Hello, On 12/22/2006 4:36 AM, Dan Trainor wrote: > Hi - > > I've been playing around with Bacula lately, and am quite simply amazed. > Kern, I've really got to hand it to you and your team - you have all > put together an amazing piece of software. Thank you for many, many > years to come, as I dig even deeper into Bacula. > > I've become a real stickler for RPM-based distributions as of late, > specifically RHEL and CentOS. I know that for RPMs, you can query based > on file type of a package: > > [EMAIL PROTECTED] ~]$ rpm -qc httpd-2.0.52-28.ent.centos4 > /etc/httpd/conf.d/welcome.conf > /etc/httpd/conf/httpd.conf > /etc/httpd/conf/magic > /etc/logrotate.d/httpd > /etc/rc.d/init.d/httpd > /etc/sysconfig/httpd > > ...for files marked as configuration files in the RPM preamble. I'm > sure this is possible to do with many other Linux distributions, as well. > > I was wondering if - before I re-invent the wheel here - someone has > already started on a framework for backing up application-specific, > user-modifiable files, such as in this example Apache's httpd.* files > and such.
I think this framework exists. You can use a script to generate the list of files to be backed up. Now, if your script queries the rpm database for configuration files (or, perhaps better, for files modified as compared to what rpm installed) and returns that list you've got what you want. I think. This would be possible with minimal shell scripting and using awk or sed, I assume. > I think that this would help someone who's in the situation > of not wanting to back up an entire package, directory, or something of > that nature - someone who simply wants to backup configuration files, > for a drop-in replacement, when it hits the fan and comes time to restore. Yes, although I prefer to have most of the system restored from tape. Doing a minimal system install (or even starting from a recovery disk of any kind) and then starting the restore seems more straightforward to me than first installing all packages I need, then restoring the backup of modified files, and then all user data etc. In your httpd example, you'd also have to back up and restore the actual data files httpd serves, and finding all these automatically is IMO next to impossible without parsing all of the httpd configuration. > Has anyone looked into this at all? If so, I'd like to talk with you - > if not, then I'd like to make some noise about doing something like > this, because I can see how beneficial it might be. Make some noise anyway :-) Arno > Thanks! > -dant > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share your > opinions on IT & business topics through brief surveys - and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > _______________________________________________ > Bacula-users mailing list > Bacula-users@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/bacula-users -- IT-Service Lehmann [EMAIL PROTECTED] Arno Lehmann http://www.its-lehmann.de ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys - and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ Bacula-users mailing list Bacula-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/bacula-users
