Describe the issue:
As per: https://bugzilla.redhat.com/show_bug.cgi?id=1318230#c18 -- GCE persistent storage will not mount if the node's internal name (i.e. the openshift_hostname variable in the ansible installer) is not equal to the GCE instance name ( the name of the VM from the perspective of GCE ). If a cluster is installed using a fully-qualified domain name as the internal hostname (which can happen by setting the variable as such or sometimes incidentally if the variable is not set), then disks will not mount -- and changing the node name post-installation is not extremely simple.
Suggestions for improvement:
Either in this doc or in the Advanced Install doc, it needs to be made clear that this variable must be set properly.
I belive this was resolved by https://github.com/openshift/openshift-docs/pull/4044
the current docs include this:
> the nodeName must match the instance name in GCE in order for the cloud provider integration to work properly. The name must also be RFC1123 compliant.
Which I guess is the intent here.
@stwalter can you please confirm?
Agreed, that should cover it
Thanks Steven, I am closing this now.