Geoff Shang wrote:
I was thinking that I could perhaps make daily incremental tarballs,
with a full backup once a month or something, as bandwidth is a bit
limited at the office site (though exactly how much I'm about to find
out). But I've not had experience making incremental backups so would
like suggestions as to the best schemes to use, etc.
Simple answer: use duplicity.
As an example, here is the (somewhat simplified) script we run daily via
cron to do a full backup every week with incremental backups each day
between full backups, slicing the backup into chunks of 100mb each, unto
the Amazon S3 grid, while encrypting everything using GPG in symmetric
cypher mode and keeping backups for 1 year.
#!/bin/bash
export PASSPHRASE='this password is fake'
export AWS_ACCESS_KEY_ID=EXAMPLE123
export AWS_SECRET_ACCESS_KEY=EXAMPLE567
duplicity --volsize 100 --full-if-older-than 1W /back/me/up \
s3+http://bucketname/
duplicity --force remove-all-but-n-full 48 s3+http://bucketname/
Pretty cool, I think.
Of course, it's just a single example. Read Duplicity fine manual for
the details
=================================================================
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
