Red Hat Bugzilla – Bug 1296176
geo-rep: hard-link rename issue on changelog replay
Last modified: 2016-06-23 01:01:26 EDT
+++ This bug was initially created as a clone of Bug #1296174 +++
Description of problem:
hard-link creation followed by a rename causes stale hard-links if changelogs are replayed
Version-Release number of selected component (if applicable):
Steps to Reproduce:
0. create and start geo-rep on a volume
1. touch f1; and wait for 20 seconds
2. ln f1 link1; mv link1 link2; and wait for 20 seconds
3. stop and start geo-rep
4. you will see link1 getting created on slave volume; ideally it shouldn't exist
link1 exists after logs are replayed
link1 shouldn't exist
Patch for this bug is available in rhgs-3.1.3 branch as part of rebase from upstream release-3.7.9.
Verified with the build: glusterfs-3.7.9-1
Carried the steps mentioned in the description and didn't see link1 be present at slave after geo-rep worker restarts.
[root@dj master]# touch f1; sleep 20 ; ln f1 link1 ; mv link1 link2 ; sleep 20
[root@dj master]# ls
[root@dj slave]# ls
After geo-rep restarts it starts history crawl and becomes changelog
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, and where to find the updated
files, follow the link below.
If the solution does not work for you, open a new bug report.