stable-to-next specs fail with missing memcached relation on n-c-c

[Ryan Beisner]

stable-to-next specs fail with missing memcached relation on n-c-c

#### 2019 07 16

Several mojo jobs were at > 10hrs run time in OSCI, with stuck volumes.
Manually canceled the jobs and cleaned up the instances and volumes on:

## SV13 - A mojo job was held in blocked state for 10+hrs due to missing memcache relation to n-c-c. (a) Why didn't the job die after it waited for more than a reasonable amount of time?; and (b) We need to update the bundle that that spec is using.
    Build #19166 specs/full_stack/stable_to_next_ha xenial-ocata (Jul 16, 2019 7:09:25 AM)
    http://10.245.162.58:8080/job/mojo_runner/19166/

## SV04 - Same issue as SV13.
    Build #19168 specs/full_stack/stable_to_next_ha xenial-newton (Jul 16, 2019 7:18:28 AM)
    http://10.245.162.58:8080/job/mojo_runner/19168/

## SV02 - Same issue as SV13.
    Build #19167 specs/full_stack/stable_to_next_ha bionic-queens (Jul 16, 2019 7:16:54 AM)
    http://10.245.162.58:8080/job/mojo_runner/19167/

## SV07 - Same issue as SV13.
    Build #19165 specs/full_stack/stable_to_next_ha xenial-pike (Jul 16, 2019 6:55:48 AM)
    http://10.245.162.58:8080/job/mojo_runner/19165/

[Ryan Beisner (1chb1n)]

I am disabling this spec in OSCI, with reference to this bug. When it is resolved, please re-enable the job. I just don't want to waste cloud compute cycles on a spec that won't pass, and is a huge deployment. Thank you!

[Ryan Beisner (1chb1n)]

Um, this is already identified and proposed:

https://github.com/openstack-charmers/openstack-mojo-specs/pull/116