[Bug 1840348] Re: Sharded OpWQ drops suicide_grace after waiting for work

[Kellen Renshaw]

@dbungert - Taking over this one for hillpd, we would like to pursue
this backport to Bionic/Queens (dropping Stein).

Originally held back on it since there isn't a viable reproducer. Now,
since the change has been in upstream since May 1, 2020, there is
confidence that it doesn't introduce a regression.

** Changed in: cloud-archive/stein
       Status: In Progress => Won't Fix

** Changed in: ceph (Ubuntu Bionic)
     Assignee: Dan Hill (hillpd) => Kellen Renshaw (krenshaw)

** Changed in: cloud-archive/queens
     Assignee: Dan Hill (hillpd) => Kellen Renshaw (krenshaw)

-- 
You received this bug notification because you are a member of Ubuntu
Sponsors Team, which is subscribed to the bug report.
https://bugs.launchpad.net/bugs/1840348

Title:
  Sharded OpWQ drops suicide_grace after waiting for work

Status in Ubuntu Cloud Archive:
  Fix Released
Status in Ubuntu Cloud Archive queens series:
  In Progress
Status in Ubuntu Cloud Archive rocky series:
  Won't Fix
Status in Ubuntu Cloud Archive stein series:
  Won't Fix
Status in Ubuntu Cloud Archive train series:
  Fix Released
Status in ceph package in Ubuntu:
  Fix Released
Status in ceph source package in Bionic:
  In Progress
Status in ceph source package in Eoan:
  Won't Fix
Status in ceph source package in Focal:
  Fix Released

Bug description:
  [Impact]
  The Sharded OpWQ will opportunistically wait for more work when processing an empty queue. While waiting, the heartbeat timeout and suicide_grace values are modified. The `threadpool_default_timeout` grace is left applied and suicide_grace is disabled.

  After finding work, the original work queue grace/suicide_grace values
  are not re-applied. This can result in hung operations that do not
  trigger an OSD suicide recovery.

  The missing suicide recovery was observed on Luminous 12.2.11. The
  environment was consistently hitting a known authentication race
  condition (issue#37778 [0]) due to repeated OSD service restarts on a
  node exhibiting MCEs from a faulty DIMM.

  The auth race condition would stall pg operations. In some cases, the
  hung ops would persist for hours without suicide recovery.

  [Test Case]
  I have not identified a reliable reproducer. Currently testing the fix by exercising I/O. 

  Recommend letting this bake upstream before considering a back-port.

  [Regression Potential]
  This fix improves suicide_grace coverage of the Sharded OpWq.

  This change is made in a critical code path that drives client I/O. An
  OSD suicide will trigger a service restart and repeated restarts
  (flapping) will adversely impact cluster performance.

  The fix mitigates risk by keeping the applied suicide_grace value
  consistent with the value applied before entering
  `OSD::ShardedOpWQ::_process()`. The fix is also restricted to the
  empty queue edge-case that drops the suicide_grace timeout. The
  suicide_grace value is only re-applied when work is found after
  waiting on an empty queue.

  - In-Progress -
  Opened upstream tracker for issue#45076 [1] and fix pr#34575 [2]

  [0] https://tracker.ceph.com/issues/37778
  [1] https://tracker.ceph.com/issues/45076
  [2] https://github.com/ceph/ceph/pull/34575

To manage notifications about this bug go to:
https://bugs.launchpad.net/cloud-archive/+bug/1840348/+subscriptions