A couple of days ago I came across this excellent tip to download your GMail mail to your computer with IMAP. Before you continue reading here, I assume you read that article first. In this article, the path /path/to/gmail/backup is the path of the local repository, which you configured for offlineimap.

In the comments of that article, I stated that there's a problem with duplicate files. GMail has no real folders, but you can label e-mails. This implies that the same e-mail may possibly end up in more than one folder. When you download all folders, it is likely you store that e-mail multiple times.

Due to circumstances I had to look for a new way to backup my home directory. I was using rdiff-backup(1) before, but I wasn't completely satisfied with it. The overhead of incremental data was quite large, especially if you moved files around. But overall it does its job quite well.

However, since a year or so I'm quite charmed of Git, the stupid file content tracker. It's a great tool for software development, but due to its advantages over other VCSs it's also suitable for backing up structures like a user's home directory.

Some time ago I wrote an article of my backup solution using dar(1). But somehow this started to get itchy because dar is not really a portable format. Whenever something happens to my original files, I need dar in order to restore them. So was basically bound to quite specific tools.

In order to decrease our backup size we decided to skip some files and directories which were not critical. With chflags(1) you can skip entire directories:

Situation: a FreeBSD webserver needs a backup system. dump and restore are your friends then. Restoring all that data is not that trivial, if you don't know how dump operates. But the manpages of these commands should help a lot already.