Destroying and rebuilding odcs-stg first.

Getting stuck getting gluster to set itself back up after rebuild.

Trying a second time in staging now.

OK - odcs-stg is almost back up. The latest issue is that the ro nfs mount of /fedora_koji is failing for reasons I don't understand.

I see a very interesting ansible commit from back in January: b3df9abc8acd3e3bbb745a504ceb111ff0e18944 :)

Struggling now to get eth1_ip issues sorted out with staging still. prod seems to have gotten stood up just fine.

OK - I'm stumped. I did everything I could think of to rebuild odcs-stg like odcs-prod, but it still won't mount /fedora_koji. I'm stuck.

The nodes are rebuilt, but stg looks broken without its ro /fedora_koji mount. (And @cverna needs staging to work to test some new OSBS stuff).

I'm going to close this as done (since the nodes are technically rebuilt on f28) and open a new one tracking the nfs mount problem for @cverna's use case.