yahoo-eng-team team mailing list archive

[Dan Genin <daniel.genin@xxxxxxxxxx>]

This appears to be a duplicate of
https://bugs.launchpad.net/nova/+bug/1094134, which was fixed in

commit 02ea0f9f9e5c7f022b465a96ba3a4f089c633bee
Merge: cd2008c 9d3f524
Author: Jenkins <jenkins@xxxxxxxxxxxxxxxxxxxx>
Date: Fri Jan 11 23:35:10 2013 +0000

Merge "Correct the calculating of disk size when using lvm disk
backend."


** Changed in: nova
       Status: Triaged => Fix Released

-- 
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/1198831

Title:
  Compute service wrong report disk usage when  libvirt_images_type is
  LVM

Status in OpenStack Compute (Nova):
  Fix Released

Bug description:
  Heloo,

  I obsereved wrong compute node disk usage when I setup
  libvirt_images_type to lvm.  The compute service still reports
  filesystem usage (instances_path) instead of volume group usage
  (libvirt_images_volume_group).

  Fo exmample:
    instances_path ~= 50GB
    libvirt_images_volume_group = foo_vg, where foo_vg ~= 1000GB
    create few Instances (6GB root volume + 300GB ephemeral),
    create a few volumes in same VG (by cinder or nova-volumes of 580GB size)

  The status shows:
  ---------------------
      local_gb_used = OK
      local_gb = Wrong
      free_disk_gb = Wrong
      disk_available_least = Wrong

  select s.availability_zone,c.local_gb,c.local_gb_used,c.free_disk_gb,c.disk_available_least from compute_nodes as c, services as s where s.id=c.service_id and s.availability_zone='foo_av_zone';
  +-------------------+----------+---------------+--------------+----------------------+
  | availability_zone | local_gb | local_gb_used | free_disk_gb | disk_available_least |
  +-------------------+----------+---------------+--------------+----------------------+
  | some_av_zone      |       49 |           306 |         -257 |                   49 |
  +-------------------+----------+---------------+--------------+----------------------+

  Instead of:
  ------------
      local_gb_used = OK (ephemeral + root volume)
      local_gb = OK (whole space)
      free_disk_gb =  OK (free = local_gb - ephemeral - root volume)
      disk_available_least = OK (free space in vg disk_available_least = local_gb - ephemeral - root volume - volume)

  select s.availability_zone,c.local_gb,c.local_gb_used,c.free_disk_gb,c.disk_available_least from compute_nodes as c, services as s where s.id=c.service_id and s.availability_zone='foo_av_zone';
  +-------------------+----------+---------------+--------------+----------------------+
  | availability_zone | local_gb | local_gb_used | free_disk_gb | disk_available_least |
  +-------------------+----------+---------------+--------------+----------------------+
  | foo_av_zone       |     1003 |           306 |          697 |                  117 |
  +-------------------+----------+---------------+--------------+----------------------+

  I attached a patch which "fix" the reporting.

To manage notifications about this bug go to:
https://bugs.launchpad.net/nova/+bug/1198831/+subscriptions