Project

General

Profile

Fix #57040

osd: Update osd's IOPS capacity using async Context completion instead of cond wait.

Added by Sridhar Seshasayee about 1 year ago. Updated 11 months ago.

Status:
Resolved
Priority:
Normal
Category:
-
Target version:
-
% Done:

0%

Source:
Tags:
backport_processed
Backport:
quincy
Reviewed:
Affected Versions:
ceph-qa-suite:
Component(RADOS):
OSD
Pull request ID:
Crash signature (v1):
Crash signature (v2):

Description

The method, OSD::mon_cmd_set_config(), sets a config option related to
mClock during OSD boot-up. The method waits on a condition variable
until the mon acks the command. This is generally not a problem. But
there could be scenarios where monitor could be slow to respond, or due
to a flaky network, response could be delayed. The OSD could therefore
be blocked from booting-up. To avoid this, the conditional wait can be
replaced with an async Context completion.

Moreover, persisting this in the monitor store is not very critical. An
existing fallback mechanism stores this value in the in-memory "values"
map of the config subsystem. This can be read by the OSD at any point
during its operation.

The issue of the OSDs being blocked from booting-up properly was
observed when running tests with failure injections during OSD boot-up.


Related issues

Copied to RADOS - Backport #57443: quincy: osd: Update osd's IOPS capacity using async Context completion instead of cond wait. Resolved

History

#1 Updated by Sridhar Seshasayee about 1 year ago

  • Status changed from New to Fix Under Review
  • Backport set to quincy
  • Pull request ID set to 47456

#2 Updated by Sridhar Seshasayee about 1 year ago

  • Status changed from Fix Under Review to Pending Backport

#3 Updated by Backport Bot about 1 year ago

  • Copied to Backport #57443: quincy: osd: Update osd's IOPS capacity using async Context completion instead of cond wait. added

#4 Updated by Backport Bot about 1 year ago

  • Tags set to backport_processed

#5 Updated by Sridhar Seshasayee 11 months ago

  • Status changed from Pending Backport to Resolved

Also available in: Atom PDF