Azure Pipelines trigger failing with "Client.Timeout exceeded while awaiting headers"

[shayki5]

Report

I configured the Azure Pipelines trigger (with scaledJob) as explained here for few small pools in my organization and it worked well.
When I configured it to another pool that running many pipelines the scaling didn't work.
I checked Keda logs and saw:

{"error": "Get \"https://my-azure-devops/_apis/distributedtask/pools/52/jobrequests\": context deadline exceeded (Client.Timeout exceeded while awaiting headers)"}

When I tried to access the above api URL I succeeded to get results, but - it took 4 seconds because the result count is 760.

So I guess this is the issue and because it takes 4 seconds it failed with Client.Timeout exceeded.

Expected Behavior

Scaling up/down of the scaledJob according to the waiting pipelines.

Actual Behavior

Nothing happened, new Jobs wasn't created.

Steps to Reproduce the Problem

1. Configure scaledJob with Azure DevOps pool that running many pipelines.
2. Keda not auto scaling.

Logs from KEDA operator

{"error": "Get \"https://my-azure-devops/_apis/distributedtask/pools/52/jobrequests\": context deadline exceeded (Client.Timeout exceeded while awaiting headers)"}

KEDA Version

2.9.3

Kubernetes Version

< 1.23

Platform

Other

Scaler Details

Azure Pipelines

Anything else?

No response