Project

General

Profile

Documentation #23354

doc: osd_op_queue & osd_op_queue_cut_off

Added by Konstantin Shalygin about 6 years ago. Updated about 3 years ago.

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

0%

Tags:
Backport:
Reviewed:
Affected Versions:
Pull request ID:

Description

In docs:

osd_op_queue default is: `prio`. Real is `wpq`. So this is a doc's bug.

If I understand properly: if osd_op_queue `wpq`, osd_op_queue_cut_off should be `high`. Real is `low`. This negative impact for queue or all as designed?

History

#1 Updated by Patrick Donnelly almost 6 years ago

  • Tracker changed from Support to Documentation
  • Project changed from Ceph to RADOS
  • Subject changed from osd_op_queue & osd_op_queue_cut_off to doc: osd_op_queue & osd_op_queue_cut_off
  • Category deleted (OSD)
  • Target version deleted (v12.2.4)
  • Release deleted (luminous)
  • Affected Versions deleted (v12.2.2, v12.2.3)

#2 Updated by Anthony D'Atri over 3 years ago

The default value for `osd_op_queue_cutoff` changed to `high` with Octopus and was documented as such.

https://github.com/ceph/ceph/pull/30441/files

The default for `osd_op_queue` in `options.cc` appears to have `wpq` back at least as far as Luminous.

Luminous and Mimic `doc/rados/configuration/osd-config-ref.rst` claim `prio`; the docs were updated to `wpq` with the Nautilus release.

@zdover23 So I think there's nothing left to do here.

#3 Updated by Konstantin Shalygin about 3 years ago

  • Status changed from New to Resolved
  • Target version set to v16.0.0

Also available in: Atom PDF