httpd-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Robert Steinmetz <>
Subject Re: [users@httpd] Reverse Proxy URL confuion.
Date Thu, 05 Feb 2009 04:04:27 GMT
Matt McCutchen wrote:
> On Wed, 2009-02-04 at 17:00 -0600, Robert Steinmetz AIA wrote:
>> Matt McCutchen wrote:
>>> Two issues:
>>> 1. The error you pasted suggests that your ProxyHTMLURLMap rule is not
>>> taking effect at all.  My guess is that you need to include the
>>> configuration file that defines all the places links appear in HTML:
>> I agree it looks like the ProxyHTMLURLMap has no effect.
>> However, in /etc/apache2/mods-available I have proxy_html.conf and  
>> proxy_html.load so it should be working unless something else needs to 
>> be done in my configuration to enable it
>> The proxy_html.conf appears identical to the link you provided, and I am 
>> not getting any errors when I load apache or when I test the 
>> configuration, so I think it is being recognized.
> OK, good.
>>> 2. Your ProxyHTMLURLMap rule is backwards, since you want to *remove*
>>> the /application prefix from the links generated by the back-end (which
>>> is the reverse of the scenario in the HOWTO, as you can see by comparing
>>> your ProxyPass directive to those in the HOWTO).  The rule should be:
>>> ProxyHTMLURLMap / /application
>>> (I don't understand what the capital S was doing at the end of your
>>> rule.)
> Sorry, that was a bad thinko; I got distracted by the S and didn't get
> the pattern and replacement in the right order.  The rule should be:
> ProxyHTMLURLMap /application /
OK changed it back, no luck.
> Try adding another rule in case the link is a full URL, like case 3 of
> the HOWTO:
> ProxyHTMLURLMap /he URL is 
Added it, No Luck
> If it still doesn't work, use your browser's command to view the HTML
> source of the login page and look for the link that contains
> the /application string.  You'll need to design a ProxyHTMLURLMap rule
> based on that link.
I've looked at the page source and the references seem to be generated 
by calling scripts. The links reported in the buttons are in the form


I'm beginning to think that the links are generated by the application 
in a way that isn't being caught.
*Robert Steinmetz, AIA*
*Steinmetz & Associates*

The official User-To-User support forum of the Apache HTTP Server Project.
See <URL:> for more info.
To unsubscribe, e-mail:
   "   from the digest:
For additional commands, e-mail:

View raw message