Troubleshooting an install, the user was unable to gather boostrap logs with the error: FATAL could not find outputs file "cluster.tfvars.json": stat cluster.tfvars.json: no such file or directory Bootstrap logs should be gatherable with just the bootstrap node; we shouldn't require control plane nodes. We have required masters since gather was created, so I might be missing something; but requiring control plane nodes could also be an assumption from AWS, where control plane nodes are created before the bootstrap node. The relevant failing call begins here: https://github.com/openshift/installer/blob/master/cmd/openshift-install/gather.go#L119 The ultimate cause of the failure I was troubleshooting was a bad pull secret. So this issue should be reproduceable when using a bad pull secret in Azure. The relevant slack thread is here: https://coreos.slack.com/archives/CHFF1LCUR/p1657178183932829
check with 4.12.0-0.nightly-2022-07-27-133042 no cluster.tfvars.json, Bootstrap logs should be gathered with just the bootstrap node.
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory (Moderate: OpenShift Container Platform 4.12.0 bug fix and security update), and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://access.redhat.com/errata/RHSA-2022:7399