[cvsnt] RE: cvsnt Digest, Vol 6, Issue 41

Bo Berglund bo.berglund at telia.com
Wed Jun 25 07:02:49 BST 2003


Ask these persons if they did a cvs update on their sandbox before
starting the commit. Typically CVS is rather terse in its messages
when procedure has not been followed. I have seen this also when I
have a binary file that needs to be committed but the local revision
is not HEAD on the server (someone else has committed his own changes
before me). In this case the response is this very non-informative
message.
I think that if the same happens with a normal text file then CVS will
tell you that you need to update first.

But update of the sandbox in this case will replace your edited xls
file with the one from the server and you will have to get your
version back by renaming the file that cvs moves it into (something
starting with a #). And of course this destroys whatever changes the
other person has done...
A side effect of the fact that binary files cannot be concurrently
edited and version controlled.

/Bo


On Tue, 24 Jun 2003 15:46:25 -0700, "Drew F. Sousa"
<DrewS at EDIFECS.COM> wrote:

>Hello, I'm having a problem I would like some help with.  Recent migration
>to CVSNT has gone pretty smoothly and most of my users are fine.
>
>We have users in Eastern Europe who are getting this message:
>
>*****CVS exited normally with code 1*****
>
>cvs -z9 commit -m "Add Case ID for  X12 and DEIFACT syntax rule\n( 09 , 29
>- Invalid syntax rule)" templateupdates.xls (in directory
>Z:\eTools\Validator\Docs\ECErrorReportDocs\)
>cvs [commit aborted]: end of file from server (consult above messages if
>any)
>
>
>
>
>Any help would be appreciated.
>
> 
>Drew	
>DrewSousa - AIM
>


/Bo
(Bo Berglund, developer in Sweden)


More information about the cvsnt mailing list