This is kind of a basic question. I am a newbie to CVS.
Don't synchronize from your CVS root tree on the server. That's a bad idea. Use "cvs export" to pull a tree without the CVS directories, and then synchronize with that.
The ,v means it's a cvs repository file - i.e., when you add foo.c, cvs creates foo.c,v in the repository.
P.S. Beware cvs export - it's very easy to mangle binary files (images, etc.) through inadvertent keyword expansion.
Binary files should be checked in as binary, which annotates their ',v' files with a flag turning off keyword expansion.
What about doing the check out of the entire repository to a mirror site and then doing the file synchronization between the mirror site and live site?
Also, why use synchronisation software at all? Just export directly from the CVS repository to the live site.
Just want to use the mirror site as a layer of protection and test the mirror site before making it live. Also want to keep the live site separate from dev. site, not associated in any way to CVS. So want to use a different program for file synchronization and comaprison.
I'd suggest to use "cvs export" instead of "csv co" for your mirror site as to avoid any possibility to check-in any changes, and to avoid all the CVS dirs.
Fog Creek Home