Azure Pipelines scaler consuming too many ADO resources (TSTUs) #5968
Unanswered
sgtnuggets
asked this question in
General
Replies: 0 comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
At 10pm Pacific on 7/15/2024 we saw a huge spike in error rates in our Keda pods because of client timeouts when hitting the Azure DevOps API. Investigation led us to see that the service account behind the personal access token we're using for the Keda integration was accumulating a massive amount of TSTUs (an accumulated metric for API usage in Azure DevOps). This cased the delay in API response time which causes the client timeouts. We haven't changed anything on our end and have been using version 2.14.0 for over a week now with no issue so I'm confident that the issue isn't in Keda itself. I'm suspecting it's related to a change on Azure DevOps side but would like to hear if any other users of this scaler are seeing similar behavior.
Beta Was this translation helpful? Give feedback.
All reactions