[racket] Web server warnings?

From: Jay McCarthy (jay.mccarthy at gmail.com)
Date: Tue Sep 25 12:22:59 EDT 2012

I just pushed a patch that I think will remove the error message.

Jay

On Mon, Sep 24, 2012 at 3:04 PM, Neil Van Dyke <neil at neilvandyke.org> wrote:
> I think I am seeing a similar error message, with a mobile Web app being
> viewed in Firefox.  Based on the timing, looks like the client is doing
> keepalive (or similar), and eventually timing out the connection.  If so, I
> think that would be normal Web browser behavior, not something that I'd
> expect to see an error message about.
>
> [stop sign icons]
> ../../../../usr/local/racket-5.3/lib/racket/collects/web-server/private/dispatch-server-unit.rkt:74:2:
> read-byte: input port is closed
>
> Neil V.
>
> Jay McCarthy wrote at 09/23/2012 02:46 PM:
>
>> Ya, the problem is that the Web server assumes that every HTTP/1.1
>> client will send multiple requests. So after the first, it keeps
>> listening, but then the client drops and the peek-byte fails. I could
>> probably change the code to also sync on the port closed event and not
>> throw any error, but I haven't invested in doing that. If it really
>> bothers you, I could look doing it.
>>
>> Jay
>>
>> On Sun, Sep 23, 2012 at 7:11 AM, Thomas Chust<chust at web.de>  wrote:
>>
>>>
>>> On Sun, 23 Sep 2012, Jordan Schatz wrote:
>>>
>>>
>>>>
>>>> [...]
>>>>
>>>> For awhile now the webserver spits out abunch of these messages:
>>>>
>>>> peek-byte: input port is closed
>>>>   context...:
>>>>
>>>>
>>>> /home/jordan/bin/lib/racket/collects/web-server/private/dispatch-server-unit.rkt:74:2:
>>>>   connection-loop
>>>>
>>>> I think it only happens when the server is using SSL, is it an
>>>> indication
>>>> that
>>>> my code has a problem, or something inherent in the server?
>>>> [...]
>>>>
>>>
>>>
>>> Hello,
>>>
>>> in my web applications I also see this behaviour, both with and without
>>> active SSL support in the Racket web server and no matter whether the web
>>> application is accessed directly or through another web server acting as
>>> a
>>> proxy.
>>>
>>> I think this is not related to problems in the web application but simply
>>> happens when clients drop connections, but I haven't investigated this in
>>> detail, since it hasn't caused any trouble apart from log spam ;-)
>>>
>>> Ciao,
>>> Thomas
>>>
>>>
>>> --
>>> When C++ is your hammer, every problem looks like your thumb.
>>> ____________________
>>>    Racket Users list:
>>>    http://lists.racket-lang.org/users
>>>
>>>
>>
>>
>>
>>



-- 
Jay McCarthy <jay at cs.byu.edu>
Assistant Professor / Brigham Young University
http://faculty.cs.byu.edu/~jay

"The glory of God is Intelligence" - D&C 93

Posted on the users mailing list.