[sheepdog-users] About snapshots

[MORITA Kazutaka]

At Sat, 17 Aug 2013 09:14:44 +0200,
Ing. Luca Lazzeroni - Trend Servizi Srl wrote:
> 
> 
> Il giorno 17/ago/2013, alle ore 03:43, MORITA Kazutaka <morita.kazutaka at gmail.com> ha scritto:
> 
> > At Fri, 16 Aug 2013 17:03:54 +0800,
> > Liu Yuan wrote:
> >> 
> >> On Fri, Aug 16, 2013 at 10:48:17AM +0200, Ing. Luca Lazzeroni - Trend Servizi Srl wrote:
> >>> Hi,
> >>> I'm trying to use external snapshot feature of QEMU/libVirt with sheepdog baked vdi to take VDI live backups without using sheepdog snapshots.
> >>> This is to avoid the problem of disk-space unreleased on snapshot deletion.
> >> 
> >> Maybe you can simply use qemu-img sheepdog:image local_iamge to read the image
> >> out of cluster?
> >> 
> >> The release of unused objects from deleted snapshots will be supported in the
> >> up comming series of v0.7.x, I guess in about one month or two.
> > 
> > I don't think the object reclaim patches should go into stable-0.7.
> > The feature changes the underlying store layout and breaks backward
> > compatibility.
> 
> Does it mean that the whole cluster needs to be updated simultaneously or that we need to reformat it after install ?

No need to reformat it, but please update all the sheep daemons after
stopping sheepdog.  Their protocol has no compatibility between 0.6.x
and 0.7.

> Still the protocol be compatible with qemu block driver used in qemu 1.5 or we need to upgrade qemu too ? 

IIUC, sheepdog 0.7 is still compatible with qemu 1.5.

> 
> > 
> > I'll resend the series after rebasing and adding a document.  Testing
> > and reviewing would be greatly appreciated.  Those are the missing
> > things to merge the feature.
> > 
> 
> Ok, thanks.
> One question; did you implement a commit/rollback feature on snapshot deletion and space reclaim ?

What do you mean by "a commit/rollback feature"?

Thanks

Kazutaka