Lack of performance

[Trustin Lee]

I've just checked in the fix for the problem you have reported.  Detailed
analysis of the root cause is described here:

  *  https://jira.jboss.org/jira/browse/NETTY-114
https://jira.jboss.org/jira/browse/NETTY-114 

I'm also attaching the modified version of your test application, which
fixes resource leak.  If you encounter too many sockets with TIME_WAIT
state, you might want to turn on 'net.ipv4.tcp_tw_recycle' in Linux.

Download:  http://n2.nabble.com/file/n2254178/test-fixed.zip test-fixed.zip 

HTH,
Trustin


Trustin Lee-3 wrote:
> 
> Hi, I was able to reproduce the problem and I'm currently looking into
> the problem.  The work around at the moment is to wait for the
> ChannelFuture returned by Channel.getCloseFuture() for each connection
> attempt.  Let me get back to you when this issue is fixed.
> 
> Thanks!
> 
> — Trustin Lee, http://gleamynode.net/
> 
> 
> 
> On Tue, Jan 27, 2009 at 8:28 PM, jaimemm <jaimemm at gmail.com> wrote:
>>
>> Hi Trustin,
>>
>> Please find attached the source code.
>>
>> Configure the client via the clientProperties file.
>>
>> I experienced lack of performance with 1 client, 10000 connections and 10
>> repeats.
>>
>> If un-comment the content of messageReceived method inside the Server
>> RequestHandler and configure the tester with: 3 clienta and 10000
>> connections I get the Exception of my other post.
>>
>>
>> With regar http://n2.nabble.com/file/n2223728/test.zip test.zip ds,
>>
>>
>>
>> Trustin Lee-3 wrote:
>>>
>>> What version of Netty are you using?  Could you try the latest SVN
>>> snapshot?  How was the CPU / RAM consumption?
>>>
>>> Also, it would be great if you can provide something that lets me
>>> reproduce the problem by myself.
>>>
>>> — Trustin Lee, http://gleamynode.net/
>>>
>>>
>>>
>>> On Mon, Jan 26, 2009 at 9:44 PM, jaimemm <jaimemm at gmail.com> wrote:
>>>>
>>>> Hi Trustin,
>>>>
>>>> WIth the idea of reaching a cut in terms of connections, I experienced
>>>> a
>>>> lack of performance under the following circumstances:
>>>>
>>>> - Running the client and the server within the same machine.
>>>> - The client sends requests continously (Factory shared for all the
>>>> request
>>>> and built with default parameters)
>>>> - The server does not answer none of them (on purpose)
>>>>
>>>>
>>>> After 20000 the client and servers szlows down amazingly.
>>>>
>>>>
>>>> Any ideas?
>>>>
>>>> With regards,
>>>> --
>>>> View this message in context:
>>>> http://n2.nabble.com/Lack-of-performance-tp2217685p2217685.html
>>>> Sent from the Netty User Group mailing list archive at Nabble.com.
>>>>
>>>> _______________________________________________
>>>> netty-users mailing list
>>>> netty-users at lists.jboss.org
>>>> https://lists.jboss.org/mailman/listinfo/netty-users
>>>>
>>>
>>> _______________________________________________
>>> netty-users mailing list
>>> netty-users at lists.jboss.org
>>> https://lists.jboss.org/mailman/listinfo/netty-users
>>>
>>>
>>
>> --
>> View this message in context:
>> http://n2.nabble.com/Lack-of-performance-tp2217685p2223728.html
>> Sent from the Netty User Group mailing list archive at Nabble.com.
>>
>>
>> _______________________________________________
>> netty-users mailing list
>> netty-users at lists.jboss.org
>> https://lists.jboss.org/mailman/listinfo/netty-users
>>
> 
> _______________________________________________
> netty-users mailing list
> netty-users at lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/netty-users
> 
> 


-----
Trustin Lee, Principal Software Engineer, JBoss, a division of Red Hat
--
what we call human nature is actually human habit
--
http://gleamynode.net/
-- 
View this message in context: http://n2.nabble.com/Lack-of-performance-tp2217685p2254178.html
Sent from the Netty User Group mailing list archive at Nabble.com.