Hi,
We're doing offsite backups using rsync, more or less the cookbook example
using:
rsync --numeric-ids --compress --rsh=/usr/bin/ssh --recursive --archive \
--relative --sparse --one-file-system \
--compare-dest=/vol/backup/$HOSTNAME/current $HOSTNAME:$DIRECTORY \
/vol/backup/$HOSTNAME/$DATE
The script also does a week-rotation, when "current" is backed up
elsewhere, and we rsynch the current directory (where we also include the
--delete option).
The problem we have arises form logfiles. We prefer batches of monthly
logfiles, but this means that since this script is running daily in a
week rotation, the logfiles (most notable the access_log's from Apache)
always change, and so are reconstructed on the backup server daily.
Now, I do realise this is still fairly efficient on our network, and that's
not my problem. My problem is more the diskspace all these logfiles take
up. Now I can't believe I'm the first one to have this problem, and unless
everyone else switched to daily rotating their logs, someone has already
come up with a clever answer for this (or hacked rsynch to stop and save the
'diff state' instead of rebuilding the file from older copy and diff).
So, how do you people manage not to keep all the incremental logfiles around,
yet have the latest copy of the logs on backup?
Paul
