Connection reset by peer (104)

classic Classic list List threaded Threaded
6 messages Options
Reply | Threaded
Open this post in threaded view
|

Connection reset by peer (104)

Noah Campbell
Running grizzly 1.6.1 http on windows XP

SelectorThread server = new SelectorThread();
                    server.setPort(9999);
                    server.setDisplayConfiguration(verbose);
                    server.setAdapter(new StaticResourcesAdapter("."));
                   
                    server.initEndpoint();
                    server.startEndpoint();

A file named index.html exists in the directory.

ab -c 100 -n 100000 -p file.out -k http://ncampbell-pc:9999/ on RHEL

file.out is 512 bytes.

$ ab -c 100 -n 100000 -p file.out -k http://ncampbell-pc:9999/
This is ApacheBench, Version 2.0.41-dev <$Revision: 1.141 $> apache-2.0
Copyright (c) 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/
Copyright (c) 1998-2002 The Apache Software Foundation, http://www.apache.org/

Benchmarking ncampbell-pc (be patient)
Completed 10000 requests
apr_recv: Connection reset by peer (104)
Total of 12438 requests completed

I'm pretty sure I'm doing something naive, but I need your guidance.

-Noah

Reply | Threaded
Open this post in threaded view
|

Re: Connection reset by peer (104)

Oleksiy Stashok
Hello Noah,

looks like some issue in Grizzly, but there is not enough information.
Is it all info/exceptions you have?

If yes, can you pls. pack and send (or publish) the testing apps., so I
will be able to reproduce this?

Thanks.

WBR,
Alexey.

Noah Campbell wrote:

> Running grizzly 1.6.1 http on windows XP
>
> SelectorThread server = new SelectorThread();
>                     server.setPort(9999);
>                     server.setDisplayConfiguration(verbose);
>                     server.setAdapter(new StaticResourcesAdapter("."));
>                    
>                     server.initEndpoint();
>                     server.startEndpoint();
>
> A file named index.html exists in the directory.
>
> ab -c 100 -n 100000 -p file.out -k http://ncampbell-pc:9999/ on RHEL
>
> file.out is 512 bytes.
>
> $ ab -c 100 -n 100000 -p file.out -k http://ncampbell-pc:9999/
> This is ApacheBench, Version 2.0.41-dev <$Revision: 1.141 $> apache-2.0
> Copyright (c) 1996 Adam Twiss, Zeus Technology Ltd,
> http://www.zeustech.net/
> Copyright (c) 1998-2002 The Apache Software Foundation,
> http://www.apache.org/
>
> Benchmarking ncampbell-pc (be patient)
> Completed 10000 requests
> apr_recv: Connection reset by peer (104)
> Total of 12438 requests completed
>
> I'm pretty sure I'm doing something naive, but I need your guidance.
>
> -Noah
>

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Connection reset by peer (104)

Jeanfrancois Arcand-2
Hi,

Oleksiy Stashok wrote:
> Hello Noah,
>
> looks like some issue in Grizzly, but there is not enough information.
> Is it all info/exceptions you have?

I don't think this is an issue :-) ab is a flooding tool...stressing
Grizzly without any ramp up time will always produce bad results. A
better way to test Grizzly is explained here:

http://weblogs.java.net/blog/sdo/archive/2007/03/ab_considered_h.html

Faban will gives much better results:

http://weblogs.java.net/blog/sdo/archive/2007/04/simple_benchmar.html
http://weblogs.java.net/blog/sdo/archive/2007/05/how_to_test_con.html

Also you might want to increase the number of threads (default is 20).
SelectorThread.setMaxThreads(..) :-)

Thanks!

-- Jeanfrancois


>
> If yes, can you pls. pack and send (or publish) the testing apps., so I
> will be able to reproduce this?
>
> Thanks.
>
> WBR,
> Alexey.
>
> Noah Campbell wrote:
>> Running grizzly 1.6.1 http on windows XP
>>
>> SelectorThread server = new SelectorThread();
>>                     server.setPort(9999);
>>                     server.setDisplayConfiguration(verbose);
>>                     server.setAdapter(new StaticResourcesAdapter("."));
>>                                        server.initEndpoint();
>>                     server.startEndpoint();
>>
>> A file named index.html exists in the directory.
>>
>> ab -c 100 -n 100000 -p file.out -k http://ncampbell-pc:9999/ on RHEL
>>
>> file.out is 512 bytes.
>>
>> $ ab -c 100 -n 100000 -p file.out -k http://ncampbell-pc:9999/
>> This is ApacheBench, Version 2.0.41-dev <$Revision: 1.141 $> apache-2.0
>> Copyright (c) 1996 Adam Twiss, Zeus Technology Ltd,
>> http://www.zeustech.net/
>> Copyright (c) 1998-2002 The Apache Software Foundation,
>> http://www.apache.org/
>>
>> Benchmarking ncampbell-pc (be patient)
>> Completed 10000 requests
>> apr_recv: Connection reset by peer (104)
>> Total of 12438 requests completed
>>
>> I'm pretty sure I'm doing something naive, but I need your guidance.
>>
>> -Noah
>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [hidden email]
> For additional commands, e-mail: [hidden email]
>

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Connection reset by peer (104)

Noah Campbell
Interestingly, the same code scales to ab -c 2000 .... on linux (given the appropriate ulimit support).

-Noah

Jeanfrancois Arcand-2 wrote
Hi,

Oleksiy Stashok wrote:
> Hello Noah,
>
> looks like some issue in Grizzly, but there is not enough information.
> Is it all info/exceptions you have?

I don't think this is an issue :-) ab is a flooding tool...stressing
Grizzly without any ramp up time will always produce bad results. A
better way to test Grizzly is explained here:

http://weblogs.java.net/blog/sdo/archive/2007/03/ab_considered_h.html

Faban will gives much better results:

http://weblogs.java.net/blog/sdo/archive/2007/04/simple_benchmar.html
http://weblogs.java.net/blog/sdo/archive/2007/05/how_to_test_con.html

Also you might want to increase the number of threads (default is 20).
SelectorThread.setMaxThreads(..) :-)

Thanks!

-- Jeanfrancois


>
> If yes, can you pls. pack and send (or publish) the testing apps., so I
> will be able to reproduce this?
>
> Thanks.
>
> WBR,
> Alexey.
>
> Noah Campbell wrote:
>> Running grizzly 1.6.1 http on windows XP
>>
>> SelectorThread server = new SelectorThread();
>>                     server.setPort(9999);
>>                     server.setDisplayConfiguration(verbose);
>>                     server.setAdapter(new StaticResourcesAdapter("."));
>>                                        server.initEndpoint();
>>                     server.startEndpoint();
>>
>> A file named index.html exists in the directory.
>>
>> ab -c 100 -n 100000 -p file.out -k http://ncampbell-pc:9999/ on RHEL
>>
>> file.out is 512 bytes.
>>
>> $ ab -c 100 -n 100000 -p file.out -k http://ncampbell-pc:9999/
>> This is ApacheBench, Version 2.0.41-dev <$Revision: 1.141 $> apache-2.0
>> Copyright (c) 1996 Adam Twiss, Zeus Technology Ltd,
>> http://www.zeustech.net/
>> Copyright (c) 1998-2002 The Apache Software Foundation,
>> http://www.apache.org/
>>
>> Benchmarking ncampbell-pc (be patient)
>> Completed 10000 requests
>> apr_recv: Connection reset by peer (104)
>> Total of 12438 requests completed
>>
>> I'm pretty sure I'm doing something naive, but I need your guidance.
>>
>> -Noah
>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@grizzly.dev.java.net
> For additional commands, e-mail: users-help@grizzly.dev.java.net
>

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@grizzly.dev.java.net
For additional commands, e-mail: users-help@grizzly.dev.java.net
Reply | Threaded
Open this post in threaded view
|

Re: Connection reset by peer (104)

Jeanfrancois Arcand-2
Hi,

Noah Campbell wrote:
> Interestingly, the same code scales to ab -c 2000 .... on linux (given the
> appropriate ulimit support).

Thanks for the report. I'm really not an expert with XP, but could it be
the ulimit is too low as well? Can you run ab using 'ab -v4' it will
logs the exceptions and send them here?

Are you seeing any exception on the server side?

Thanks

-- Jeanfrancois


>
> -Noah
>
>
> Jeanfrancois Arcand-2 wrote:
>> Hi,
>>
>> Oleksiy Stashok wrote:
>>> Hello Noah,
>>>
>>> looks like some issue in Grizzly, but there is not enough information.
>>> Is it all info/exceptions you have?
>> I don't think this is an issue :-) ab is a flooding tool...stressing
>> Grizzly without any ramp up time will always produce bad results. A
>> better way to test Grizzly is explained here:
>>
>> http://weblogs.java.net/blog/sdo/archive/2007/03/ab_considered_h.html
>>
>> Faban will gives much better results:
>>
>> http://weblogs.java.net/blog/sdo/archive/2007/04/simple_benchmar.html
>> http://weblogs.java.net/blog/sdo/archive/2007/05/how_to_test_con.html
>>
>> Also you might want to increase the number of threads (default is 20).
>> SelectorThread.setMaxThreads(..) :-)
>>
>> Thanks!
>>
>> -- Jeanfrancois
>>
>>
>>> If yes, can you pls. pack and send (or publish) the testing apps., so I
>>> will be able to reproduce this?
>>>
>>> Thanks.
>>>
>>> WBR,
>>> Alexey.
>>>
>>> Noah Campbell wrote:
>>>> Running grizzly 1.6.1 http on windows XP
>>>>
>>>> SelectorThread server = new SelectorThread();
>>>>                     server.setPort(9999);
>>>>                     server.setDisplayConfiguration(verbose);
>>>>                     server.setAdapter(new StaticResourcesAdapter("."));
>>>>                                        server.initEndpoint();
>>>>                     server.startEndpoint();
>>>>
>>>> A file named index.html exists in the directory.
>>>>
>>>> ab -c 100 -n 100000 -p file.out -k http://ncampbell-pc:9999/ on RHEL
>>>>
>>>> file.out is 512 bytes.
>>>>
>>>> $ ab -c 100 -n 100000 -p file.out -k http://ncampbell-pc:9999/
>>>> This is ApacheBench, Version 2.0.41-dev <$Revision: 1.141 $> apache-2.0
>>>> Copyright (c) 1996 Adam Twiss, Zeus Technology Ltd,
>>>> http://www.zeustech.net/
>>>> Copyright (c) 1998-2002 The Apache Software Foundation,
>>>> http://www.apache.org/
>>>>
>>>> Benchmarking ncampbell-pc (be patient)
>>>> Completed 10000 requests
>>>> apr_recv: Connection reset by peer (104)
>>>> Total of 12438 requests completed
>>>>
>>>> I'm pretty sure I'm doing something naive, but I need your guidance.
>>>>
>>>> -Noah
>>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: [hidden email]
>>> For additional commands, e-mail: [hidden email]
>>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: [hidden email]
>> For additional commands, e-mail: [hidden email]
>>
>>
>>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

Re: Connection reset by peer (104)

Noah Campbell
 ab -c 300 -v4 -n 1000 http://ncampbell-pc:9999/goat/blah
This is ApacheBench, Version 2.0.41-dev <$Revision: 1.141 $> apache-2.0
Copyright (c) 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/
Copyright (c) 1998-2002 The Apache Software Foundation, http://www.apache.org/

Benchmarking ncampbell-pc (be patient)
INFO: POST header ==
---
GET /goat/blah HTTP/1.0
User-Agent: ApacheBench/2.0.41-dev
Host: ncampbell-pc:9999
Accept: */*


---
apr_recv: Connection refused (111)

Is it due to XP limiting incoming requests (i'm not running a server, just the workstation version)
Jeanfrancois Arcand-2 wrote
Hi,

Noah Campbell wrote:
> Interestingly, the same code scales to ab -c 2000 .... on linux (given the
> appropriate ulimit support).

Thanks for the report. I'm really not an expert with XP, but could it be
the ulimit is too low as well? Can you run ab using 'ab -v4' it will
logs the exceptions and send them here?

Are you seeing any exception on the server side?

Thanks

-- Jeanfrancois


>
> -Noah
>
>
> Jeanfrancois Arcand-2 wrote:
>> Hi,
>>
>> Oleksiy Stashok wrote:
>>> Hello Noah,
>>>
>>> looks like some issue in Grizzly, but there is not enough information.
>>> Is it all info/exceptions you have?
>> I don't think this is an issue :-) ab is a flooding tool...stressing
>> Grizzly without any ramp up time will always produce bad results. A
>> better way to test Grizzly is explained here:
>>
>> http://weblogs.java.net/blog/sdo/archive/2007/03/ab_considered_h.html
>>
>> Faban will gives much better results:
>>
>> http://weblogs.java.net/blog/sdo/archive/2007/04/simple_benchmar.html
>> http://weblogs.java.net/blog/sdo/archive/2007/05/how_to_test_con.html
>>
>> Also you might want to increase the number of threads (default is 20).
>> SelectorThread.setMaxThreads(..) :-)
>>
>> Thanks!
>>
>> -- Jeanfrancois
>>
>>
>>> If yes, can you pls. pack and send (or publish) the testing apps., so I
>>> will be able to reproduce this?
>>>
>>> Thanks.
>>>
>>> WBR,
>>> Alexey.
>>>
>>> Noah Campbell wrote:
>>>> Running grizzly 1.6.1 http on windows XP
>>>>
>>>> SelectorThread server = new SelectorThread();
>>>>                     server.setPort(9999);
>>>>                     server.setDisplayConfiguration(verbose);
>>>>                     server.setAdapter(new StaticResourcesAdapter("."));
>>>>                                        server.initEndpoint();
>>>>                     server.startEndpoint();
>>>>
>>>> A file named index.html exists in the directory.
>>>>
>>>> ab -c 100 -n 100000 -p file.out -k http://ncampbell-pc:9999/ on RHEL
>>>>
>>>> file.out is 512 bytes.
>>>>
>>>> $ ab -c 100 -n 100000 -p file.out -k http://ncampbell-pc:9999/
>>>> This is ApacheBench, Version 2.0.41-dev <$Revision: 1.141 $> apache-2.0
>>>> Copyright (c) 1996 Adam Twiss, Zeus Technology Ltd,
>>>> http://www.zeustech.net/
>>>> Copyright (c) 1998-2002 The Apache Software Foundation,
>>>> http://www.apache.org/
>>>>
>>>> Benchmarking ncampbell-pc (be patient)
>>>> Completed 10000 requests
>>>> apr_recv: Connection reset by peer (104)
>>>> Total of 12438 requests completed
>>>>
>>>> I'm pretty sure I'm doing something naive, but I need your guidance.
>>>>
>>>> -Noah
>>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: users-unsubscribe@grizzly.dev.java.net
>>> For additional commands, e-mail: users-help@grizzly.dev.java.net
>>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: users-unsubscribe@grizzly.dev.java.net
>> For additional commands, e-mail: users-help@grizzly.dev.java.net
>>
>>
>>
>

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@grizzly.dev.java.net
For additional commands, e-mail: users-help@grizzly.dev.java.net