httpd-users mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Zac Stevens <...@cryptocracy.com>
Subject Re: [users@httpd] Vary: Host mess
Date Wed, 06 Nov 2002 08:21:50 GMT
Hi Jurgen,

On Wed, Nov 06, 2002 at 02:57:50AM -0500, Jurgen wrote:
> It is the mod_rewrite thing:
> You see, for some reason, mod_rewrite adds a header "Vary: Host" to the web server output.

This behavior complies with the HTTP spec, although from what I have read
it seems like it shouldn't be doing so in recent Apache revisions.

> This was still fine for me when I discovered that browsers reaload files every single
time, even pictures, if the files come from a mod_rewritten directory. So I thought "Fine,
then I will unset this header for non cgi-bin directories". This worked well for a while.

This is certainly broken, but that's a client issue.

> Now it is getting worse, even bad.
> 
> I have spent a long time figuring out how to generate files according to Acrobats FDF
specification through a cgi script. It just didn't work.
> I downloaded the output of my script and compared it with a working fdf file and I didn't
find any difference. Then I used this saved download in my non cgi-bin directory and it worked
as well.
> So I thought it is a header thing. It just has to be. I tried with text/plan and application/vnd.fdf
and it just didn't work.
> I decided to look one last time at the HTML headers and the only wierd thing there was
again this "Vary: Host".
> 
> Would you think that this header would affect the Acrobat Reader? How would be looking
there? What is this header for anyway?

It might well be causing the problem - it wouldn't be the first time that
a client has choked on a perfectly legal response.  I don't really
understand the specifics of what you're trying to do here.  Is the URL
loaded directly from Acrobat, or opened by the browser and passed through
to Acrobat?  If the latter, does this relate to the Acrobat Reader
plugin/control, or the full application?

> Anyway, I removed this header from this non cgi-bin directory and it worked.
> 
> Now I just would like to know what this header is for. It is just a bad header. I want
it to be removed.

I'm guessing you meant that you removed the header from the cgi-bin
requests?

This is a pretty arcane topic so extra details would be useful here.  Which
version of Apache are you running?  Which clients are being used (product
and version)?  Is there a proxy server involved?  What rewrites are being
applied to the CGI requests?

The section of the HTTP/1.1 specification relating to the Vary: header is
http://www.w3.org/Protocols/rfc2616/rfc2616-sec14.html#sec14.44

Thus far I have only found IE-related problems with this header, which are
discussed in the following bug report:
http://bugs.apache.org/index.cgi/full/4118

HTH,


Zac


---------------------------------------------------------------------
The official User-To-User support forum of the Apache HTTP Server Project.
See <URL:http://httpd.apache.org/userslist.html> for more info.
To unsubscribe, e-mail: users-unsubscribe@httpd.apache.org
   "   from the digest: users-digest-unsubscribe@httpd.apache.org
For additional commands, e-mail: users-help@httpd.apache.org


Mime
View raw message