Merv
Sun Dec 28 19:54:13 CST 2003
There have been reports in this NG that these two work well as "on-the-fly"
imaging solutions:
PowerQuest V2i
Acronis True Image server
No experience here, though.
--
Merv Porter [SBS MVP]
===================================
"D H Harris" <dixon@sohelpme.info> wrote in message
news:eKn1sgazDHA.1908@TK2MSFTNGP10.phx.gbl...
> Merv Porters excellent advice includes:
>
> "Some form of RAID
> Daily backups to tape
> Regular imaging
> Storage of backup media offsite."
>
> If there is any imaging software that supports server imaging I have not
> heard of it. It's
> still a good move though, in spite of that official declaimer it seems to
> me. Shoot, how bad can it be as long as it is not the sole backup method
> being relied on?
>
> Question for all: Does any imaging software support server imaging?
>
> TIA
>
> DHH
>
>
> "Merv Porter [SBS-MVP]" <mwport@hotmail.com_no_spam> wrote in message
> news:eaLa30MzDHA.3216@TK2MSFTNGP11.phx.gbl...
> > Hi Tracey:
> >
> > Disaster Recovery (DR) means storing backups and images offsite in case
of
> > fire, theft, water damage, earthquakes, meteorite impacts, asteroid
> > collisions, aardvark attack, etc. Of course, any event that destroys
your
> > server is going to be a major pain since, most probably, the replacement
> > server will not have the same hardware components as the original. This
> > means that you're backup tapes will probably the most useful if the
entire
> > server has to be replaced (assuming you can get the same type of tape
> backup
> > drive for the new server). However, IMHO, under most DR circumstances,
> > imaging in conjunction with tape backups seems to be the quickest/most
> cost
> > effective way to restore your server.
> >
> > Assuming you have less than about 40 GB of data to back up (system,
> Exchange
> > store, SQL, user data, etc.), if you buy an external USB 2.0 hard drive
> (and
> > a USB PCI adapter if you're server doesn't support USB 2.0 natively),
> you'll
> > be able to fairly easily image the server to it using Ghost 2003 by
> downing
> > the server and booting up on a Ghost boot CD or Ghost floppy disks.
This
> > should be done whenever the server undergoes any major changes (new
> > hardware, new software apps, OS SPs, multiple patches). In general, a
> > once/month image schedule should be sufficient. Multiple images can be
> > stored on a single external USB HD for redundancy since you never really
> > know if an image is "good" until you restore it. Of course, this puts
all
> > you eggs in one "external" basket, so alternating multiple external USB
> HDs
> > would even be better. Just make sure to store the image(s) offsite.
> >
> > As you noted, routine daily full backups of all your data using multiple
> > tapes are also a must to be able to bring the server back up with a
> minimum
> > amount of down time. Again, these tapes need to be stored offsite.
> >
> > Now, as Keith mentioned, incorporating RAID (minimum RAID1 software or
> > hardware) is also a must if you need minimum down time on the server.
> This
> > will cost you at least another HD. The downside to software RAID1 is
that
> > you need to convert the HDs to Dynamic (from their original Basic
> > configuration). This is not a problem for Ghost 2003 as you will be
> booting
> > to DOS and creating the partition image(s) as type Basic and then
> converting
> > back to type Dynamic after you have restored the image(s).
> >
> > Before imaging, I would suggest breaking the mirror (and maybe removing
> the
> > secondary "mirroring" HD from the server). Then regenerate the mirror
> after
> > the images have been created. This is a precaution in case things go
> south
> > with the imaging operation.
> >
> > So, depending on your level of risk tolerance, you need these elements
for
> a
> > DR program:
> >
> > Some form of RAID
> > Daily backups to tape
> > Regular imaging
> > Storage of backup media offsite
> >
> >
> > Just my $0.02.
> >
> > --
> > Merv Porter [SBS MVP]
> > ===================================
> >
> > "tracey" <anonymous@discussions.microsoft.com> wrote in message
> > news:007601c3ccba$752a84d0$a601280a@phx.gbl...
> > > Hi
> > >
> > > i am trying to find out the cheapest and quickest way to
> > > get my server up and running in an event of a failure.
> > >
> > > What do you think of my solution? I was thinking i
> > > should buy a second hard disk. Image the disk i have now
> > > and then restore the image on the new drive and keep it
> > > in the cupboard. Keep regular backups of my current data
> > > and in the case of a failure, pop the second disk in and
> > > restore the data and continue as normal.
> > >
> > > Every six months create another image of the current hard
> > > disk and restore it to the one in the cupboard.
> > >
> > >
> >
> >
>
>