Hello!
I've an issue that is impacting our wkfs: we noticed that some delivery activities are running for serveral minutes before being sent to customers. The delivery templates are email templates with barely any personalisation (first name and some other basic personalisation).
Here can you see a delivery with 800 customers that took 6 min to process. How can I see what's causing this? Is there a way to see the logs of a delivery in a wkf? When checking under "Journal" I don't get any additional info:
In the delivery logs I cannot see anything strange either. It takes 1 min and not 6 min as in the wkf.
Start of logs:
End of logs:
How can I troubleshoot and see what's causing this slowness?
Solved! Go to Solution.
Topics help categorize Community content and increase your ability to discover relevant content.
Views
Replies
Total Likes
I tried to delete all the delivery nodes in the wkf and create new ones. So far it's working, didn't encounter the same bug yet.
Another problem/bug I encountered maybe related to this is that some deliveries have been triggered and sent twice to the same target group despite the wkf has been just triggered once:
Here can you see that the wkf gets triggered a first time through a signal:
Then it gets a warning out of nowhere and it is automatically resumed causing maybe this double send-out:
There are like 5 other deliveries and the other 4 do not get triggered twice, just this delivery2264 did.
Views
Replies
Total Likes
Honestly, when we have seen some of the warnings like the one you refer to there, we often mitigate simply by duplicating the workflow and running that instead. No clue why. If that does not help, my suggestion would be to pause elements one by one, and monitor when error dissapears, to narrow down to what activity is the issue.
in terms of the long delivery time, I don't have any take on that, maybe other in this forum could help with that 🙂
Hope it helps!
.
I tried to delete all the delivery nodes in the wkf and create new ones. So far it's working, didn't encounter the same bug yet.