Re: Re: Help regarding Wincvs 2.0.2.4

On Fri, 01 Jun 2007 23:19:14 -0000, "connect_to_smita"
<[email protected]> wrote:

>
>Thanks a lot for the help I could setup server adn client they are 
>able to communicate using sspi protocol, but one problem i am seeing 
>now is.
>
>I have cvsed a directory on the server. 

What exactly do you mean by this statement? What did you do? And what
was the result on the server? Your wording is very strange....

>The client PC can cvs login 
>to server and can see all the files which have been cvsed in the 
>directory,

What do you mean by "see all the files"???
Please state *exactly* what you are doing.

>when I execute "cvs status" it asks to cvs checkout first.

Of course! cvs status compares the status of the local file set with
the server files and gives you and indication if the files you have
locally are up to date or if there is newer data on the server that
you need to´update to. For this to make sense you *of course* need to
have a locally checked out set of files....

>but if i checkout then another directory is created which is local 
>to this client and change commited is stored as a new revision in 
>that m/c only not communicated to sever.

It seems you have not even grasped the very fundamentals of the way
CVS works...
The procedure for a developer using CVS to develop code is this:

1. Put the sources on to the CVS repository server from the client
using either the command "cvs import" (works without a loacl checked
out file set) or "cvs add" inside an already checekd out folder.
In the case of "cvs add" you need to also do a cvs commit.

2. For existing sources on the server or for an imported source tree
you need to do "cvs checkout" to receive to the local computer a
working file set from the server. THis is called a "sandbox" in CVS
lingo.

3. Now you work on the files as usual and when you are done you send
your changes back to the server using the "cvs commit" command.
THis places your changes into the files in the server so others can
get them.

>Is there are way to mount 
>the path in server to all clients so that commit on client is 
>communicated to server.

???? Are you suggesting that clients should have direct access to the
server file system in the repository area? This is in general *A VERY
BAD* idea! Normally tyhe server should be locked down such that there
is no share points at all on the disk where the repository is located.
*All* CVS operations should go through the client/server protocol CVS
uses so that the CVS server is the sole manager of the files on the
server!
>
>Please help me out its in the last stage I guess then my setup 
>should be fully functional.

I don't believe you are anywhere near to what you think! You
definitely need to read a good book on CVS itself to understand how it
works. There is a good web based introduction here:
http://cvsbook.red-bean.com/cvsbook.html

It is UNIX centric, but accurate. The fact that you use WinCvs as a
graphic front-end does not change the way CVS works in the background
and you should know about it in order to administrate the server.

The CVSNT developers at March Hare also have created an eBook that is
more up-to-date and contains information on how to use various
front-ends as well. It is available here (not for free though):
http://www.march-hare.com/cvsnt/features/book/



HTH

/Bo
(Bo Berglund, developer in Sweden)



 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/cvsgui/

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/cvsgui/join
    (Yahoo! ID required)

<*> To change settings via email:
    mailto:[email protected] 
    mailto:[email protected]

<*> To unsubscribe from this group, send an email to:
    [email protected]

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/