Mailing-List: contact cloudstack-users-help@incubator.apache.org; run by ezmlm
Precedence: bulk
Reply-To: cloudstack-users@incubator.apache.org
Received-SPF: pass (nike.apache.org: local policy)
Date: Thu, 25 Oct 2012 00:22:25 +0200 (CEST)
From: Julien Garet <julien.garet@inria.fr>
To: cloudstack-users@incubator.apache.org
Message-ID: <50990836.29726476.1351117345482.JavaMail.root@inria.fr>
In-Reply-To: <508637A5.40409@upcnet.es>
Subject: Re: backup a cloudstack deployment
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----=_Part_29726475_1906762031.1351117345481"

------=_Part_29726475_1906762031.1351117345481
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

Hello,=20

> De: "Paco Orozco" <paco.orozco@upcnet.es>
> =C0: cloudstack-users@incubator.apache.org
> Envoy=E9: Mardi 23 Octobre 2012 08:22:29
> Objet: Re: backup a cloudstack deployment

> Hi Julien,

> Good point! We have a private IaaS cloud and we have the same
> question.
> At the beginning documentation and forums recommends to backup MySQL
> DB
> and the secondary storage.

> I'm using Cloudstack and Xen servers with LVM as a primary storage.
I'm using KVM along with Cloudstack with iSCSI and GFS2 (on Dell Equallogic=
) as primary storage.=20
As secondary storage, I am thinking at using disk based backup appliance wi=
th inline deduplication (Quantum, EMC DataDomain, ...), has anyone tried su=
ch a setup ? Are performances good enough or do I need some kind of speedy =
NFS server ?=20

> I'm doing it but I've got some not solved questions.

> 1) Geoff says "You should have every VM running a Scheduled Snapshot
> routine, that way if you have a disaster, you can recreate your
> architecture, and restore VMs from Snapshots / Templates". When you
> schedule this snapshot routine it will create snaps on secondary
> storage. Does this snaps count as a consumed resources for users?
> Must
> they will pay for it? Can users delete them?
As far as I know, a snapshot is a resource as others, it has its own limits=
, different from the volumes. Users can delete their own snapshots. The ser=
vice offer we are planning to provide is to tell users they have all the to=
ols to plan their own backup (via the snapshots) and are responsible to set=
 it up for their VMs. We ensure that in case of disaster on primary storage=
, they will have tools to recover failed VMs.=20

I've done some testing around snapshots (manual or scheduled) and it appear=
s that, vm running or not, I have a lot of errors. For example, I've schedu=
led hourly snapshots on a VM, keeping the last 3, and I currently have : th=
e last backup is in error, the two others are OK on the root disk, and for =
the data disk only the last is good.=20

Does anyone face this kind of snapshots problems ?=20
Another question around that, are there means to have a copy of a volume on=
 secondary storage or only snapshots are stored on it ?=20

> 2) I'm now testing a solution for PHD in order to do VM backups from
> Xen
> Server. Are you using Xen or KVM/VMware?
I did not look for kvm specific backup solutions, I have a little backgroun=
d with XenServer and last time I looked at it, I finished with a perl scrip=
t that did vm snapshoting before moving the volume to a dedup appliance (Qu=
antum DXi)=20
> 3) I need to protect myself from harware failures (mirroring) and
> software or human errors (backup). Are you only using mirror?
For human errors, in Cloudstack, I think that the cleanup delay before remo=
ving destroyed VMs can help a lot. I'm also quite afraid of data corruption=
 on the raid array, e.g. many disks failing at the same time because of a b=
ad series... How long will it take to recover the whole platform if only fr=
om secondary storage ? And recover will only be on the user behalf ?=20

How will Cloudstack behave if a lot of VMs on a primary storage are not ava=
ilable ?=20

> Thanks
> --
> Paco Orozco (paco.orozco@upcnet.es)
> SIP: paco.orozco@upcnet.es
> Backoffice. =C0rea de Serveis TIC
> UPCNet
> Edifici V=E9rtex - Pl. Eusebi G=FCell, 6
> Tel=E9fon centraleta: 93.40.11600
> GPG Key ID: 0x3EDEC0AC

------=_Part_29726475_1906762031.1351117345481--