yahoo-eng-team team mailing list archive

[Takashi NATSUME <natsume.takashi@xxxxxxxxxxxxx>]

** Also affects: nova/rocky
   Importance: Undecided
       Status: New

** Changed in: nova/rocky
     Assignee: (unassigned) => Takashi NATSUME (natsume-takashi)

** Changed in: nova/rocky
   Importance: Undecided => Medium

-- 
You received this bug notification because you are a member of Yahoo!
Engineering Team, which is subscribed to OpenStack Compute (nova).
https://bugs.launchpad.net/bugs/1816938

Title:
  Misleading log message "Booting with blank volume" in nova-compute
  when booting from real volume

Status in OpenStack Compute (nova):
  In Progress
Status in OpenStack Compute (nova) rocky series:
  New
Status in OpenStack Compute (nova) stein series:
  In Progress

Bug description:
  Something is broken here:

  https://github.com/openstack/nova/blob/af78b13c24d4abf393d17ac57e9135204ef12b73/nova/virt/block_device.py#L836

  Because I'm seeing this getting logged:

  https://github.com/openstack/nova/blob/af78b13c24d4abf393d17ac57e9135204ef12b73/nova/virt/block_device.py#L855

  When I know I'm booting from an existing volume. This is also all over
  the CI logs:

  http://logstash.openstack.org/#dashboard/file/logstash.json?query=message%3A%5C%22Booting%20with%20blank%20volume%20at%5C%22%20AND%20tags%3A%5C%22screen-n-cpu.txt%5C%22&from=7d

  My guess is something with dict get() access changed on the
  DriverVolumeBlockDevice code with this change:

  https://github.com/openstack/nova/commit/b958bf1126aea8b88ccebb43a330fc1a44717145
  #diff-40dadeaa854473fb72fa4bf3491a434f

  If I change the code to check:

  if bdm.volume_id:
     ...

  Then I get the expected log message but that is probably dangerous if
  we have a non-volume BDM.

To manage notifications about this bug go to:
https://bugs.launchpad.net/nova/+bug/1816938/+subscriptions