OVB BMC failing to start properly

[Ben Nemec]

The BMC is failing to configure the network properly. This is probably related to the keystone v3 changes that recently went in to OVB.

[Ben Nemec (bnemec)]

This is because for some reason there are domains set in the environment:

export OS_USER_DOMAIN_ID=openstack-nodepool_domain

This is wrong with our keystone v2 cloud. I need to figure out where those are coming from. Previously they would have been ignored because we didn't support keystone v3 at all.

[Ben Nemec (bnemec)]

Bleh, it was a str_replace collision. I must have added that after I tested against keystone v2. Should be fixed by https://github.com/cybertron/openstack-virtual-baremetal/commit/409d01d9591b5501e3d7b5fea5b8051db38d4263

[Ben Nemec (bnemec)]

Okay, BMCs appear to be working again. Going to leave this open until all the broken jobs clear out of the system so people can figure out what's going on.

[Ronelle Landy (rlandy)]

OVB is working again on the downstream jobs.
Thanks for the quick fix.

[Ben Nemec (bnemec)]

Yeah, jobs appear to be passing again and this has been alerted in the #tripleo channel a few times, so hopefully the people affected have had a chance to see it. I'm going to go ahead and close the bug.

Apologies for the trouble. I've switched my test cloud back to keystone v2 to hopefully avoid this happening again.