DNS Rollout Failure During Upgrade after changing Gateway Mode in RHOCP 4
Issue
- DNS daemonset fails to communicate with Kubernetes API during upgrade
- Rollout of cluster operators, specifically DNS operator, gets stuck
- Error message indicates "i/o timeout" when trying to reach API
- Switching back to
routingViaHost: trueallows successful rollout and upgrade
Environment
- Red Hat OpenShift Container Platform (RHOCP) 4
- OVNKubernetes Shared Gateway Mode (
routingViaHost: false) - Default route not via the
br-exinterface
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.