Project

General

Profile

Actions
Copy link

Bug #21042

closed

Cleanup stale bucket instances

Added by Robin Johnson over 6 years ago. Updated over 2 years ago.

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

0%

Source:
Tags:
Backport:
jewel
Regression:
No
Severity:
3 - minor
Reviewed:
Affected Versions:
ceph-qa-suite:
rgw
Pull request ID:
Crash signature (v1):
Crash signature (v2):

Description

This is for non-multisite/non-federated.

Legacy/large Ceph clusters end up with a LOT of bucket instances that don't point to any remaining buckets.
In Dreamhost production cluster, there are 24M bucket instances, but less than 100k buckets.

Need a plan that can clean them up.

  1. List all bucket instances
  2. List all buckets
  3. For each bucket instance:
    • Does ANY bucket still exist? If not, add to removal list.
    • Does the bucket point to this bucket instance still? If not, add to removal list.
  4. For each item in removal list
    • Purge bucket index shards.
    • Remove bucket index shards?
    • Remove bucket instance.
Actions
Copy link
 #1

Updated by Casey Bodley over 2 years ago

  • Status changed from New to Resolved

it sounds like `radosgw-admin reshard stale-instances list` will identify these, and `radosgw-admin reshard stale-instances rm` will clean them up

Actions
Copy link

Also available in: Atom PDF