[Gluster-users] how to deal with data when a peer restart

[Joe Julian]

Yes, gluster does handle this. /How/ it handles this I don't have time 
to write this morning. As for the source code, look at the afr 
translator (xlators/cluster/afr/src).

On 03/19/2016 09:26 AM, 袁仲 wrote:
>
> I have two peers with two bricks on each of them, and each two 
> bricks(on different peer) make up a replicate volume, and two 
> replicate volumes make up a distribute volume.
>
> I mount the volume on the client,and start a vm(virtual machine).
>
>
> when a peer restart (shutdown -r now). and my question is:
>
>  1.  how does glusterfs deal with the data on the peer, and does the
>     “restart” operation would trigger some process of data healing or
>     some process to keep the data (on different peers with a same
>     replicate volume) identical.
>  2. if it does, and can you show me a function in the source code?
>
>
> thanks,
>
> yz
>
>
>
> _______________________________________________
> Gluster-users mailing list
> Gluster-users at gluster.org
> http://www.gluster.org/mailman/listinfo/gluster-users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.gluster.org/pipermail/gluster-users/attachments/20160319/4dd9db5d/attachment.html>