[revServer] process timeout issue

[Pierre Sahores]

Dear Friends,

the command i purpose to use at shell prompt, today, in beetwin 15:00 and 15:05, Dallas Texas time :

$ ab -kc 10 -t 30 http://woooooooords.com/index.html
$ ab -kc 10 -t 30 http://woooooooords.com/index.irev

Both command are pointed on the same index.irev (trought revrite rules)

For the participants : thanks for reporting the results.

All the best,

Pierre

My own test :

$ ab -kc 10 -t 30 http://woooooooords.com/index.irev
This is ApacheBench, Version 2.3 <$Revision: 655654 $>
Copyright 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/
Licensed to The Apache Software Foundation, http://www.apache.org/

Benchmarking woooooooords.com (be patient)
Finished 1557 requests


Server Software:        Apache/2.0.63
Server Hostname:        woooooooords.com
Server Port:            80

Document Path:          /index.irev
Document Length:        417 bytes

Concurrency Level:      10
Time taken for tests:   30.000 seconds
Complete requests:      1557
Failed requests:        0
Write errors:           0
Non-2xx responses:      1557
Keep-Alive requests:    1547
Total transferred:      1215577 bytes
HTML transferred:       649269 bytes
Requests per second:    51.90 [#/sec] (mean)
Time per request:       192.680 [ms] (mean)
Time per request:       19.268 [ms] (mean, across all concurrent requests)
Transfer rate:          39.57 [Kbytes/sec] received

Connection Times (ms)
              min  mean[+/-sd] median   max
Connect:        0    2  21.6      0     209
Processing:   181  190  20.7    188     768
Waiting:      181  189  14.6    188     684
Total:        181  192  33.4    188     893

Percentage of the requests served within a certain time (ms)
  50%    188
  66%    190
  75%    192
  80%    194
  90%    196
  95%    198
  98%    201
  99%    375
 100%    893 (longest request)

$ ab -kc 10 -t 30 http://woooooooords.com/index.html
This is ApacheBench, Version 2.3 <$Revision: 655654 $>
Copyright 1996 Adam Twiss, Zeus Technology Ltd, http://www.zeustech.net/
Licensed to The Apache Software Foundation, http://www.apache.org/

Benchmarking woooooooords.com (be patient)
Finished 1558 requests


Server Software:        Apache/2.0.63
Server Hostname:        woooooooords.com
Server Port:            80

Document Path:          /index.html
Document Length:        417 bytes

Concurrency Level:      10
Time taken for tests:   30.005 seconds
Complete requests:      1558
Failed requests:        0
Write errors:           0
Non-2xx responses:      1558
Keep-Alive requests:    1548
Total transferred:      1216358 bytes
HTML transferred:       649686 bytes
Requests per second:    51.92 [#/sec] (mean)
Time per request:       192.586 [ms] (mean)
Time per request:       19.259 [ms] (mean, across all concurrent requests)
Transfer rate:          39.59 [Kbytes/sec] received

Connection Times (ms)
              min  mean[+/-sd] median   max
Connect:        0    3  35.1      0    1101
Processing:   182  189   5.7    188     339
Waiting:      182  189   5.7    188     339
Total:        182  192  36.3    188    1285

Percentage of the requests served within a certain time (ms)
  50%    188
  66%    190
  75%    193
  80%    194
  90%    195
  95%    197
  98%    200
  99%    376
 100%   1285 (longest request)




Le 2 août 2010 à 19:13, Pierre Sahores a écrit :

> Mark,
> 
> Good suggestion ! 
> 
> Andre,
> 
> I'm OK for a test purpose and incremental distributed attack against woooooooords.com if we can share a knowed test code (yours ?) in beetwin some of us.
> 
> 100, 1000, 10 000, ... parallelised requests to see how it goes with a code setup to run automatically in a temporized mode. 
> 
> This would probably let us know more about the abilities of the revServer and (loki.)on-rev.om to handle trafic.
> 
> Any thought welcome,
> 
> Kind Regards, 
> 
> Pierre
> 
> 
> Le 2 août 2010 à 18:45, Mark Wieder a écrit :
> 
>> The only way to get meaningful results
>> from load-testing a server is to use multiple client machines. If you
>> could post your test code somewhere maybe more of us could run
>> simultaneous tests against the server for a distributed attack.
> 
> --
> Pierre Sahores
> mobile : (33) 6 03 95 77 70
> 
> www.woooooooords.com
> www.sahores-conseil.com
> 
> 
> 
> 
> 

--
Pierre Sahores
mobile : (33) 6 03 95 77 70

www.woooooooords.com
www.sahores-conseil.com