Stefan Knoll wrote: > yes server and client are runnning on the same machine. The files which > get backed up are on an ext3 filesystem and the disk where the snapshots > get written to is a reiserfs filesystem.
A few words there - if you have to have both reiser and ext3, run the live data on reiser and the backup on ext3. Reiser works well and is really fast - 'til it stops, and then it's really, really, really no fun to repair it, if it works at all. Having your live data on the slower system and the backups on the less reliable really makes no sense as far as I can tell. The private drama around Reiser won't help the project have much of a future either, as the driving force behind the FS is his company. Honestly, I'd just stay with ext3 for everything. > You're right, if I check the size for each snapshot it shows me around > 3.6GB, if I check the size via du for the bank incl. all the daily > snapshots it also shows me 3.6GB. OK, that's kinda what I expected. > What I find a bit strange, and that's why I initially noticed it is, > that df shows the space as occupied, e.g. after each run, 3,6GB are gone. That's weird, though - when df runs out, you're out of space, so you *do* seem to lose space. Have you tried the ls -i thing I suggested? Yours, Bernd _______________________________________________ Dirvish mailing list [email protected] http://www.dirvish.org/mailman/listinfo/dirvish
