Bob, Your reply just made me notice that I replied to you directly, instead of to the mailing-list. So, I resent my message for the benefit of the others. On Wed October 24 2007 20:06, Robert Smits wrote:

What does the first line with "#!/bin/bash" do? I understand it's a descriptive comment, but what does the !/bin/bash do for you.

It is not a comment. While # starts comments anywhere else in a bash script, if the script starts with "#!/bin/bash", bash will be used to interpret the script, even if someone's login shell is csh or tcsh. For example, if I had used the Bourne shell syntax, I should start my script with #!/bin/sh.

Second, I've been using rsync in the form rsync -a source file location destination file location. Is it because you want to do this interactively that you've added all these other options?

The fundamental difference is that both ends are active. If I was just mirroring my home to a passive backup destination, then a pure --archive option would suffice and I would probably do it on my entire $HOME. But in my case, I synchronize only key directory trees between the 3 machines and any of the 3 can have a more recent version of a document. --compress: speeds up synchronization between office and home. --update: prevents overwriting a newer version on the destination, if I forgot to update it in the source first. Say I change Projects/ABC/file1 in the office and forget I changed it. Then I work on Projects/ABC/file2 at home. If I didn't use update, I would revert to the old version of file1 when I archive Projects/ to the office. --partial --delay-updates: keep partial transfers if the connection breaks, so a restart only transfers the rest. The problem is the time of a partial file is always newer, since the time stamp is adjusted at the end. So, at restart a partial file would not only look complete, but also more recent, and when I synchronize back, I would corrupt the source, hence delay-updates. --progress: just to see that the connection is not hanging in case of large files.

...

I found that sometimes I forgot that I created a new file in the office, say under Projects/ABC/, and when I work from home and I want to "syncsend Projects", it asks if I want to delete that new file. I then say "no" and subsequently run "syncget Projects" to bring that file to my desktop at home.

But if i use the rsync command without the --delete parameter it won't delete any files at all on the receiving side, right? And if I have an extra one in either receiving or sending side, it'll add it, right?

Your synchronization is always one way, never back and forth at once. Without --delete, rsync will notice that there is an extra file at the destination, which would normally mean you deleted the corresponding file in the source, but rsync will not synchronize this deletion. rsync cannot know that in my case the destination side is not a pure backup site and that it can also generate new content. I have to run "rsync office home", which is "syncget" in this case, for rsync to bring the new file to my home machine.

Well.my scripts were much more primitive and I'll study yours to see how they add functionality. Once I have the scripts working I just want to run them daily with cron.

These are just different scenarios. I would use a cron job to sync to a true, passive backup site. Since all my sites are active and I am somewhat absent-minded, my initially simple interactive script evolved to this. :) -- Carlos FL Who is General Failure, and why is he reading my disk? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org