Hi,
On Mon, Mar 13, 2017 at 9:54 PM, Sadegh Solati <solati.sadegh at
gmail.com>
wrote:
> Hi,
> I am completely new to gluster.
>
Welcome to Gluster :)
> I have 3 nodes pure replicate with about 600 GB data in small files about
> 20-60KB. I have some questions.
>
> 1- my bricks logs is filled with lines like this:
>
> [2017-03-13 16:08:18.484745] E [MSGID: 115050]
> [server-rpc-fops.c:156:server_lookup_cbk] 0-images-server: 49204346:
> LOOKUP /storage/2/f1c0064d53d7ee55338c9b471f5aa64f.jpg
> (fb60183f-8359-4e97-bd85-448d3ccef8da/f1c0064d53d7ee55338c9b471f5aa64f.jpg)
> ==> (No data available) [No data available]
> This file exist on all the nodes.
>
> is it related to gfid attribute on the file? I created the volume when the
> files existed on all 3 servers thus they don't have any gfid.
>
Yes it is because when lookup has happened, there was no gfid assigned with
the files. After you run the heal command the files will get gfids assigned
and when the next lookup happens it won't give those error messages in the
log.
>
> 2- I started a self-heal and it started successfully.
> "gluster volume heal images info" works, but these commands fail:
> gluster volume heal images info heal-failed
> gluster volume heal images info healed
> It gives me this error:
> *Gathering list of healed entries on volume images has been unsuccessful
> on bricks that are down. Please check if all brick processes are running.*
>
This is a known issue and there is a patch[1] for this, which addresses the
issue.
>
> All the nodes and bricks are up.
>
> I am using glusterfs 3.8.9 on ubuntu 14.04
>
> Any help is appreciated.
>
[1] https://review.gluster.org/#/c/15724/
Regards,
Karthik
>
> _______________________________________________
> Gluster-users mailing list
> Gluster-users at gluster.org
> http://lists.gluster.org/mailman/listinfo/gluster-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.gluster.org/pipermail/gluster-users/attachments/20170314/3e103f09/attachment.html>