Need a script to update directories

Aleksandar Milivojevic alex at milivojevic.org
Tue Jan 2 18:08:58 UTC 2007


Quoting "Burke, Thomas G." <tg.burke at ngc.com>:

> Hey all,
>
> 	Background:  I do all my web page work in a personal directory
> so as not to FUBAR the real web pages.  Problem is keeping the two
> synchronized.
>
> 	I have thrown together a simple "copy script" that runs as part
> of my nightly cron job - it copies *everything* from my working
> directory to the actual web directory every night.  These are both on
> the same machine, so access is not a problem.
>
> 	Problem:  Copying everything is slow and, generally, painful.
> There's a couple 100 meg of data there.  I'm looking for a script to
> update the web directory with newer/missing files.  I don't mind at all
> if some stuff stays there, a true sync isn't necessary.  I just need to
> get all the new content pushed to the appropriate place.

I can think of two obvious solutions.

One is to use rsync to keep things in sync.  Obvious advantage is that  
it is simple to setup.

The other one is to use CVS or some other revision control system.   
You may wish to block access to CVS directories (in case of CVS) it  
will create in httpd.conf.  It is more complicated to setup, but if  
you already use CVS or some other revision control system to manage  
those 100 MB of HTML than you already have everything in place.  Just  
do "cvs co whatever-you-call-it" to copy current version out of  
repository.  You are using some kind of revision control system to  
manage those 100 MB of HTML?  Right?

Either will transfer only the updated files.






More information about the redhat-list mailing list