Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Re: Disaster Recovery Plan on Failed Server

2 views
Skip to first unread message

Alex Warmerdam

unread,
Feb 9, 2010, 4:52:58 PM2/9/10
to
On Thu, 04 Feb 2010 10:46:02 GMT, akeaveney
<akea...@no-mx.forums.novell.com> wrote:

Hi,

We have this dr setup working, netware servers.

We use the software from eVault (i365/seagate) for this purpose
together with .

The biggest question you will have to answer, how much downtime is
acceptable.

What we do, we backup the fysical servers to a vault outside the
building and do a test restore daily on virtual servers. This means
that the backup and restore speed are the important ones. Data
transmitted is only the delta's. The bandwith usage between live and
dr location is a 2Mb line. In case of a dr we get 1Gb connection speed
for provisioning the vdi's to the users.

For us this is acceptable.

From the live enviroment, users and servers can not access the dr
servers.

When needed, clients have to connect to the vdi's in the virtual
enviroment to access the dr site.


>
> Hi
>
> We have been asked to investigate possible scenarios on recovering a
> failed Server if it was to go down at one of our sites. In the past I
> have gone to the site in question and fixed that Server or put a
> replacement Server in which can obviously take a few days depending on
> the location.
>
> Someone outside of our company and experience working with
> Novell/NetWare has suggested having another Server onsite elsewhere as a
> backup and for this to come online and take over the failed Server's
> role if the original was to fail. I don't see how this can work having
> to configure the new Server to be the same as the old and the time it
> would take to remove the old one from the tree. Then if having to do the
> same thing again to get the original or replacement one working back on
> the original site.
>
> Is there any other ideas that you do yourselves in your own companies
> to cover this scenario as I have been asked to provide a response? The
> best solution I can think of so far is to have another NetWare Server
> onsite (at every site) for redundancy so users can still login if one
> fails.
>
> Any ideas or thoughts would be greatly appreciated.
>
> Kind regards,
> Ant

0 new messages