[cvsnt] Problem committing file.
gordon.boast at ieee.org
Mon May 5 11:46:43 BST 2008
From: Gordon Boast [ <mailto:gordon.boast at ieee.org>
mailto:gordon.boast at ieee.org]
Sent: Sunday, May 04, 2008 3:17 PM
To: gordon.boast at ieee.org
Subject: Re: Problem committing file.
On Sat, 03 May 2008 17:41:01 +0200 Bo Berglund <bo.berglund at telia.com>
>It is *not* an error.....
>Someone has changed te readonly property of the file in his sandbox by
issuing the cvs edit command.
>THis makes the file read/write in his sandbox. But it also tells CVSNT that
this user is the owner of
>the edit token on the file and that noone else may commit any changes to
the file until he is done
>and has committed his own changes.
>So you need to find out who has done this and tell him to commit or unedit
Thank you for your comments. The file in question did checkout as
read-only. As stated in my original description, I changed this by
clearing the read-only flag in the file's properties window. I am the only
person with the file checked out. (I am the only person using this module)
How can someone else own the edit token? Based on these comments I issued an
"edit" for myself -- this allowed me to perform the commit!
It seems to me that the real issue is that some of the files in this module
checkout in readonly. As stated in my original email, this behavior started
when I changed access protocols. Does anyone know why this is happening? How
can I get *all* of the files to checkout so that I am the default editor of
>By the way, your server is set up wrongly.
>Your CVSROOT connect string should not contain any drive specification like
you have now. CVSNT
>should manage the location of the repository all by itself and only publish
a symbolic name of
Yes. I don't know how this came to be and I don't know how to fix it. I read
somewhere that this can cause problems for UNIX/LINUX. We are entirely
Windows here. Still, I would like to fix it if someone can tell me how.
Thanks for the help,
More information about the cvsnt