Bug 765572 (GLUSTER-3840)

Summary: Detect and repair root gfid when a volume is created from a former volume's parent dir
Product: [Community] GlusterFS Reporter: Joe Julian <joe>
Component: posixAssignee: Divya <divya>
Status: CLOSED CURRENTRELEASE QA Contact: Vijaykumar Koppad <vkoppad>
Severity: medium Docs Contact:
Priority: medium    
Version: 3.2.4CC: bbandari, gluster-bugs, vijay
Target Milestone: ---   
Target Release: ---   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Fixed In Version: glusterfs-3.5.0 Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2014-04-17 11:37:55 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: DP CRM:
Verified Versions: 3.3.0qa45 Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:

Description Joe Julian 2011-12-02 13:12:15 UTC 
The following was reported via IRC where the user had created a volume using /hekafs-export/www as the brick directory. Later, he removed that volume and used /hekafs-export as the brick directory. This led to an endless series of this error message:

Dec  2 01:32:44 hakafs01 GlusterFS[17549]: [2011-12-02 01:32:44.661161] C [inode.c:232:__is_dentry_cyclic] 0-posix-acl-autoload/inode: detected cyclic loop formation during inode linkage. inode (1/00000000-0000-0000-0000-000000000001) linking under itself as www

getfattr -m . -d -e hex /hekafs-export /hekafs-export/www
getfattr: Entferne führenden '/' von absoluten Pfadnamen
# file: hekafs-export
security.selinux=0x756e636f6e66696e65645f753a6f626a6563745f723a64656661756c745f743a733000
trusted.afr.plenty-client-2=0x000000000000000000000000
trusted.afr.plenty-client-3=0x000000000000000000000000
trusted.afr.plentyTest-client-2=0x000000000000000000000000
trusted.afr.plentyTest-client-3=0x000000000000000000000000
trusted.gfid=0x00000000000000000000000000000001
trusted.glusterfs.dht=0x00000001000000007fffffffffffffff
trusted.glusterfs.test=0x776f726b696e6700

# file: hekafs-export/www
security.selinux=0x73797374656d5f753a6f626a6563745f723a64656661756c745f743a733000
trusted.gfid=0x00000000000000000000000000000001
trusted.glusterfs.dht=0x00000001000000007fffffffffffffff
trusted.glusterfs.test=0x776f726b696e6700

The user did not understand the error message, nor was he able to deduce from the error what needed to be done to correct the problem. It seems like this should be able to be self-corrected.
Comment 1 Anand Avati 2011-12-26 05:41:17 UTC 
CHANGE: http://review.gluster.com/781 (extras: clean up a brick's gfid xattr) merged in master by Vijay Bellur (vijay)
Comment 2 Anand Avati 2012-01-05 17:18:45 UTC 
CHANGE: http://review.gluster.com/2514 (extras: add check for brick path existence) merged in master by Vijay Bellur (vijay)
Comment 3 Vijaykumar Koppad 2012-06-07 13:56:29 UTC 
This patch doesn't seem to do any behavioural changes. Its just a script clear_xattr.sh in the glusterFS source  which can be used to remove xattr from the bricks previously belong to some other volume. I think this should be documented properly. Moving it to DP .
Comment 4 Niels de Vos 2014-04-17 11:37:55 UTC 
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-3.5.0, please reopen this bug report.

glusterfs-3.5.0 has been announced on the Gluster Developers mailinglist [1], packages for several distributions should become available in the near future. Keep an eye on the Gluster Users mailinglist [2] and the update infrastructure for your distribution.

[1] http://thread.gmane.org/gmane.comp.file-systems.gluster.devel/6137
[2] http://thread.gmane.org/gmane.comp.file-systems.gluster.user