(E-Mail Removed) (Chas) wrote in
news:(E-Mail Removed) om:
> We have two Redhat Fedora servers. One runs a mission critical app,
> the other is a web server. We have office locations on the east and
> west coast of the U.S. connected by T-1's. What I have been asked to
> do is set up two new servers on the east coast, that will exactly
> mirror - either in real time, or on a tight schedule (i.e. hourly),
> the data on the west coast. I would like to initiate failover
> automagically, or have a procedure in place so that this can be done
> within a very short period of time. I was originally thinking of just
> using rsync, is there a better way?
> Has anyone done this? Could you point me in the right direction to
> start looking? Im open to software solutions if someone has a good one
> that they have used successfully.
>
> Thanks,
> Chas.
>
It is way more than just mirroring servers.
Keeping the data in synch is only part of the challenge.
My initial reaction is that what you have described must have been thought
up by Dilbert's PHB who thinks the way to solve disk space problems is by
emailing pieces of the database to users who have spare disk space on their
desktop PCs. Somebody knows enough buzz words & acronyms to be DANGEROUS.
Nothing is impossible for the person who does not have to do it!
Here is what I think "you" want or desire to happen.
The mission critical app. server of the Left Coast melts down.
Automagically the Left Coast clients will detect this failure
and their packets will immediately find their way to the "hot standby"
server on the East Coast. This scenario will stipulate that you have
been successful at figuring out how to keep the data in synch.
I doubt that you'll ever see this pipedream become reality.
What kind of clients do you have?
How are they going to detect the failure & redirect their connections
to the new server?