[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [cobalt-users] Cobat GUI Backup... what's really backupped?



Dave asked:
> Quick Question:
>
> I know that the cobalt scheduled backup from the interface
> allows backup of all sites, users, and mail directories....
>
> now, if I perform a
>
> du -hcs site* under /home/sites  -> I get a value of 209 MB Used
>
> du -hcs /var/spool/mail -> I get a value of 15 MB Used
>
> How come then that my backup is 340 MB?
>
> Does anyone know what directories are really back-upped with
> the standard cobalt backup?
>
> Cheers!
>
> Dave.
The easiest way to answer your question is probably to give you the tools to
look at it yourself. Its a normal gzip'd tar file with some garbage thrown
in to keep things proprietary. :)

Jeff Bilicki from cobalt net.com posted a script a long time ago to strip
the headers out of the file and allow you to work with it like normal... I
am sure he won't mind if I repost as its probably in the archives somewhere.

Script starts here....
#!/usr/bin/perl
# Jeff Bilicki <email deleted out to avoid spam harvesters>
# removes the header out of a Qube 2 and RaQ 2 backup file

use strict;

my ($infile);
my ($outfile) = "out.tar.gz";
my ($end) = "\%\%END_XML";
my ($begin) = "\%\%BACKUP_HEADER";

if (@ARGV) {
    $infile = $ARGV[0];
} else {
    print "usage: stripheader.pl <file name>\n";
    exit 1;
}

open (INFILE, $infile) or die "Can't open: $!\n";
open (OUTFILE, ">$outfile") or die "Can't open $!\n";

while (<INFILE>) {
    if ( /^$begin/ ... /^$end/ ) {
 next;
    } else {
 print OUTFILE $_;
    }
}
close(INFILE);
close(OUTFILE);
exit 0;