ProvisioningIP with disabled ProvisioningNetwork causing metal3 pods to fail
Issue
- After a cluster upgrade from 4.6.x to higher, the metal3 pods are in CrashLoopBackOff state.
- The machines are reporting 'Provisioning' state, even when nodes are ready and accessible right after the installation.
- The metal3 containers report following messages when all
metal3-image-cache
pod replicas are in Pending/CrashLoopBackOff state:
2022-05-09T17:41:52.259696150Z Waiting for <provisioning_ip> to be configured on an interface
2022-05-09T17:41:53.264638140Z Waiting for <provisioning_ip> to be configured on an interface
Environment
- Red Hat Openshift Container Platform [RHOCP]
- 4.7+
- BareMetal IPI
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.