On Fri, 15 Apr 2011 13:11:21 +0000 (GMT) Kai Zimmer <[email protected]> wrote:
> Hi list, > > i've been using amanda for many years now (stuck at version > 2.5.2p1-4). She has been recovering my/our files very reliably for a > long time - thanks a lot for that great piece of software! But by the > time files and file systems are getting bigger (14TB of data on the > main backup server right now) and i can't cope with splitting up my > DLEs into chunks of ~20GB anymore, which i have to do with the > (ancient) version of amanda i'm still using. I'm not exactly sure > anymore, why exactly had to split up my DLEs - probably because of > timeout problems. So, i somehow have to reorganize my backup and > therfore have some questions about newer versions of amanda: > - Is splitting up DLEs manually into smaller chunks still necessary > with current versions of amanda? No. A typical reason to split up DLEs is tape size, where the DLE is too large for the tapes. Another way to deal with that is to allow tape spanning, i.e. allow amanda to use multiple tapes per run. You will need a changer or vtapes. If timeouts were the original problem, you should be able to increase them. > - Is there an option for DLEs not to follow symlinks to other > filesystems? I would think so, but don't know off-hand. I know you can tell GNU tar to do this. > - Is there a way to save files regularly to disk and only > occasionally to tape? Have two configurations, one for tape and one for vtape. Or have only the vtape configuration, and separately copy the vtape directory to physical tape as needed. If you are going to do the latter, you may also want to get the amanda metadata as well. > - Is there a way to allow DMZ backups without opening a whole range > of ports? ssh tunnels? I haven't done this, so I am guessing here. > > thanks, > Kai -- Charles Curley /"\ ASCII Ribbon Campaign Looking for fine software \ / Respect for open standards and/or writing? X No HTML/RTF in email http://www.charlescurley.com / \ No M$ Word docs in email Key fingerprint = CE5C 6645 A45A 64E4 94C0 809C FFF6 4C48 4ECD DFDB
