Red Hat Bugzilla – Bug 1476212
[geo-rep]: few of the self healed hardlinks on master did not sync to slave
Last modified: 2017-08-21 09:42:02 EDT
Description of problem:
In the following scenario, the sync of hardlinks do not happen to slave.
1. Create geo-rep between master and slave
2. Mount the volume
3. Create a file (file1)
4. Let the file sync to slave
5. kill one set of replica for a subvolume containing file1
6. create a hardlink of file1 (ln file1 file2).=> Ensure that the file2 hashes to the same subvolume of file1
7. Start the master volume forcefully to heal file2 . Wait for heal to happen
8. Kill the other set of the replica (than the step 5)
9. Start the geo-replication
In the above scenario the hardlinks are not synced to slave and there are no errors.
Step 1 to Step 5 remains same
6. create a hardlink of file1 (ln file1 file2).=> Ensure that the file2 hashes to the different subvolume of file1
Step 7 to Step 8 remains same
In this scenario, sync happens as follows:
a. If both the bricks active are (selfhealed bricks) which has recoreded MKNOD. Sync happens.
b. If the self healed brick containing MKNOD for sticky bit file becomes PASSIVE, hardlinks are not synced.
Version-Release number of selected component (if applicable):
Always with the above steps.
--- Additional comment from Worker Ant on 2017-07-26 08:24:07 EDT ---
REVIEW: https://review.gluster.org/17880 (geo-rep: Fix syncing of self healed hardlinks) posted (#1) for review on master by Kotresh HR (firstname.lastname@example.org)
REVIEW: https://review.gluster.org/17907 (geo-rep: Fix syncing of self healed hardlinks) posted (#1) for review on release-3.10 by Kotresh HR (email@example.com)
COMMIT: https://review.gluster.org/17907 committed in release-3.10 by Shyamsundar Ranganathan (firstname.lastname@example.org)
Author: Kotresh HR <email@example.com>
Date: Wed Jul 26 08:09:31 2017 -0400
geo-rep: Fix syncing of self healed hardlinks
In a distribute replicate volume, if the hardlinks
are created when a subvolume is down, it gets
healed from other subvolume when it comes up.
If this subvolume becomes ACTIVE in geo-rep
there are chances that those hardlinks won't
be synced to slave.
AFR can't detect hardlinks during self heal.
It just create those files using mknod and
the same is recorded in changelog. Geo-rep
processes these mknod and ignores it as
it finds gfid already on slave.
Geo-rep should process the mknod as link
if the gfid already exists on slave.
> Change-Id: I2f721b462b38a74c60e1df261662db4b99b32057
> BUG: 1475308
> Signed-off-by: Kotresh HR <firstname.lastname@example.org>
> Reviewed-on: https://review.gluster.org/17880
> Smoke: Gluster Build System <email@example.com>
> CentOS-regression: Gluster Build System <firstname.lastname@example.org>
> Reviewed-by: Aravinda VK <email@example.com>
(cherry picked from commit d685e4238fafba8f58bf01174c79cb5ca35203e5)
Signed-off-by: Kotresh HR <firstname.lastname@example.org>
Smoke: Gluster Build System <email@example.com>
CentOS-regression: Gluster Build System <firstname.lastname@example.org>
Reviewed-by: Aravinda VK <email@example.com>
This bug is getting closed because a release has been made available that should address the reported issue. In case the problem is still not fixed with glusterfs-3.10.5, please open a new bug report.
glusterfs-3.10.5 has been announced on the Gluster mailinglists , packages for several distributions should become available in the near future. Keep an eye on the Gluster Users mailinglist  and the update infrastructure for your distribution.