Endpoints scaled down to zero because runpod thinks they are not in use.
You guys scaled down my workers on multiple endpoints with the following emails:
"Your endpoint has not received any requests for 30 days."
"Your endpoint has not received any requests for 149 days."
"Your endpoint *** has not received any requests for 205 days."
These numbers are NOT true. All of these endpoints are actively in use!
This must be a bug, right? I already made a ticket on runpod, just checking here if someone else has the same issue.
I even checked if runpod properly tracks my jobs in the metrics tab and everything looks totally fine there.
Please fix this.
5 Replies
I got the same issues, i just realized I had no workers on my endpoint
I had to manually reset the max number of workers , which was set to 0 back to my normal one
raised this to eng to make sure we validate the numbers, ty
@Unknown_User @Redfill The Canadien could you provide me your endpoint ids?
sent them to you via pm
i already made a support ticket and they said they gonna forward it to the escalation team
and i also manually added the workers back.
Sorry about that, we’re actively working on fixing the issue. For now, we’ve temporarily disabled the auto scale down system. I’ll let you know as soon as we’ve resolved it. @Unknown_User