-Atin
Sent from one plus one
On Aug 4, 2015 10:27 PM, "shacky" <shacky83@xxxxxxxxx> wrote:
>
> Thank you Mathieu for your answer!
>
> 2015-08-03 20:24 GMT+02:00 Mathieu Chateau <mathieu.chateau@xxxxxxx>:
> > date is not the same but is content different ?
>
> Yes, the content is different.
> The file on the node1 is the updated one, the other two nodes had the
> old version.
>
> > You may have disable the mtime attribue to get better perf ?
>
> No, I just have a "default" configuration for that volume, I had not
> disable the mtime attribute.
>
> > What are these 2 GFID ?
> > You can use this script to find who they are:
> > https://gist.github.com/semiosis/4392640
>
> Thank you so much.
>
> I tried to delete that file on the node1, and I verified that it was
> deleted on node2 and node3 too.
> So I copied the correct file on node1 and it correctly replicated on
> node2 and node3.
>
> So now the problem seems to be solved, but..... How I could monitor
> this? I realized that just by chance!
>
> Another question: That volume is mounted in /mnt/data and the real
> data are located in /data.
> So every nodes are mounting the data on localhost:data.
> If I write some files in /mnt/data they are correctly replicated, but
> if I write them in /data not of course
> Is there a way to avoid this? I want /data to be accessible to Gluster
> only, not to real users!
Creating data from back end is definitely not recommended as features like replication heavily depends on the external attributes set on the file and having data created directly from back end wouldn't have the xattrs set.
>
> Thank you very much for your help!
> _______________________________________________
> Gluster-users mailing list
> Gluster-users@xxxxxxxxxxx
> http://www.gluster.org/mailman/listinfo/gluster-users
_______________________________________________ Gluster-users mailing list Gluster-users@xxxxxxxxxxx http://www.gluster.org/mailman/listinfo/gluster-users
