Project

General

Profile

Actions
Copy link

Bug #55026

closed

multisite: expose config option for data sync's polling interval

Added by Casey Bodley about 2 years ago. Updated over 1 year ago.

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

0%

Source:
Tags:
Backport:
Regression:
No
Severity:
3 - minor
Reviewed:
Affected Versions:
ceph-qa-suite:
Pull request ID:
45784
Crash signature (v1):
Crash signature (v2):

Description

in rgw_data_sync.cc this interval is hard-coded as:

#define INCREMENTAL_INTERVAL 20
Actions
Copy link
 #1

Updated by Casey Bodley about 2 years ago

  • Status changed from New to Fix Under Review
  • Pull request ID set to 45784
Actions
Copy link
 #2

Updated by Casey Bodley about 2 years ago

  • Assignee set to Casey Bodley
Actions
Copy link
 #3

Updated by Casey Bodley almost 2 years ago

  • Status changed from Fix Under Review to Resolved
Actions
Copy link
 #4

Updated by Alexander Patrakov over 1 year ago

This looks like an important fix for the situation where the multi-site replication is going over a high-latency link, and the buckets to be replicated contain a lot of small objects. Can we have this backported to previous releases?

Right now I am looking at a pair of Pacific clusters that have 40 ms latency between them. One of them had an extended downtime, yet the catch-up process only uses ~50 Mbps out of 1 Gbps available. That's sad.

Actions
Copy link

Also available in: Atom PDF