Thank you for the response. I like the approach. The goal (for me)
is to produce something that can be managed via a web-browser, provide
user friendly file selection, backup and restore interfaces and works
so long as you can write scripts to produce the desired backup
results.
Amanda looks cool and I'll check it out, but from what I can tell, can
only be managed from another nix box. I could be wrong (and usually
am :)
I've played with the Webmin tools, but they are not very intuitive for
the average user trying to perform a restore operation.
Thanks again for the post.
I'll let you know if I find time to work on this and post the results
back.
Quote:
> Interesting idea Mark,
> Considering that I know how to implement enterprise systems with BrightStor
> and NetBackup DataCenter with ability to restore any file that was backed up
> within 6 months in about 5 minutes, it might be an interesting challenge and
> a relatively easy one to implement.
> Any programming language capable of executing system commands and connecting
> to databases can be used here.
> There is another product you should consider, it's called Amanda.
> http://sourceforge.net/projects/amanda/
> In its simplest form this is how I would do this concept with PHP and ADODB
> (stifling any DB wars in the process :).
> Exec a backup utility. Parse a backup utility's logs into the DB with a
> current time stamp, current backup file name, backed up file's mtime,
> current tape index number.
> On restore, run a select, locate the record in my DB, instruct the backup
> program of choice to restore it to either the original location or location
> of choice for the user.
> The real fun is to take something like tar or star, hack direct DB support
> into it, and then have a commercial-strength product not much unlike that of
> Computer Associates or Veritas offerings.
> Leonid