Hello,
I beg your pardon for bringing this topic to the table again. I have read this
entire thread (even the scientific debate around thrusting probabilities
calculus) and have tried the main suggestions detailed by you to solve this
problem (except the use of raid mirror detailed by Les Mikesel, which does not
fit our needs).
What worked best for us was the strategy of copying the 'cpool' with an
standard rsync (v.3) and then use BackupPC_tarPCCopy to re-create the sets of
backups. I have even refined that approach creating a script to break down the
rsync of the 'cpool' in multiple rsyncs (per sub-directory) and to "smartly"
run BackupPC_tarPCCopy for only the new backup sets of each pc.
This approach works best in the environments we use Coraid boxes and that we
can mount both the regular BackupPC partition and the backup partition in the
same server, so as to BackupPC_tarPCCopy directly to the backup partition. In
the other cases we need to create a tar file with BackupPC_tarPCCopy, copy it
to the backup server over the network, and then untar the file - that adds a
new level of complexity to the solution.
Anyway, the first step in this approach (rsync of cpool) works considerably
fine (for cpools containing a few Terabytes of data). What doesn't always works
fine is that BackupPC_tarPCCopy sometimes produces tar files that are too big
(several Gb, as for the backup of Zimbra and database servers), which brings to
the following question:
* Why BackupPC_tarPCCopy sometimes produces big tar files ? *
For what I understand, after the completion of a backup, BackupPC_link is run
to transfer all the non-BackupPC-system-files (e.g., attrib files) to the
'cpool', replacing then by hard links. A tar file originated from a 'linked'
backup set would then contain mainly a list of relations between the files that
compose the data set and their relative position in the cpool, so we can use
the same tool to untar the file and re-create the data set with hard links.
Which brings to my second and last question:
* Is this (BackupPC_tarPCCopy creating big files) happening because the
interval between the execution of my backups (the end of one and the start of
the next one) is not giving enough time to BackupPC to run BackupPC_link and
finally the files in the data sets are not being 'linked' in the cpool ? *
Ideally I would like to have all files in the cpool so the rsync of this
directory would make for most of the trouble. If the second question is true
then it would be great to have a way to manually run BackupPC_link to "clean"
the backup sets.
I appreciate your view in those two questions.
Regards,
--
Fernando Laudares Camargos
Révolution Linux
http://www.revolutionlinux.com
---------------------------------------
* Tout opinion et prise de position exprimée dans ce message est celle
de son auteur et pas nécessairement celle de Révolution Linux.
** Any views and opinion presented in this e-mail are solely those of
the author and do not necessarily represent those of Révolution Linux.
Peter Walter a écrit :
> All,
>
> I have implemented backuppc on a Linux server in my mixed OSX / Windows
> / Linux environment for several months now, and I am very happy with the
> results. For additional disaster recovery protection, I am considering
> implementing an off-site backup of the backuppc server using rsync to
> synchronize the backup pool to a remote server. However, I have heard
> that in a previous release of backuppc, rsyncing to another server did
> not work because backuppc kept changing the file and directory names in
> the backup pool, leading the remote rsync server to having to
> re-transfer the entire backup pool (because it thinks the renamed files
> are new files).
>
> I have searched the wiki and the mailing list and can't find any
> discussion of this topic. Can anyone confirm that the way backuppc
> manages the files and directories in the backup pool would make it
> difficult to rsync to another server, and, if so, can anyone suggest a
> method for "mirroring" the backuppc server at an offsite backup machine?
>
> Regards,
> Peter
>
> ------------------------------------------------------------------------------
> Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, CA
> -OSBC tackles the biggest issue in open source: Open Sourcing the Enterprise
> -Strategies to boost innovation and cut costs with open source participation
> -Receive a $600 discount off the registration fee with the source code: SFAD
> http://p.sf.net/sfu/XcvMzF8H
> _______________________________________________
> BackupPC-users mailing list
> BackupPC-users@lists.sourceforge.net
> List: https://lists.sourceforge.net/lists/listinfo/backuppc-users
> Wiki: http://backuppc.wiki.sourceforge.net
> Project: http://backuppc.sourceforge.net/
------------------------------------------------------------------------------
Come build with us! The BlackBerry® Developer Conference in SF, CA
is the only developer event you need to attend this year. Jumpstart your
developing skills, take BlackBerry mobile applications to market and stay
ahead of the curve. Join us from November 9-12, 2009. Register now!
http://p.sf.net/sfu/devconf
_______________________________________________
BackupPC-users mailing list
BackupPC-users@lists.sourceforge.net
List: https://lists.sourceforge.net/lists/listinfo/backuppc-users
Wiki: http://backuppc.wiki.sourceforge.net
Project: http://backuppc.sourceforge.net/
