[Rear-users] Questions on handling

[Schlomo Schapiro]

Hi Werner,

well, the documentation is somewhat lacking still, though I hoped that the 
general part would be sufficient to cover your question (feel free to submit 
better documentation :-))

Basically it goes like this: mkrescue creates a rescue media and mkbackup 
creates a rescue media and does the backup, but only for those backup methods 
that are actually an internal part of rear :-) At the moment this is only NETFS 
which creates a tar (or pax ...) archive on a network share.

The reason why one needs to specify the backup method (and indeed all possible 
options are specified in the /etc/rear/default file) is that for some backup 
solutions (like TSM) there is special support to
1) back up the rear mkrescue image with the backup solution
2) include the backup agent/client in the rescue system
3) restore the entire system from backup with the specified backup solution.

In all cases except NETFS the backup (usually a full backup) has to be taken 
independantly of rear. But rear will then be able to automatically restore the 
system by recreating the disk layout and invoking the backup client to restore 
the files from the last backup.

Implementing support for another backup solution is fairly simple, took me 2-3 
days each time. Of course I don't implement the support for a backup solution in 
my free time but as part of my consulting work, that is why there is only 
support for TSM and GALAXY (yet). If you will send me a patch for NetBackup then 
I will be happy to include it. Otherwise I will also be happy to implement it 
for you (Email me off-list for conditions & details).

The backup methods REQUESTRESTORE and EXTERNAL are special because they don't 
take anything into the rescue system but expect some "magic" to restore the data 
when requested or allow specifying your own custom scripts. But it is always 
better to write a support module for your backup software because that would 
allow you to fully utilize the power of rear and also contribute your effort 
back to the project.

One popular example for REQUESTRESTORE is having the backup done with rsync 
(e.g. RBME) and then obviously the backup server has to push the data back to 
the system (using rsync over ssh). If you are looking for a simple backup 
solution besides NetBackup, do take a look at RBME as it simplifies the rsync 
backup.

Regards,
Schlomo

PS: If you want to quickly test the workings of rear, simply use
BACKUP=NETFS
NETFS_URL=nfs://host/share/path
and make sure your system can NFS mount that url. The SLAC demo was just the 
same, the demo film on the website also shows this in detail.

Werner Flamme wrote:
> Hi list(eners), ;-)
> 
> I'm trying to use ReaR in earnest now - and I'm struggling with the
> documentation :-(
> 
> When I understand /etc/rear/default in the right way, the difference
> between "rear mkrescue" and "rear mkbackup" is that the latter does not
> only save the data necessary to restore the system but saves "the whole
> machine" too. So, mkrescue is part of mkbackup.
> 
> To achieve a backup, I must set BACKUP= to the right value. What values
> are possible, and what is the effect of the respective setting? Per
> default it's BACKUP=REQUESTRESTORE. I do not see a difference between
> mkbackup and mkrescue here - what does REQUESTRESTORE mean/do? Other
> possibilities seem to be GALAXY, TSM, NETFS, and EXTERNAL. Since we use
> ArcServe's NetBackup in the company, I think I am bound to EXTERNAL - if
> the client is scriptable, which I think it is not...
> 
> And when using EXTERNAL, I am completely free what I want do do? As
> EXTERNAL_BACKUP I can give some shellscript that looks around the disk
> to see what's worth saving ;-), moves/copies it to a collecting
> directory tree (for example /forbackup/*) and have it maybe transported
> to another machine without the knowledge of ReaR? Same for
> EXTERNAL_RESTORE then?
> 
> I really loved the demo on SLAC2007 :-) - everything was so simple, but
> I never asked for details...
> 
> Regards,
> Werner
>