Bug 2145022 - rgw: reports of put ops recreating former bucket index objects after resharding
Summary: rgw: reports of put ops recreating former bucket index objects after resharding
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: Red Hat Ceph Storage
Classification: Red Hat Storage
Component: RGW
Version: 5.2
Hardware: Unspecified
OS: Unspecified
unspecified
high
Target Milestone: ---
: 5.3
Assignee: J. Eric Ivancich
QA Contact: Anuchaithra
Akash Raj
URL:
Whiteboard:
Depends On:
Blocks: 2126049
TreeView+ depends on / blocked
 
Reported: 2022-11-22 23:04 UTC by J. Eric Ivancich
Modified: 2023-09-19 04:30 UTC (History)
11 users (show)

Fixed In Version: ceph-16.2.10-78.el8cp
Doc Type: Bug Fix
Doc Text:
.PUT object operation writes to the correct bucket index shards Previously, due to a race condition, a PUT object operation would rarely write to a former bucket index shard. This caused the former bucket index shard to be recreated, and the object would not appear in the proper bucket index. Therefore, the object would not be listed when the bucket was listed. With this fix, care is taken to prevent various operations from creating bucket index shards and recover when the race condition is encountered. PUT object operations now always write to the correct bucket index shards.
Clone Of:
Environment:
Last Closed: 2023-01-11 17:42:24 UTC
Embargoed:


Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
Red Hat Issue Tracker RHCEPH-5672 0 None None None 2022-11-22 23:04:30 UTC
Red Hat Product Errata RHSA-2023:0076 0 None None None 2023-01-11 17:43:38 UTC

Description J. Eric Ivancich 2022-11-22 23:04:09 UTC
Description of problem:

On pacific we had reports of heavy PUT workloads recreating former bucket index objects. Although this has not been reproduced in upstream main or ceph 5.x, we believe that potential still exists.


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


How reproducible: Difficult


Steps to Reproduce:
1. On pacific was reproduced when under heavy PUT OBJ workload resharding would occur.
2. There's speculation that another worthwhile test would be a recursive bucket delete while same bucket under heavy PUT OBJ workload.
3.

Actual results:

Formerly deleted bucket index objects are recreated with data for newly PUT objects.

Expected results:

Newly PUT objects should appear in the bucket index objects created as a result of resharding.

Additional info:

Comment 1 RHEL Program Management 2022-11-22 23:04:21 UTC
Please specify the severity of this bug. Severity is defined here:
https://bugzilla.redhat.com/page.cgi?id=fields.html#bug_severity.

Comment 29 errata-xmlrpc 2023-01-11 17:42:24 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 (Moderate: Red Hat Ceph Storage 5.3 security update and Bug Fix), 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-2023:0076

Comment 30 Red Hat Bugzilla 2023-09-19 04:30:21 UTC
The needinfo request[s] on this closed bug have been removed as they have been unresolved for 120 days


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