On Mon, Jun 30, 2003 at 02:35:57PM -0400, Neil Schemenauer wrote:
> On Mon, Jun 30, 2003 at 04:51:23PM +0200, Hugo van der Merwe wrote:
> > I'm now getting the odd effect that when I click on the link
> > submitting the form at url A a second time, I just get A again,
> > reloaded. _q_lookup is called for url B, my page rendering code is
> > called, the apache logs seem to indicate (by page size) that the page
> > page is being returned properly, with status code 200, why would
> > mozilla take be directly back to url A then?
>
> I think you need to determine exactly what are the requests and
> responses.  The best is to use something like ethereal to capture all
> the traffic.  If you can't do that, looking at the URLs and status codes
> in the Apache access log will tell you more.

I had looked at the apache logs (as I mentioned), but I have now also
checked the headers - using Mozilla's "Live HTTP headers"
(Tools->Web Development->Live HTTP headers) - the times when an
incorrect redirect happens looks *identical* to when it doesn't
happen (other than Date and Keep-Alive values). And then it doesn't
*always* result in an incorrect redirect, though I'm not sure what
conditions determine whether it will redirect or not.

I've considered adding headers along these lines:

Expires: Thu, 19 Nov 1981 08:52:00 GMT
Cache-Control: no-store, no-cache, must-revalidate, post-check=0, pre-check=0
Pragma: no-cache

(as seen from a php app's headers), but I will probably only bother to
do this when the problem really bothers me (or my users), or manifests
with e.g. Internet Explorer as well. It is looking like a mozilla bug
to me though. I'll play with Konqueror a bit as well.

Thanks,
Hugo van der Merwe
--
The earth's surface is accelerating outward at 9.8 meters per sec^2
1024D/60715698: 5F2E 8EC2 E0A4 5D25 0569  F281 4A6C D76D 6071 5698