I think it's a new issue. Please file a JIRA issue.

On Dec 7, 2007 3:49 AM, Luc Willems <[EMAIL PROTECTED]> wrote:
>
>         Hello all ,
>
> i have some problem using the HttpClient protocol handlers from 2.0.0 M1 
> (from 5/12/2007) .
> i'm creating a kind of HTTP load tool which have around 10 threads using 1000 
> and more connections
> to send HTTP request at a fixed rate (xxx TPS)
>
> on low/medium traffic there is now problem but on high TPS rates when using 
> multible connections i get
> decode errors like :
>
> 2007-12-06 19:45:17,390|main|WARN |minaHTTPLoad.run(39)|banging
> 2007-12-06 
> 19:45:20,723|pool-4-thread-1|ERROR|HTTPAgent.exceptionCaught(59)|error 
> :Unsupported HTTP version:  (Hexdump: 00 A1 00 00 00 00 00 80 80 80 BF ... )
>         at 
> org.apache.mina.filter.codec.http.HttpCodecUtils.throwDecoderException(HttpCodecUtils.java:156)
>         at 
> org.apache.mina.filter.codec.http.HttpVersionDecodingState$1.finishDecode(HttpVersionDecodingState.java:46)
>         at 
> org.apache.mina.filter.codec.statemachine.ConsumeToDynamicTerminatorDecodingState.decode(ConsumeToDynamicTerminatorDecodingState.java:83)
>         at 
> org.apache.mina.filter.codec.http.HttpVersionDecodingState.decode(HttpVersionDecodingState.java:62)
>         at 
> org.apache.mina.filter.codec.statemachine.DecodingStateMachine.decode(DecodingStateMachine.java:77)
>         ....
>         at 
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:907)
>         at java.lang.Thread.run(Thread.java:619)
>
> the request i send is http://<server>/gif/l_arrow.gif
>
> now from the hexdump it seems that the decoder is decoding the gif data as 
> the first HTTP response line.
> i checked the communication with wireshark , and response of the apache 
> server was OK.
> i use HTTP keep-alive .
>
> i tried to "lock" the sending thread until the response of the previous 
> request was recieved. this helps a bit but there are still some Exceptions.
> main disadvantage is ofcourse  that it also "blocks" my thread so , NIO (non 
> blocking IO) is not really the case here.
>
> Is this a know issue with the current MINA 2.0.0 code base ?
>
>         luc
>
>
>



-- 
what we call human nature is actually human habit
--
http://gleamynode.net/
--
PGP Key ID: 0x0255ECA6

Reply via email to