Jeff Greenland wrote:
> I'm sure this doesn't help, but we had the same problem with
> the 3.x series.  It went away when we upgraded to 4.x and 5.x.

Yoo-hoo!  Just struggled my way through setting up Tomcat 5.0.16 (never
could figure out how to get mod_jk2 to work, so fell back to mod_jk).
And the problem seems to have disappeared.  Don't really understand how,
as I mentioned in my original note if I bypass Apache and go to directly
to Tomcat (3.2.3), via :8080, I didn't see the problem.  That led me to
conclude that Tomcat wasn't the problem.  But I'm not going to argue
with success.  Thanks, Jeff (and all others who offered a suggestion.)

> 
> Good luck,
> 
> Jeff
> 
> 
> -----Original Message-----
> From: Guy Rouillier [mailto:[EMAIL PROTECTED]
> Sent: Monday, January 19, 2004 5:36 PM
> To: Tomcat Users List
> Subject: Mozilla showing JSP source code
> 
> 
> I've tried to do due diligence on this issue, searching the
> archives as well as Google.  I'm sure it is a common problem,
> but I found several questions and no definitive responses, so
> here goes.  Our website works fine with IE, but we're having
> a significant problem with Mozilla (and derivatives like
> Galeon).  I've tried various versions, including 1.5 as well
> as the brand new 1.6.  I'm seeing this problem both from a
> Windows XP/2000 host and a Solaris host.  I've tried Mozilla
> both on Windows XP clients and Mandrake Linux 9.2 i586
> clients.  All exhibit the same behavior.
> 
> As the title says, when using Mozilla, I'll frequently see
> source code in the browser window.  If I hit reload, in most
> cases, I'll see the page properly displayed (99% of the time -
> rarely, I'll see the source again.)  We are still using Apache
> 1.3.27, Tomcat 3.2.3 and 
> mod_jk 1.2.4.  Here, for example, is one I can produce very
> regularly: ======= HTTP/1.1 200
> Date: Tue, 20 Jan 2004 00:14:25 GMT
> Server: Apache/1.3.26 (Unix) mod_jk mod_ssl/2.8.9 OpenSSL/0.9.6c
> Keep-Alive: timeout=15, max=99
> Connection: Keep-Alive
> Transfer-Encoding: chunked
> Content-Type: text/html;charset=ISO-8859-1
> 
> 1172
> 
> <html>
> =======
> 
> I'm still working on getting Tomcat 5.0.19 configured with
> JK2, in the hopes this will magically go away, but I've
> having problems getting JK2 configured properly.  Here are
> some things I've tried or noticed:
> 
> (1) I first tried going directly to Tomcat, bypassing Apache (using
>> 8080) and this works.  All our pages are displayed in
> Mozilla without any problem.  That leads me to conclude that
> the problem is either in Tomcat delivering the pages to
> Apache via mod_jk, or in Apache delivering the pages to the
> browser.  The first seems more likely.
> 
> (2) We specify no buffer clause on our page directive.  On
> some pages, specifying buffer="64kb" seems to work.
> Frustratingly, this solution seems to work on some pages but
> not others, and on some systems and not others for the same page.
> 
> (3) I also tried changing the KeepAliveTimeout in Apache
> httpd.conf. The default value is 15.  As a test, I upped it
> to 150.  Again, this solved the problem on some pages on some
> clients, but not reliably or predictably.  It also caused
> Mozilla to spin its wheels for about the whole 2 1/2 minutes.
> 
> (4) The most reliable way to see this fail is via a redirect.
>  This happens most frequently on the secure half of our
> website (https).  On those pages, we have an authentication
> header at the top of each page:
> 
> <%@ include file="/includes/authenheader.jspf" %>
> 
> Inside this file, we check some session variables, and if
> they don't have the right set of values (or those values
> don't exist) we
> response.sendRedirect() to a login page.  During this redirect, I
> *always* see the source for the login page - the login page
> has not come up cleanly one time.  This page is very small,
> and neither the buffer or timeout changes help.
> 
> If anyone has any ideas on how to address this problem, I'm
> willing to try anything.  I really am out of ideas and don't
> know where to go next. Thanks for any help.


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to