Googles Chrome and Freescale web server

[alager]

Has anyone tried Chrome on their embedded web servers? I get the error:

This webpage is not available.

The webpage at http://192.168.10.209/ might be temporarily down or it may have moved permanently to a new web address.

More information on this error
Below is the original error message

Error 320 (net::ERR_INVALID_RESPONSE): Unknown error.



Aaron

[bkatt]

---

alager wrote:
Has anyone tried Chrome on their embedded web servers? I get the error:

...

Error 320 (net::ERR_INVALID_RESPONSE): Unknown error.

---

I just tried Chrome on *my* embedded web server, and it works great. The only thing I noticed was that if you use digest authentication, Chrome does not allow a blank user name. Some other browsers (but not all) have the same limitation.

If you are using the emg web browser that freescale provides with niche lite, I'm not surprised that you got an unknown error, as this code does not do what the rfc standards require.

[J2MEJediMaster]

Does your  embedded web server work with other browsers? Of course it does. So, now you have to ask: Does the server side code look for a particular browser type? If there's no match on the browser type, what browser type does the server software default to? Can you change the user-agent ID of Chrome to that of another browser and see if the server responds properly?

---Tom

[alager]

I'm just trying the standard demo that comes with the 52235evb. Looking at index.html, there are no javascript calls, nor any other browser specific stuff, just plain vanilla html.

Yes the web server does work with firefox and IE.

User Agents:

Firefox:
Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.1) Gecko/2008070208 Firefox/3.0.1

Chrome:
Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US) AppleWebKit/525.13 (KHTML, like Gecko) Chrome/0.2.149.29 Safari/525.13


I don't see any way of changing the user agent in chrome.
Aaron

Message Edited by alager on 2008-09-10 06:40 PM

[bkatt]

Before the "plain vanilla html", emg provides the http header, like:

HTTP/1.1 200 OK
Server: EMG/1.1.0
Cache-control: no-cache
HTTP Server By Eric Gregori Version 1.1.0
Content-type: text/html
Content-length: 9640

That 4th line (the long one) is not valid, and you would have to change the "emg_static_ffs" pc program or similar to get rid of it. I don't know whether this is causing your problem with Chrome...

[alager]

I did find that the request for the favicon.ico generated a different response from the server.
***************************
HTTP/1.x 200 OK
Last-Modified: Wed, 15 Mar 2006 10:42:00 GMT
Server: EMG_HTTP 1.0.0
Content-Type: text/html
***************************

From what I can tell the server line should be:
Server: EMG_HTTP/1.0.0

But I'm not sure if that is invalid or not as it is.

Aaron

[alager]

I'm not too sure that's it.  Here are the headers coming from the eval board:

Code:

HTTP/1.x 200 OKServer: EMG/1.1.0Cache-Control: no-cacheContent-Type: text/plainContent-Length: 28

 

This does not seem out of spec...unless there is something subtle I'm missing.

*After I wrote this I realized that first the browser sends a big set of data before the server responds with the above.  The above sample was taken from Firefox.  So it's feasible that the web server is sending something different when Chrome does the GET.



Message Edited by alager on 2008-09-10 11:50 PM