Mail archiving/restoring??

My requirements are as follows:

In our company "openwebmail" is the software used as mail client for the intranet
website and the mail server is sendmail redhat linux 7.2.

The backup server is running redhat linux 7.2

"openwebmail" is written is perl.
It is accessible from the Internet explorer.

I have to develop an archiving facility for the mails. The mails should be archived
automatically on per user basis when the quota granted to the user exceeds and should also be archived explicitly by user if he wants.

I have to archive the mails on the backup server mentioned above.

How should I begin? What should be the programming language and which platform should I use?
Is there a need for a database backend?

Please help me out,
akaash...

First off, is there some reason you have to use a distribution that's almost a decade old? You should look into upgrading the operating systems on both of those servers.

Openwebmail is nothing more than a client. This could just as easily be Squirrelmail, Horde, Outlook or tons of other mail clients, it has nothing to do with how mail is stored on your server. That said, what you need to do is figure out where the actual mail is being stored on the server. This might be in user's home directories, a specialized user account set up for it, somewhere under /var, some local or remote database or any number of other places.

Once you've figured out where and how mail is stored to begin with, you can then setup a backup system for it. Like everything else Linux, there are about 5 million different ways to do this and the proper methods will largely be determined by how your mail is stored as well as how coherent of a backup system you require. Generally speaking (and this isn't always the case) email is stored in a directory structure (though it may be in a database) that can be backed up like anything else on the server.

Redundancy solutions don't generally require a lot of custom programming (again, situations vary) as there are tons of tools you can use to help with this such as rsync, DRBD, and so on. If you do end up having to write any code, it'll likely be bash and/or perl scripts.