[pve-devel] pve-qemu-kvm: enable glusterfs backend

[Alexandre DERUMIER]

>>For example, a reboot of any gluster server member will cause a
>>resync of all dirty files. This re-read the whole file to compute checksums
>>(like rsync -c).
>> 
>>But it seems nobody else think that is a problem - why?

Don't known, but for me, glusterfs was not created some year ago for hosting vm files.
So, redhat try to improve this step by step.

Maybe this is why they are working on BD xlator using lvm-dm-thin as backend ?



BTW, do you you plan to upgrade to qemu 1.6 soon ?



----- Mail original ----- 

De: "Dietmar Maurer" <dietmar at proxmox.com> 
À: "Alexandre DERUMIER" <aderumier at odiso.com> 
Cc: pve-devel at pve.proxmox.com 
Envoyé: Mardi 13 Août 2013 09:22:03 
Objet: RE: [pve-devel] pve-qemu-kvm: enable glusterfs backend 

> I can't build pve-qemu-kvm with theses packages.... (maybe some missing 
> headers files in glusterfs-common package ?) 

Ah - OK. 

I already committed your glusterfs patches - seems to work good so far. 

The only problem I see it that re-sync large files can take a long time, 
but seems that is expected behavior. 

For example, a reboot of any gluster server member will cause a 
resync of all dirty files. This re-read the whole file to compute checksums 
(like rsync -c). 

But it seems nobody else think that is a problem - why?