Why iptables rules are not cleaned up by kubeproxy when the pods are terminated/rolling update?
Issue
- While performing the rolling update, kubeproxy does not clean the iptable rules for the old pods immediately and on average it takes ~10 minutes to clean up those rules.
Environment
- Openshift v3.11.318 and before
- Rolling strategy as below
rollingUpdate:
maxSurge: 1
maxUnavailable: 0
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.
