BackupPC-users

Re: [BackupPC-users] Offsite copy of backuppc data for disaster recovery

2009-06-29 06:38:17
Subject: Re: [BackupPC-users] Offsite copy of backuppc data for disaster recovery
From: Adam Goryachev <mailinglists AT websitemanagers.com DOT au>
To: "General list for user discussion, questions and support" <backuppc-users AT lists.sourceforge DOT net>
Date: Mon, 29 Jun 2009 20:33:49 +1000
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Randy Orrison wrote:
> I'm currently using rsync to copy that /backup partition to an
> off-site server (it's on a site we own, and the connection is over a
> hardware VPN, so I'm not worrying about encryption either on the
> off-site server or en-route).
>
> What I'm wondering is - what will be my restore procedure if I need
> to use the offsite copy?  I've read through the archives, but
> haven't found exactly the same scenario.
>
> I could copy the files directly out of the pc directories, but the
> filename mangling and lack of filling makes that unappealing (this
> is an operation where rdiff-backup shines).  The easiest way seems
> to be through the web interface, but at present I don't actually
> have backuppc installed on the off-site server.
>
> What it boils down to is this: Can I install BackupPC on the offsite
> server in such a way that the archive is accessible through the web
> interface, but it doesn't actually try to do backups itself?  I'd
> like to install it now and be able to test restoring from the
> offsite server, so that I have one less thing to worry about if I
> ever need it.
>
Yes, you could install backuppc on the remote server in the same way
you installed it locally (presumably not a package install or else you
should be using /var/lib/backuppc on debian). Just copy across the
config files (/etc/backuppc if using the package) but edit them to set
BackupsDisable = 2. This way the remote server will not attempt any
backups, and will automatically purge old backup data as needed... I
think restore should work fine from the web interface, or use the
download tar/zip options...
> Finally, one message I did find in the archive suggested that I
> didn't need to copy the pool directory itself.  If I could skip
> this, it would make the offsite rsync a bit faster, and bandwidth is
> a concern.  (Yes, I know I'll be copying the data anyway, but it's a
> lot less links for rsync to maintain.)  Would restoring with the web
> interface work if the pool directory was completely missing?

Then you should have seen all the messages saying that using rsync to
copy the backuppc doesn't scale well. In any case, you will need to
enable the hardlinks flag to rsync (-H I think) so it doesn't matter
much if there is one extra link in the pool/cpool directory, certainly
it won't change the amount of data being transferred). You should
ensure you are using rsync v3 at both ends, and you may want to adjust
the encryption option rsync sends to ssh since you are not concerned
about data security in-transit (this is just to decrease cpu utilisation).

Other than that, all I can suggest is you read the mailing list
archives, and whatever is on the wiki/faq about this topic as it is
discussed regularly, although there is still no perfect/one solution.

Regards,
Adam
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAkpImIwACgkQGyoxogrTyiUWlACg1CpYR3cyXIKak9MZRvEvMn/7
AJIAoLElGkEjVYDcTwJTvYaXv6IjuZek
=eFq2
-----END PGP SIGNATURE-----


------------------------------------------------------------------------------
_______________________________________________
BackupPC-users mailing list
BackupPC-users AT lists.sourceforge DOT net
List:    https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki:    http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/

<Prev in Thread] Current Thread [Next in Thread>