tomcat-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Milt Epstein <>
Subject RE: context configuration file being overwritten
Date Thu, 05 Feb 2004 16:06:06 GMT
On Wed, 4 Feb 2004, Shapira, Yoav wrote:

> >Hate to repost this unchanged, but I'm surprised this hasn't gotten
> >any response.  Any Tomcat developers on the list?  Am I missing
> >something about how context configurations files are supposed to
> >work?  Because otherwise this looks like a bug.
> Yes, there are tomcat developers who watch this list.  There's
> always the possibility no one cares enough to respond to your post,
> or no one is interested enough to research it.

Yes, I'm well aware of that possibility :-).

> If you think it's a bug, feel free to write and suggest a patch.  If

Again, the "fix it yourself" response.  Like I said, I don't want to
get into a discussion of this (I've been in on them in the past :-),
but, I'll just say that that reasoning only goes so far.

> you don't think this is a bug, or don't feel like writing a patch,

Actually, I don't have a problem investigating it -- and in fact I
have started to.  But it would be nice if I first at least got some
insight into how/why things work the way they do and/or a pointer,
even a rough one, to where the relevant code exists.

> change your deployment practices.

Well, in fact, I have changed them (because I've been forced to).  But
I don't like this response.  For one thing, Tomcat is a community
effort, and reporting info about suspected anomalies/bugs is one way
of participating in that community effort.

>                                    Personally I don't deploy
> anything other than a packed WAR file to tomcat's webapps directory.
> I don't make use of symlinks nor put anything in the conf directory,
> as support for these is not mandated by the servlet specification
> and I want my app and its deployment to be portable.

This is another Pandora's box.  Not everything Tomcat does is
controlled by the spec, so some choices it makes, and some choices you
make in using it, are not going to be covered by the spec, and hence
subject to the potential of not being portable.  Further, if Tomcat
behaves some way, I think it is important to try to make sure it
behaves that way consistently, and is not subject to misbehaviors
(whether they're at the design level or the implementation level).

Milt Epstein
Research Programmer
Integration and Software Engineering (ISE)
Campus Information Technologies and Educational Services (CITES)
University of Illinois at Urbana-Champaign (UIUC)

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message