|
Home > Archive > Apache Server configuration support > July 2006 > Weird IE6 / header? problems
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Weird IE6 / header? problems
|
|
|
| Hi all!
I'm having some problems and I don't know how to debug them or search for a
solution on Google. I am not sure this is the best newsgroup to ask, but I
suspect the culprit is Apache config, so...
I have built a site nad it works OK on my server. The pages are simple and I
go to great lengths to ensure they are valid XHTML and contain no dubious
JS.
When I uploaded it to the ISP server and tested it and it worked ok - on all
browsers except IE6 that is. On IE some of the pages flicker and then show
themselves, some of them just keep flickering,... And: if I "view source"
on the ones that show themselves I get incomplete source.
The page, needless to say, opens OK if I save it on disk and open it in IE.
I checked in FF (Linux and Windows), Konqueror, even wget - and all show no
signs of error. I even checked the response headers and they look OK to me:
$ wget -S -U "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1; .NET
CLR 1.1.4322)" http://some.domain.com/index.php
....
HTTP request sent, awaiting response...
HTTP/1.1 200 OK
Date: Thu, 27 Jul 2006 09:44:16 GMT
Server: Apache
X-Powered-By: PHP/4.4.2
Cache-Control: no-store, no-cache, must-revalidate, post-check=0,
pre-check=0
Set-Cookie: PHPSESSID=840b0915f65905c13657523796ea03
91; path=/
Expires: Thu, 19 Nov 1981 08:52:00 GMT
Pragma: no-cache
Vary: Accept-Encoding
Connection: close
Content-Type: text/html; charset=ISO-8859-2
Length: unspecified [text/html]
What the &#$>?[@}#! is going on? Some content negotiation error? Problems
with the server configuration? Could it be the "Vary:" header?
I hope someone can help me out - I've run out of ideas. 
Best,
Anze
| |
|
| "Anze" <anzenews@volja.net> schreef in bericht
news:w20yg.4451$oj5.1558087@news.siol.net...
> When I uploaded it to the ISP server and tested it and it worked ok
> - on all browsers except IE6 that is.
> On IE some of the pages flicker and then show
> themselves, some of them just keep flickering,...
> And: if I "view source" on the ones that show themselves I get
> incomplete source.
Incomplete? Missing fragment(s) or missing tail or ... ?
What is the size received and expected?
Might some files have been corrupted during upload?
Check end-of-line conversion on Windows to *nix uploads.
> I checked in FF (Linux and Windows), Konqueror, even wget - and all show
no
> signs of error. I even checked the response headers and they look OK to
me:
>
> $ wget -S -U "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1; SV1; .NET
> CLR 1.1.4322)" http://some.domain.com/index.php
> ...
> HTTP request sent, awaiting response...
> HTTP/1.1 200 OK
> Date: Thu, 27 Jul 2006 09:44:16 GMT
> Server: Apache
> X-Powered-By: PHP/4.4.2
> Cache-Control: no-store, no-cache, must-revalidate, post-check=0,
> pre-check=0
> Set-Cookie: PHPSESSID=840b0915f65905c13657523796ea03
91; path=/
> Expires: Thu, 19 Nov 1981 08:52:00 GMT
> Pragma: no-cache
> Vary: Accept-Encoding
> Connection: close
> Content-Type: text/html; charset=ISO-8859-2
> Length: unspecified [text/html]
>
>
> What the &#$>?[@}#! is going on? Some content negotiation error? Problems
> with the server configuration? Could it be the "Vary:" header?
Unlikely, http://www.w3.org/Protocols/rfc2616...4.html#sec14.44
tells
' The Vary field value indicates the set of request-header fields that fully
determines, while the response is fresh, whether a cache is permitted to use
the response to reply to a subsequent request without revalidation.'
Content-Encoding commonly reveils to the client the use of serverside
compression.
This compression may cause problems on some versions and brands of browsers
if applied to content other then xml, xhtml or html-pages, eg linked
stylesheets and javascript
http://httpd.apache.org/docs/2.0/mod/mod_deflate.html
If your page is generated by a php script doing compression _and_ your
provider is doing that two, some browsers may fail ...
No more suggestion untill the real URL is disclosed here
HansH
| |
|
|
HansH, thank you for your reply!
> Incomplete? Missing fragment(s) or missing tail or ... ?
> What is the size received and expected?
> Might some files have been corrupted during upload?
> Check end-of-line conversion on Windows to *nix uploads.
When I click View -> Source in IE I get Notepad with source, but about half
of it (at the end) is missing.
I repeat: this is strictly IE6 thing, other browsers work OK, so I guess it
must be some HTTP connected problem.
> This compression may cause problems on some versions and brands of
> browsers if applied to content other then xml, xhtml or html-pages, eg
> linked stylesheets and javascript
> http://httpd.apache.org/docs/2.0/mod/mod_deflate.html
>
> If your page is generated by a php script doing compression _and_ your
> provider is doing that two, some browsers may fail ...
But that should be visible in headers, right? How can I see if this is the
problem?
> No more suggestion untill the real URL is disclosed here
http://domex.netinet.si/
Again, thank you for the reply - I really appreciate it.
Best,
Anze
| |
|
|
Ha! Solved it!
I tried using lynx and saw in 'status line' that it fetches the pages
from .gz files. So I figured - let's check those .gz files to see if we
have some double compression or sth. like that. And sure enough, the
temporary .gz files had some trailing garbage.
As part of debugging / profiling process I added a php function that
displayed the profiling times:
register_shutdown_function("display_profiling_times");
It looks like output was first encoded and only then was output from this
function appended.
IE6 is clearly not as forgiving as other browsers and was reloading the page
over and over.
Interesting problem - I disabled the output as it has no place in a finished
application anyway, but I sure learnt something today. 
HansH, thank you again for your help. Your advice pointed me in the right
direction.
Best,
Anze
| |
|
| "Anze" <anzenews@volja.net> schreef in bericht
news:2o2yg.4455$oj5.1559464@news.siol.net...
> I repeat: this is strictly IE6 thing, other browsers work OK, so I guess
it
> must be some HTTP connected problem.
>
>
> But that should be visible in headers, right?
> How can I see if this is the problem?
You'll see a response header like
Content-Encoding: gzip
ONLY if the request had this header like
Accept-Encoding: gzip,deflate
However, it does NOT tell you it is done TWICE
>
HansH
| |
|
|
"Anze" <anzenews@volja.net> schreef in bericht
news:Ci3yg.4458$oj5.1560338@news.siol.net...
>
>
> Ha! Solved it!
>
> I tried using lynx and saw in 'status line' that it fetches the pages
> from .gz files. So I figured - let's check those .gz files to see if we
> have some double compression or sth. like that. And sure enough, the
> temporary .gz files had some trailing garbage.
>
> As part of debugging / profiling process I added a php function that
> displayed the profiling times:
> register_shutdown_function("display_profiling_times");
> It looks like output was first encoded and only then was output from this
> function appended.
As per documentation
http://nl3.php.net/manual/nl/functi...wn-function.php
behaviour varies per version of php :-(
BTW your '404 page not found' is missing
HansH
|
|
|
|
|