Bug 2356050 - [7.1.z backport] osd/scrub: discard repair_oinfo_oid() [NEEDINFO]
Summary: [7.1.z backport] osd/scrub: discard repair_oinfo_oid()
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: Red Hat Ceph Storage
Classification: Red Hat Storage
Component: RADOS
Version: 7.1
Hardware: Unspecified
OS: Unspecified
unspecified
high
Target Milestone: ---
: 7.1z4
Assignee: Ronen Friedman
QA Contact: skanta
URL:
Whiteboard:
Depends On:
Blocks: 2356515
TreeView+ depends on / blocked
 
Reported: 2025-03-30 09:10 UTC by Alex
Modified: 2025-05-07 13:25 UTC (History)
13 users (show)

Fixed In Version: ceph-18.2.1-309.el9cp
Doc Type: Enhancement
Doc Text:
.PG scrub performance improved by removing unnecessary object ID repair check Previously, every PG scrub invoked `repair_oinfo_oid()`, a function meant to fix mismatched object IDs in rare cases linked to a historical filesystem bug. This added unnecessary overhead, as the check applied only under very specific conditions. With this enhancement, the check was removed, improving deep scrub performance by over 10%. Shallow scrubs are expected to benefit even more.
Clone Of:
: 2356515 (view as bug list)
Environment:
Last Closed: 2025-05-07 12:49:12 UTC
Embargoed:
skanta: needinfo? (rfriedma)


Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
Ceph Project Bug Tracker 18409 0 None None None 2025-03-30 09:10:01 UTC
Ceph Project Bug Tracker 20471 0 None None None 2025-03-30 09:10:01 UTC
Github ceph ceph pull 61590 0 None Merged osd/scrub: discard repair_oinfo_oid() 2025-03-30 09:10:01 UTC
Red Hat Issue Tracker RHCEPH-11000 0 None None None 2025-03-30 09:10:43 UTC
Red Hat Knowledge Base (Solution) 7114155 0 None None None 2025-04-01 14:20:14 UTC
Red Hat Product Errata RHSA-2025:4664 0 None None None 2025-05-07 12:49:23 UTC

Description Alex 2025-03-30 09:10:02 UTC
Description of problem:
PG not scrubbed in time
PG not deep-scrubbed in time

Version-Release number of selected component (if applicable):
RHCS 7.1

How reproducible:


Steps to Reproduce:
1.
2.
3.

Actual results:


Expected results:


Additional info:

Comment 1 Storage PM bot 2025-03-30 09:10:14 UTC
Please specify the severity of this bug. Severity is defined here:
https://bugzilla.redhat.com/page.cgi?id=fields.html#bug_severity.

Comment 7 Raimund Sacherer 2025-04-01 08:27:24 UTC
Hi, 

@Icemanny, @Ronen, here is a KCS to spread out scrubs correctly for RHCS 5/6 and RHCS 7+:
https://access.redhat.com/solutions/7109348

Ronen, the specifics of scrub scheduling changed in 7+, but it does not actually spread the scrubs evenly, should I open a BZ so we can investigate that? It's explained in the KCS above and I played around with it, it's rather difficult to get a more uniform scrub scheduling and it's very clustered (well, more like a bell curve).

BR
Raimund

Comment 15 errata-xmlrpc 2025-05-07 12:49:12 UTC
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory (Important: Red Hat Ceph Storage 7.1 security, bug fix, and enhancement updates), and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHSA-2025:4664


Note You need to log in before you can comment on or make changes to this bug.