[Bug 1842656] [NEW] Bitmap allocator returns duplicate entries

[Eric Desrochers]

Public bug reported:

There is a known issue where ceph commit latency can increase over time
[0], and the "fix" is to move to the new Nautilus bitmap allocator back-
ported in 12.2.12 [1].

It has been brought to my attention that one encountered a bug with the
nautilus bitmap allocator [2] that brought down their cluster. This was
due to a double-extent allocation which has been fixed and back-ported
to Luminous, but is targeted for 12.2.13.

Would be good to add it into our Ubuntu ceph package(s) where it applies
to get ahead of the op commit latency problem.

The bug is reported upstream and being discussed here [3] [4].

[0] http://tracker.ceph.com/issues/38738 
[1] https://www.mail-archive.com/ceph-users@lists.ceph.com/msg53793.html 
[2] http://lists.ceph.com/pipermail/ceph-users-ceph.com/2019-July/035799.html 
[3] https://tracker.ceph.com/issues/40080 
[4] https://github.com/ceph/ceph/pull/28644

** Affects: ceph (Ubuntu)
     Importance: Undecided
         Status: New

** Affects: ceph (Ubuntu Bionic)
     Importance: Undecided
         Status: New


** Tags: sts

** Tags added: sts

** Also affects: ceph (Ubuntu Bionic)
   Importance: Undecided
       Status: New

-- 
You received this bug notification because you are a member of Ubuntu
OpenStack, which is subscribed to ceph in Ubuntu.
https://bugs.launchpad.net/bugs/1842656

Title:
  Bitmap allocator returns duplicate entries

Status in ceph package in Ubuntu:
  New
Status in ceph source package in Bionic:
  New

Bug description:
  There is a known issue where ceph commit latency can increase over
  time [0], and the "fix" is to move to the new Nautilus bitmap
  allocator back-ported in 12.2.12 [1].

  It has been brought to my attention that one encountered a bug with
  the nautilus bitmap allocator [2] that brought down their cluster.
  This was due to a double-extent allocation which has been fixed and
  back-ported to Luminous, but is targeted for 12.2.13.

  Would be good to add it into our Ubuntu ceph package(s) where it
  applies to get ahead of the op commit latency problem.

  The bug is reported upstream and being discussed here [3] [4].

  [0] http://tracker.ceph.com/issues/38738 
  [1] https://www.mail-archive.com/ceph-users@lists.ceph.com/msg53793.html 
  [2] http://lists.ceph.com/pipermail/ceph-users-ceph.com/2019-July/035799.html 
  [3] https://tracker.ceph.com/issues/40080 
  [4] https://github.com/ceph/ceph/pull/28644

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/ceph/+bug/1842656/+subscriptions