Gluster users - Jan 2016 - posix_acl_default [Invalid argument] issue with distributed geo-rep

Hello,

I just set up distributed geo-replication to a slave on my 2 nodes'
replicated volume and so far it works but I see every 60 seconds in the
slave's geo-replication-slaves gluster log file the following message:


[2016-01-31 17:38:48.027792] I [dict.c:473:dict_get]
(-->/usr/lib/x86_64-linux-gnu/glusterfs/3.7.6/xlator/system/posix-acl.so(posix_acl_setxattr_cbk+0x26)
[0x7f2334c5c166]
-->/usr/lib/x86_64-linux-gnu/glusterfs/3.7.6/xlator/system/posix-acl.so(handling_other_acl_related_xattr+0xb0)
[0x7f2334c5c0f0]
-->/usr/lib/x86_64-linux-gnu/libglusterfs.so.0(dict_get+0x93)
[0x7f233c04b0c3] ) 0-dict: !this || key=system.posix_acl_default [Invalid
argument]

The exact log file name is the following:
/var/log/glusterfs/geo-replication-slaves/d11ac2ca-439b-4b23-ba3a-18f3849f83ed:gluster%3A%2F%2F127.0.0.1%3Amyvolume-geo.gluster.log.

Because I have a lot of files this log file is already 16 MB big after not even
24 hours of running geo-replication. Does anyone understand what issue is? As
far as I understand it has to do with the file system's ACL. Let me detail
my setup below:

OS: Debian GNU/Linux 8.3
Gluster version: 3.7.6
Filesystem on master nodes: ZFS (with "acltype" parameter set to
"posixacl")
Filesystem on slave node: XFS


Volume info:

Type: Replicate
Volume ID: d11ac2ca-439b-4b23-ba3a-18f3849f83ed
Status: Started
Number of Bricks: 1 x 2 = 2
Transport-type: tcp
Bricks:
Brick1: gfs1a.domain.com:/data/myvolume/brick
Brick2: gfs1b.domain.com:/data/myvolume/brick
Options Reconfigured:
changelog.changelog: on
geo-replication.ignore-pid-check: on
geo-replication.indexing: on
performance.readdir-ahead: on
nfs.disable: on

Any ideas? I would be really interested to find a solution here and also make
sure my data is fine.

Best regards
ML

On 31/01/16 23:25, ML mail wrote:
> Hello,
>
> I just set up distributed geo-replication to a slave on my 2 nodes'
replicated volume and so far it works but I see every 60 seconds in the
slave's geo-replication-slaves gluster log file the following message:
>
>
> [2016-01-31 17:38:48.027792] I [dict.c:473:dict_get]
(-->/usr/lib/x86_64-linux-gnu/glusterfs/3.7.6/xlator/system/posix-acl.so(posix_acl_setxattr_cbk+0x26)
[0x7f2334c5c166]
-->/usr/lib/x86_64-linux-gnu/glusterfs/3.7.6/xlator/system/posix-acl.so(handling_other_acl_related_xattr+0xb0)
[0x7f2334c5c0f0]
-->/usr/lib/x86_64-linux-gnu/libglusterfs.so.0(dict_get+0x93)
[0x7f233c04b0c3] ) 0-dict: !this || key=system.posix_acl_default [Invalid
argument]
This is not an error to be afraid of. Above log can be poped up on every 
setxattr call.
You can track the fix in http://review.gluster.org/#/c/13325/
--
Jiffin
> The exact log file name is the following:
/var/log/glusterfs/geo-replication-slaves/d11ac2ca-439b-4b23-ba3a-18f3849f83ed:gluster%3A%2F%2F127.0.0.1%3Amyvolume-geo.gluster.log.
>
> Because I have a lot of files this log file is already 16 MB big after not
even 24 hours of running geo-replication. Does anyone understand what issue is?
As far as I understand it has to do with the file system's ACL. Let me
detail my setup below:
>
> OS: Debian GNU/Linux 8.3
> Gluster version: 3.7.6
> Filesystem on master nodes: ZFS (with "acltype" parameter set to
"posixacl")
> Filesystem on slave node: XFS
>
>
> Volume info:
>
> Type: Replicate
> Volume ID: d11ac2ca-439b-4b23-ba3a-18f3849f83ed
> Status: Started
> Number of Bricks: 1 x 2 = 2
> Transport-type: tcp
> Bricks:
> Brick1: gfs1a.domain.com:/data/myvolume/brick
> Brick2: gfs1b.domain.com:/data/myvolume/brick
> Options Reconfigured:
> changelog.changelog: on
> geo-replication.ignore-pid-check: on
> geo-replication.indexing: on
> performance.readdir-ahead: on
> nfs.disable: on
>
> Any ideas? I would be really interested to find a solution here and also
make sure my data is fine.
>
> Best regards
> ML
> _______________________________________________
> Gluster-users mailing list
> Gluster-users at gluster.org
> http://www.gluster.org/mailman/listinfo/gluster-users