[Gluster-users] Geo replication on slave not showing files in brick

[ML mail]

Yes my single slave node is a single brick. Here would be the output of the volume info just in case:

Volume Name: myslavevol
Type: Distribute
Volume ID: *REMOVED*
Status: Started
Number of Bricks: 1
Transport-type: tcp
Bricks:
Brick1: gfs1geo:/data/myslavevol/brick
Options Reconfigured:
nfs.disable: on

I tried to mount this volume using FUSE but even there I don't see the files. I only see the ".gfid" directory.


Now I noticed that I did not have ACL enabled on my underlying filesystem (ZFS). Saw that from this error message:

[2015-03-04 22:27:06.100996] W [posix.c:5534:init] 0-myslavevol-posix: Posix access control list is not supported.

So now I at least activated POSIX ACL but still the files are not here. Could it be related to the ACL? or nothing to do with that?



On Friday, March 6, 2015 2:22 PM, M S Vishwanath Bhat <msvbhat at gmail.com> wrote:

On 6 March 2015 at 14:27, ML mail <mlnospam at yahoo.com> wrote:

Hello,
>
>
>I just setup geo replication from a 2 node master cluster to a 1 node slave cluster and so far it worked well. I just have one issue on my slave if I check the files on my brick i just see the following:
>
>
>drwxr-xr-x  2 root root 15 Mar  5 23:13 .gfid
>drw------- 20 root root 21 Mar  5 23:13 .glusterfs
>
>
>there should be around 10 files on that volume but it's simply not there. If I list the files in the ".gfid" directory I can see that the files are there but just have a horrible name made out of numbers.
>
>
>I don't see any special errors or warnings on the slave except maybe this warning which might be related:
>
>
>[2015-03-05 22:19:06.233956] W [glusterd-op-sm.c:3312:glusterd_op_modify_op_ctx] 0-management: Failed uuid to hostname conversion
>[2015-03-05 22:19:06.233974] W [glusterd-op-sm.c:3404:glusterd_op_modify_op_ctx] 0-management: op_ctx modification failed
>
>
>
>So my question, is this the expected behaviour? or should be able to see the files on my slave volume just like on my master volume?

You should see files in slave volume just like your master volume.


Looks like you check the files in slave bricks. Does your single node slave server has single brick? If it has more bricks, the files may be present in other brick as well.


Can you mount the slave volume and check for the files?


Best Regards,

Vishwanath



>
>Regards
>ML
>
>
>
>_______________________________________________
>Gluster-users mailing list
>Gluster-users at gluster.org
>http://www.gluster.org/mailman/listinfo/gluster-users
>
>