Default Execution time out
In the docs it say that all serverless endpoints have a 10 min default execution time out. We have had few instances that the job is stuck in processing for hours. Are the docs incorrect and we need to set the execution timeout manually?
4 Replies
@Milad Are you still experiencing this, if so - can you share an endpoint id?
Here is one of the endpoints that had this issue 5i4k4qb4m0a6if, the job was in progress for way longer than 10 mins (We have added manual execution timeout since then). It looks like it was picked up by a worker that was still initializing and was stuck on loading the docker image. Here is an example Request ID: 99ac96a1-c1cc-4b93-abe0-d28f59ed8d7f-u1
Thank you! I'll have someone look, in the meantime we identified a delay for users using 1.7.2 of the RunPod SDK. At the very least starting there would probably help.
FYI, I just checked and that endpoint is using 1.7.7