For some reason, my old backup scripts aren't working. A simpler approach is to add this sort of thing to a script called by cron:
rsync -vrmxz --progress --include='*.ipynb'
--include='*/' --exclude='*' /home/me /mnt/J/backup/
which starts at the current directory and works through all sub-directories, grabbing each .ipynb file and copying it to /backup.
The parameters:
a: archive mode. Preserve attributes.
v: verbose. Not needed for the silent cron version.
z: compress during transfer. Not needed if creating a first copy locally but useful for network xfer.
x: one filesystem: don't cross file-system boundaries.
r: recursive.
m: prune empty directories. Essential.
progress: live feedback. Not needed for silent cron version.
include: files to include
exclude: files to exclude. This is required even though we used an include command. *.* is essential.
Make neater using variables:
hour=`date +%H`
i="ipynb"
rsync -vrmxz --progress --include="*.$i" --include='*/'
--exclude='*' /home/me /mnt/J/backup/hour$hour/
Regarding ddrescue solutions, watch for squashfs: it will create a compressed backup image that can be mounted. Seems better than ddrescue piped to gzip or so.
No comments:
Post a Comment