Project

General

Profile

Actions

Bug #53737

closed

Ceph_Multi-Site : Data added in primary RGW while secondary RGW service is down does not get synced when secondary service comes up

Added by Prashant Tambe over 2 years ago. Updated over 1 year ago.

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

0%

Source:
Community (user)
Tags:
Backport:
Regression:
No
Severity:
2 - major
Reviewed:
Affected Versions:
ceph-qa-suite:
Pull request ID:
Crash signature (v1):
Crash signature (v2):

Description

  • We have one RGW on each Site and RGW multi-site replication is set between them. Data added in primary gets replicated to secondary.
  • Primary side is read-write-delete enabled while secondary side is only read-only.
  • If we make secondary RGW service down for some minutes and keep adding data at the primary side while secondary service is down, following happens :
    • Data gets added into the primary side
    • When secondary service comes up, it does not get the data from primary side which was added while secondary service was down
    • If we add New data in primary(after secondary coming up), then this time only newly added data gets synced from primary to secondary and we get a sync status showing "Data is caught up with the source"
    • In this case, data added on primary while secondary being down, does not get written on secondary side after it comes up

Does anyone know why this happen? What could be the solution to this problem?

What is the full-proof way to tell that secondary side has really caught up with the master? Thanks in advance!


Related issues 1 (0 open1 closed)

Is duplicate of rgw - Bug #50785: multisite: full sync broken for tenanted bucketsResolvedCasey Bodley

Actions
Actions #1

Updated by Josh Durgin about 2 years ago

  • Project changed from Ceph to rgw
  • Category deleted (librbd)
Actions #2

Updated by Casey Bodley about 2 years ago

  • Assignee set to Casey Bodley
Actions #3

Updated by Mule Te about 2 years ago

I think this issue is fixed by this PR: https://github.com/ceph/ceph/pull/41316

But this PR is not backport to octopus. We have tested latest pacific and the issue is resolved.

@Casey Bodley Is that possible to backport this PR?

Thanks

Actions #4

Updated by Ilya Dryomov almost 2 years ago

  • Target version deleted (v15.2.16)
Actions #5

Updated by Casey Bodley over 1 year ago

  • Is duplicate of Bug #50785: multisite: full sync broken for tenanted buckets added
Actions #6

Updated by Casey Bodley over 1 year ago

  • Status changed from New to Duplicate
Actions

Also available in: Atom PDF