Mailing-List: contact test-dev-help@httpd.apache.org; run by ezmlm
Precedence: bulk
Reply-To: test-dev@httpd.apache.org
Date: Tue, 18 Jun 2002 08:21:53 -0700
From: Aaron Bannert <aaron@clove.org>
To: test-dev@httpd.apache.org
Subject: Re: [PATCH] flood & verify_200
Message-ID: <20020618082153.G21255@clove.org>
References: <20020618173830.22b4a8da.jacek.prucia@7bulls.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20020618173830.22b4a8da.jacek.prucia@7bulls.com>
User-Agent: Mutt/1.3.23i

On Tue, Jun 18, 2002 at 05:38:30PM +0200, Jacek Prucia wrote:
> 
> I think current flood (tried HEAD) have problems with veryfing
> responses. In particular it marks 200 OK responses as 'FAIL' if they are
> sent with HTTP/1.0 rather than expected 'HTTP/1.1'. Looks like problem
> is with memcpy verification in function verify_200. To work around that
> problem i just wrote a little patch just to make things work, which i'm
> sending here for verification/blessing/commit. It might turn out that
> the patch is a piece of crap -- mostly because I'm really new both to
> apr and flood (i'm not sure if that change in profile_event_handlers is
> wise thing to do...). However it solves serious (at least for me)
> problem, and if patch is wrong -- it has to be done other way.

I will take a look at this patch and apply in the next few hours.

> Besides that, flood is a very useful tool, and i'm using it to do
> regression tests for over-sized web application. Looks like flood could
> use some more patches to be more suited for this task. In particular, it
> would be cool if flood could stop if particular url (or any url as an
> option) returns FALSE. This is what i'm going to work on in near future.

Thanks! Patches are appreciated, especially if you're also working on
boated webapps. ;)

> One more thing... looks like flood build system is slightly broken. You
> can build it, but there's no way to install it in a target directory
> (make install fails). Is this a known problem?

The install target just hasn't been implemented because at the moment
we only have one file that would need to be installed. If it would be
useful, I'm sure it could be implemented quickly (actually, I was
thinking about this just the other day).

-aaron