>From your google results, this link may be more enlightening:
http://support.microsoft.com/kb/q234067/#XSLTH3136121122120121120120

But then, there is: http://support.microsoft.com/kb/165150/

This is why I am concerned that simply removing 'Pragma: no-cache'
from the HTTP headers may cause stale data problems.

Thanks again,
Cameron 

> -----Original Message-----
> From: P T Withington [mailto:[EMAIL PROTECTED] 
> Sent: Friday, December 02, 2005 9:17 AM
> To: Cameron Brown
> Cc: [email protected]
> Subject: Re: [Laszlo-dev] Dataset loading problem over HTTPS in IE6
> 
> GTFW for 'ie cache control https' reveals:  http://
> support.microsoft.com/default.aspx?scid=kb;en-us;812935
> 
> It would appear to be an IE bug.
> 
> On 1 Dec 2005, at 21:13, Cameron Brown wrote:
> 
> >
> > Ye List,
> >
> > Executive Summary:
> >
> > Dataset loading does not work if communicating over HTTPS 
> in IE6 and 
> > the server sends a 'Pragma: no-cache' HTTP header.
> >
> > The Details:
> >
> > I've been having trouble with my serverless lzx app 
> breaking when run 
> > over HTTPS in IE6.  It works fine over HTTP in IE, and fine 
> for both 
> > HTTP and HTTPS in Firefox 1.0.7.
> >
> > I first noticed the problem a few months ago, using Laszlo 3.0.  It 
> > did not go away in 3.0.2.  I let it slide for a while since Firefox 
> > provided a workaround if HTTPS was required, but the 
> coincidence of a 
> > disgruntled IE-wed customer and the release of Laszlo
> > 3.1 brought it back to my attention.  The basic problem was 
> that over 
> > HTTPS in IE6, my datasets which load XML data from my app 
> server were 
> > not getting any data when making requests.  Instead I'd 
> just get the 
> > (3.1) error:
> >
> > LzLoadQueue.serverlessOnDataHandler load failed from URL ' /cmd/ 
> > authinfo ', no data received.
> > Failure to load data in serverless apps may be caused by 
> Flash player 
> > security policies. Check your data server crossdomain.xml file
> >
> > I only use server-relative paths for URLs, and loading 
> works fine in 
> > the other client environments, so I knew it wasn't actually a 
> > crossdomain.xml problem.  Also I could see that the server 
> was getting 
> > the request properly and replying properly.  So I tried 
> messing around 
> > with the headers of the reply and lo and behold getting rid of 
> > 'Pragma: no-cache' fixed the problem.
> >
> > The only thing I'm worried about now is whether dropping the
> > 'Pragma: no-cache' is now going to cause some version of 
> some browser 
> > to start caching the data coming from the server -- that would be a 
> > disaster.  I am still able to use the headers:
> >
> >   Expires: Sat, 1 Jan 2000 00:00:00 GMT
> >   Cache-Control: no-store, no-cache, must-revalidate, post-check=0, 
> > pre-check=0
> >
> > They haven't caused any trouble.  And I suspect that they will be 
> > sufficient to prevent caching in important browsers, right?  Any 
> > reassurances will be appreciated.
> >
> > Anyway, here's some code to demonstrate the problem:
> >
> > ===============================================================
> > PHP for the server side ("/prag-nc-test.php"):
> > ===============================================================
> > <?
> > // Pragma: no-cache breaks serverless loads under https in 
> IE, // but 
> > works for http in IE and http and https in Firefox
> >
> > header('Pragma: no-cache');
> >
> > print('<result>Irrelevant</result>');
> > ?>
> > ===============================================================
> >
> > ===============================================================
> > LZX Client:
> > ===============================================================
> > <canvas height="350" width="100%" debug="true" proxied="false">
> >   <debug x="5" y="40" height="300" width="800"/>
> >   <dataset type="http" name="testds"
> >           autorequest="false"
> >           ontimeout="canvas.handleTimeout()"
> >           onerror="canvas.handleError()"
> >           ondata="canvas.handleData()"
> >           src="/prag-nc-test.php"/>
> >
> >   <method name="handleData">
> >     Debug.write((new Date())+": Request succeeded.");
> >   </method>
> >   <method name="handleError">
> >     Debug.write((new Date())+": Request got an error -- sorry.");
> >   </method>
> >   <method name="handleTimeout">
> >     Debug.write((new Date())+": Request timed out (after 
> "+(getTimer 
> > ()-canvas.reqTime)+" ms) -- sorry.");
> >   </method>
> >
> >   <view x="5" y="5">
> >     <text valign="middle">Request URL:</text>
> >     <edittext width="300" id="srcTxt" text="$once 
> > {canvas.datasets.testds.src}"/>
> >     <button text="Start Request">
> >       <method event="onclick">
> >         canvas.reqTime = getTimer();
> >         var d = canvas.datasets.testds;
> >         if (srcTxt.text) {
> >           Debug.write((new Date())+": Setting URL to '"+srcTxt.text
> > +"'...");
> >           d.setAttribute('src', srcTxt.text);
> >           Debug.write((new Date())+": value set.");
> >         }
> >         Debug.write((new Date())+": Sending request to 
> "+d.src+"...");
> >         d.doRequest();
> >       </method>
> >     </button>
> >     <simplelayout axis="x"/>
> >   </view>
> > </canvas>
> > ===============================================================
> >
> > I created a JIRA issue for the problem: http://www.openlaszlo.org/
> > jira/browse/LPP-1156
> >
> > Regards,
> > Cameron Brown
> >
> > _______________________________________________
> > Laszlo-dev mailing list
> > [email protected]
> > http://www.openlaszlo.org/mailman/listinfo/laszlo-dev
> 

_______________________________________________
Laszlo-dev mailing list
[email protected]
http://www.openlaszlo.org/mailman/listinfo/laszlo-dev

Reply via email to