Project

General

Profile

Actions
Copy link

Bug #64017

closed

osd/scrub: 'max_osd_scrub' default of '1' is too low

Added by Ronen Friedman 4 months ago. Updated about 1 month ago.

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

0%

Source:
Tags:
backport_processed
Backport:
reef,quincy
Regression:
No
Severity:
3 - minor
Reviewed:
Affected Versions:
ceph-qa-suite:
Pull request ID:
51669
Crash signature (v1):
Crash signature (v2):

Description

Note: this is a 'back-dating' bug report, to be used as a basis for back-porting PR#51669.

The default configuration of only one concurrent scrub (as either primary or replica)
per OSD, is too conservative. Based on multiple complaints, that value results in
customers' clusters taking much too long to complete a scrub of all PGs.

Related issues 3 (1 open — 2 closed)

Related to RADOS - Bug #55009: Scrubbing exits due to error reading object headFix Under ReviewRonen Friedman

Actions
Copied to Ceph - Backport #64018: quincy: osd/scrub: 'max_osd_scrub' default of '1' is too lowResolvedRonen FriedmanActions
Copied to Ceph - Backport #64019: reef: osd/scrub: 'max_osd_scrub' default of '1' is too lowResolvedRonen FriedmanActions
Actions
Copy link

Also available in: Atom PDF