[Gluster-users] replace brick with gluster 3.6

[Pranith Kumar Karampuri]

On 03/30/2015 05:05 PM, Ml Ml wrote:
> Thanks for your quick reply.
>
> This is the output of my remaining healty peer:
>
> getfattr -d -m. -e hex /brick/raidvolb/data/
> getfattr: Removing leading '/' from absolute path names
> # file: brick/raidvolb/data/
> trusted.gfid=0x00000000000000000000000000000001
> trusted.glusterfs.dht=0x000000010000000000000000ffffffff
> trusted.glusterfs.volume-id=0x8786357b9d114c01a34baee949c116e9
On the new brick execute:
setfattr -n trusted.glusterfs.volume-id -v 
0x8786357b9d114c01a34baee949c116e9 <new-brick-path>

Before bringing glusterd, new brick up, do the following:
 From the mount point execute 'mkdir <non-existent-dir-name>; rmdir 
<same-dir-you-created-before>; setfattr -n trusted.abc -v def 
<gluster-mount-path>; setfattr -x trusted.abc <gluster-mount-path>'

Could you please post the output at this point of all the bricks? i.e. 
both the healthy bricks and old bricks? I would like to check that there 
are no mistakes in the operations you performed above.

Once I confirm things are fine, do:
start glusterd on the machine with new brick and self-heal should 
automatically start.

These steps are from latest document: http://review.gluster.com/8503
Let me know if you have any doubts.

I will try to get this in as soon as possible.

Pranith
>
>
>
>
> On Mon, Mar 30, 2015 at 12:38 PM, Pranith Kumar Karampuri
> <pkarampu at redhat.com> wrote:
>> On 03/30/2015 03:59 PM, Ml Ml wrote:
>>> Anyone?
>>>
>>> Is this a dumb question or just a hard one?
>>> I already tried:
>>>
>>> http://www.gluster.org/community/documentation/index.php/Gluster_3.4:_Brick_Restoration_-_Replace_Crashed_Server
>>>
>>> but i got stuck with the setfattr command.
>>>
>>> So i was wondering if this is the way to go?
>> could you paste output of getfattr -d -m. -e hex
>> <any-of-the-other-bricks-in-replication>.
>>
>> Pranith
>>>
>>>
>>> On Thu, Mar 26, 2015 at 10:31 PM, Ml Ml <mliebherr99 at googlemail.com>
>>> wrote:
>>>> Hello List,
>>>>
>>>> i have a 3 Peer Replica Gluster. On one of my peers the hard drive of
>>>> a brick failed.
>>>> I replaced it and formated the brick device it with ext4.
>>>>
>>>> How do i get it back into my gluster? Is there a official way how to
>>>> re-integrade it?
>>>>
>>>> Thanks,
>>>> Mario
>>> _______________________________________________
>>> Gluster-users mailing list
>>> Gluster-users at gluster.org
>>> http://www.gluster.org/mailman/listinfo/gluster-users
>>