Description of problem: After a failed baremetal install, the operator must destroy the cluster in order to retry the install. Unfortunately, destroying the cluster in ACM leaves the bootstrap vm running on the baremetal provisioning host. E.g., after destroying the "crc" cluster in our ACM 2.1.4 deployment, on the provisioning host I see: [root@neu-15-29 ~]# virsh list Id Name State ------------------------------------- 2 crc-mh2lk-bootstrap running This will cause problems in subsequent installs because the bootstrap vm is configured with the API VIP for the cluster, so having an old bootstrap vm lying around will prevent a new bootstrap vm from being configured correctly. Version-Release number of selected component (if applicable): ACM 2.1.4
Hey Lars! To be clear, you're not talking about ACM's Hub cluster. You're talking about a baremetal cluster that you wanted to manage. Is that correct? Thanks! Nathan
This is concerning the installation of a Bare Metal cluster, not ACM's Hub; reassigning to Cluster Lifecycle for triage.
G2Bsync 862667538 comment chrisahl Wed, 16 Jun 2021 19:50:51 UTC G2Bsync This will be fixed in 2.3.0