi @hanley
There was some backlog in the queue because it was throttling jobs more aggressively than it should, the issue is now fixed they should be resumed
1 Like
Thanks for the quick fix!
Still having issues this morning European time: lots of runs stuck in running or paused stage for hours
We’ve had runs queueing for long periods of time all week. Is the goal on your end to eventually upgrade the infrastructure so there’s no queuing? Or is queueing just a new reality of this platform? Some of our flows are time-sensitive, and essentially work like an API in which the caller expects a timely response. But the queuing breaks that functionality.
Hi @hanley
If the flow has been queuing for a week, then it’s a bug and stuck, please shoot an email with the run link to support@activepieces.com
The current queue takes from couple of seconds to couple of minutes (maximum)
Each queue isn’t lasting a week. I mean that our flows started regularly queueing this week.
We’ve been using AP for about a year now and until recently have been able to rely on fast execution without queuing. Is queueing (from a few seconds up to a couple minutes) the new norm now?
Now getting this issue: Code Block Failing? - #3 by gsj
Anyone else?
524s are back again yesterday and today. Call a webhook flow, and that flows’ run log indicates that it finished executing quickly and responded. But caller never actually receives response, instead receiving a 524.
One example: This flow (Activepieces) calls this flow in Step 6 (Activepieces) today, Oct 10, at 12:18 pm.
Called flow indicates that it executed and responded in about 1 second. But calling flow shows that step took 2 minutes and 5 seconds with a 524.