Re: Grizzly revision 4420 broke Atmosphere rest-chat sample

From: Sebastien Dionne <survivant00_at_gmail.com>
Date: Mon, 26 Apr 2010 06:28:54 -0400

please. do the test after the login... there are no problem before the
login. I told the procedure how the reproduce the problem in my first post.

there are no point in your test unless you do the procedure to reproduce the
problem.

I know that is only the content-type that change.. but you haev at least to
test the problem.

until you do that.. you won't be able to prove that Grizzly didn't broke
somewhere else.

2010/4/26 Oleksiy Stashok <Oleksiy.Stashok_at_sun.com>

> Sebastien,
>
> this is output from wireshark log for HTTP connection, which gets
> initialized before login.
> As I understand this is the connection, which Atmosphere uses for
> server->client notifications. And the only difference I see, is that latest
> Grizzly includes content-type header in the HTTP response for this
> connection.
>
> Otherwise, as I told, I didn't see any other difference in client<->server
> communication, using different Grizzly versions.
>
> WBR,
> Alexey.
>
>
> On Apr 26, 2010, at 2:57 , Sebastien Dionne wrote:
>
> just one question.
>
> these output are from the submit login ou simply the get from the
> index.html ?
>
> the problem appears only after you try the login button.
>
>
>
>
> 2010/4/23 Oleksiy Stashok <Oleksiy.Stashok_at_sun.com>
>
>> Hi,
>>
>> the only difference, caused by that commit is the response for the
>> following request:
>> HTTP GET /atmosphere-rest-chat/chat?0 HTTP/1.1
>>
>> The original (rev<4220) returns [1], rev >= 4220 returns [2].
>> As you see in the later, Grizzly added additional Content-Type header.
>> According to wireshark the rest of communication for both revisions goes
>> absolutely the same way.
>> So, as I understand, something unexpected is happening on a client side
>> (browser or javascript), so it doesn't tolerate content-type header, or its
>> value. Is it possible to fix on a client side?
>>
>> Thanks.
>>
>> WBR,
>> Alexey.
>>
>> [1]
>> HTTP/1.1 200 OK
>> server: grizzly/1.9.19
>> Expires: -1
>> Cache-Control: no-store, no-cache, must-revalidate
>> Pragma: no-cache
>> Transfer-Encoding: chunked
>> Date: Fri, 23 Apr 2010 21:35:34 GMT
>>
>> 834
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>>
>>
>>
>> [2]
>> HTTP/1.1 200 OK
>> server: grizzly/1.9.19
>> Expires: -1
>> Cache-Control: no-store, no-cache, must-revalidate
>> Pragma: no-cache
>> *Content-Type: text/plain; charset=iso-8859-1*
>> Transfer-Encoding: chunked
>> Date: Fri, 23 Apr 2010 21:27:59 GMT
>>
>> 834
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>> 
>>
>>
>>
>>
>> On Apr 23, 2010, at 19:06 , Sebastien Dionne wrote:
>>
>> Always fails >= 4420
>>
>> Always works with revisions below 4420.
>>
>> On Apr 23, 2010 1:03 PM, "Oleksiy Stashok" <Oleksiy.Stashok_at_sun.com>
>> wrote:
>>
>> Hi,
>>
>> I'm not saying that it's definitely not a Grizzly issue, it could be :)
>>
>> But not sure how rev. 4420 caused that.
>> BTW, the default content-type, which we set if non was set before (unless
>> it's not explicitly changed) is: "text/plain; charset=iso-8859-1".
>> Sebastien, is this bug constantly reproducible, it *always* fails?
>>
>> Thanks.
>>
>> WBR,
>> Alexey.
>>
>>
>>
>> On Apr 23, 2010, at 18:14 , jfarcand wrote:
>>
>>> > Salut,
>>> >
>>> > On 23/04/10 12:22 AM, Oleksiy Stashok wrote:
>>> >>
>>> >> Hi Sebastien,
>>> >>
>>> >> that fix was req...
>>> To unsubscribe, e-mail: dev-unsubscribe_at_grizzly.dev.java.net
>>> For additional commands, e-mail: dev-help_at_grizzly.dev.java.net
>>>
>>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe_at_grizzly.dev.java.net
>> For additional commands, e-mail: dev-help_at_grizzly.dev.java.net
>>
>>
>>
>>
>
>
> --
> -------------
> A+
>
> Sébastien.
>
> Vous pouvez me suivre sur Twitter / You can follow me on Twitter :
> http://twitter.com/survivant
>
>
>

-- 
-------------
A+
Sébastien.
Vous pouvez me suivre sur Twitter / You can follow me on Twitter :
http://twitter.com/survivant