Re: [rdiff-backup-users] Moving the backup-location to another machine

[Thomas Harold]

On 1/11/2014 12:45 PM, Ron Leach wrote:
> We're replacing our backup server and hit a problem moving our existing
> backup directory to the new machine.  Despite the new machine's disk
> space being slightly larger than the space on the existing system, we've
> run out of space on the new machine, and the move is not yet complete.
> I think we've either approached the moving of the backup incorrectly, or
> we've not created as large a filesystem as we had intended

Maybe you ran out of inodes?

When working with backup solutions like rdiff-backup, keep in mind that 
it uses up a lot of inodes.  Doubly so when backing up a source file 
system which also has lots and lots of files.  So unless you create your 
ext4 filesystem with additonal inodes, you can run into issues on that 
front.

# df -i

Typically, I create ext4 rdiff-backup volumes as:

# mkfs.ext4 -b 4096 -i 8192 /dev/vgname/lvmvolumename

...

Using "cp -a" works fine, you could have also used rsync or any other 
solution to move the rdiff-backup directory.  The rdiff-backup program 
doesn't rely on the target file system to store metadata (permissions, 
attributes, etc), it stores it in separate files.  The only important 
thing is to set permissions properly on the new location so that your 
backup clients can write to their directories.

Lastly, don't run any file system more then 80% full.  Buy larger disks 
and setup some sort of alerting system to check disk space daily/weekly 
and send emails out when you get above your threshold.