[PVE-User] HA migration behaviour vs. failures

[Dhaussy Alexandre]

I already hit a number of bugs with gluster.. I think they should perform more tests before releasing new "stable" versions.

Your experience doesn't mean much to me as long as you don't give any details.
I have done some benchmarks with iozone and glusterfs performances seemed quite honorable.

Envoyé de mon iPad

> Le 22 juil. 2014 à 21:24, "admin at extremeshok.com" <admin at extremeshok.com> a écrit :
> 
> From our experience with glusterfs, performance is terrible. Reliability was also questionable as we experienced random corruption.
> 
> I suggest you give ceph a try.
> 
> Sent from my iPhone
> 
>> On 22 Jul 2014, at 7:25 PM, Dhaussy Alexandre <ADhaussy at voyages-sncf.com> wrote:
>> 
>> 
>> 
>> Le 22/07/2014 18:56, Michael Rasmussen a écrit :
>>> On Tue, 22 Jul 2014 16:30:05 +0000
>>> Dhaussy Alexandre <ADhaussy at voyages-sncf.com> wrote:
>>> 
>>>> 
>>>> But i have some annoying issues with live migrations..
>>>> Sometimes it will work, but sometimes (with no reason) it won't.
>>>> When it fails (migration aborted), i try again then it works ! :(
>>> Could it be a saturation of network or an overburden storage?
>> 
>> I don't think so.
>> Very few load on both network and storage.
>> 
>> Network : dedicated vlan for glusterfs (for this 3 nodes only)
>> Storage : 15 vms (doing nothing) / 24 vdisks / 18 disks on 3 nodes 
>> (glusterfs with replica 3)
>> 
>> Glusterfs servers/clients and Proxmox on the same nodes.
>> 
>>> 
>>> 
>>> _______________________________________________
>>> pve-user mailing list
>>> pve-user at pve.proxmox.com
>>> http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
>> _______________________________________________
>> pve-user mailing list
>> pve-user at pve.proxmox.com
>> http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user