Hey Prefect Team,
On Friday (2024/11/08) around 5pm (AEST) I've had multiple runs unable to set their status state. This is not a call i'm making myself, rather the prefect cloud backend doing so. I've had a similar issue a month ago, see here: https://prefect-community.slack.com/archives/CL09KU1K7/p1727829012698459
This time is different as this was a 500 error occurring at the end of successful runs - indicating that prefect was attempting to change the state to 'success'. Also these errors occurring at the end of the run meant that my runs began to run indefinitely. Attached is an example of this failed run.
I am wanting to:
1. Raise this if this hasn't already been caught.
2. Wondering how this occurring and how it can be prevented in future now that this has happened a few times. When this happens, it's breaking our ETL processes, which our business is relying on.
Thanks!
r
Rob Freedy
11/12/2024, 4:13 PM
Hey @Sam Lawler!! Based on the error log you shared it looks like this could have coincided with an incident Prefect Cloud was having. For more information on the incident, you can subscribe to updates on our status page here.
s
Sam Lawler
11/13/2024, 3:24 AM
Ah, i was unaware of this status page. Thanks for linking me to this.
Bring your towel and join one of the fastest growing data communities. Welcome to our second-generation open source orchestration platform, a completely rethought approach to dataflow automation.