Community technical support mailing list was retired 2010 and replaced with a professional technical support team. For assistance please contact: Pre-sales Technical support via email to firstname.lastname@example.org.
You have totally misunderstood the CVS concept... I recommend reading this book first (on-line version): http://cvsbook.red-bean.com/cvsbook.html Then please take note the following distinctions: 1. CVS server Here is where the full history of all files in all modules are kept. The server keeps this in the "repository" in special files ending with ,v. These folders should not be directly accessed by anyone using other tools than CVS. 2. CVS client There can be many cvs clients to a single cvs server. Every client has his own personal copy of the source files, which he gets using the cvs checkout command. These files are normal files and are used by the development system. Note that at the same time there can be many developers working on the same source files, they all do this in their own private copy on their own PC. 3. CVS update Whenever a developer thinks that there might be some new changes to the files that other developers may have done he should do a cvs update command. This will get all changes from the server and merge them into the local copy files. Now these files contain the latest server revision plus the changes the developer himself has done. 4. CVS commit When a developer is done with the edits on his file and wants to send it back he has to do two things: a) cvs update so that his files all contain the latest changes from the server (plus his own changes) b) cvs commit so that the server will get his latest changes. This will increment the revision counter on the server 5. Automatic updates What you are asking is why the developer #2 does not see the changes when the developer #1 does his commit. Well, this is not the way CVS works and in fact is also impossible, just think how it could be done if the developer #2 has switched off his PC or is travelling with it. Impossible for any automated action here. No, CVS works on the commands issued by the developers, so if a developer wants the latest changes then he must himself do a cvs update. There is no other way. /Bo -----Original Message----- From: kaane [mailto:kaane at 121metre.com] Sent: den 4 februari 2003 14:19 To: cvsnt at cvsnt.org cvsnt downloads at march-hare.com @CVSNT on Twitter CVSNT on Facebook Subject: [cvsnt] how to update remote copies from remote respository Hi all, Assume that; There are machines which IP 10.200.10.11 (my machine) IP 10.200.10.12 (server which has cvsroot ) and my application server is 10.200.10.12 too. I am getting a copy by using checkout command. Then, I am changing some parts of code and, I am executing 'commit' command to make the original source updated. But CVS just updates ,v files such as main.class,v instead of main.class At that situation, there is no real code update. because of my application doesn't care the ,v files. Is there any way to do a real update by using cvs? On the other hand, I have an idea which is not succeeded by me because of an unknown key-point. Idea: I am using 'cvs -d RESPOSITORY update ' to update mu local copy. But, is there any parameter for using "cvs -d RESPOPSITORY update -xxx remote_copy_addr" (realize that -xxx is made up by me) ??? thanks for all... Kaan Erdemir _______________________________________________ cvsnt mailing list cvsnt at cvsnt.org cvsnt downloads at march-hare.com @CVSNT on Twitter CVSNT on Facebook http://www.cvsnt.org/cgi-bin/mailman/listinfo/cvsnt https://www.march-hare.com/cvspro/en.asp#downcvs