KEDA frequent reconciling causing ApiServer LIST verb Saturation

[djsly]

Report

Keda operator goes into a "restart" loop causing a constant reconciliation loop. Due to the non cache LIST operation the apiserver was not able to answer LIST request for other services as well.

Expected Behavior

Keda Operator should use cached List operation and it should also not get into reconciliation loop

Actual Behavior

Graph showing the apiserver saturation time

Steps to Reproduce the Problem

1. Hard to tell, it started after Keda Operator lost it's leader election due to a slow ApiServer response : https://github.com/operator-framework/operator-sdk/issues/5186

Logs from KEDA operator

keda-pods-apiserver-list-saturation.csv

KEDA Version

2.6.1

Kubernetes Version

1.23

Platform

Microsoft Azure

Scaler Details

Prometheus

Anything else?

Azure Support: 2209160030000259 for the impact and severity of such a bug

[zroubalik]

Hi, this is probably a duplicate of https://github.com/kedacore/keda/issues/2914

This issue has been solved most likely. By chance, could you please update KEDA version to the latest one?

[djsly]

we certainly can, what do you mean "by chance" ?

[zroubalik]

we certainly can, what do you mean "by chance" ?

the original sentece was missing a . - I hope that it makes sense now ;)

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed in 7 days if no further activity occurs. Thank you for your contributions.

[stale[bot]]

This issue has been automatically closed due to inactivity.