You are viewing a plain text version of this content. The canonical link for it is here.
Posted to users@httpd.apache.org by "Rose, John B" <jb...@utk.edu> on 2014/11/06 20:37:04 UTC
[users@httpd] http testing question
Any pro/con views, personal experience, insights on these tools, or any not listed here you think are good?
Ab - Apache
jmeter
Webpagetest
Httperf
Apache Flood
We have tried ab, webpagetest, and jmeter a bit but still learning.
Thanks
Re: [users@httpd] http testing question
Posted by "Patton, Billy N" <bi...@h3net.com>.
I’ve just finished creating a script that will test my app for content and functionality.
I use WWW::Mechanize and Test::WWW::Mechanize
Getting their nuances took me some trial and error, There are some thing it is missing, su as proof that a radio button was actually clicked
Or I havn’t figured out that nuance.
For content and functionality it does a good job.
Although I have yet to figure out why it won’t do a follow_link for one of 20, when it will follow it perfectly from the app or from the html, when I view source and click on it.
Loading would take some parallel threading to load up your server, but then you’d have to conceive some way to measure.
All this if you have time and not a budget for doing it.
On Nov 7, 2014, at 6:08 AM, Jeff Trawick <tr...@gmail.com>> wrote:
On Thu, Nov 6, 2014 at 2:37 PM, Rose, John B <jb...@utk.edu>> wrote:
Any pro/con views, personal experience, insights on these tools, or any not listed here you think are good?
Ab - Apache
jmeter
Webpagetest
Httperf
Apache Flood
We have tried ab, webpagetest, and jmeter a bit but still learning.
Thanks
What aspects do you plan to test?
Most, maybe all, can create arbitrary load on your server, at least when you use multiple clients.
Most, maybe all, can check the HTTP response code.
Some can perform detailed checks of response content.
Some can use the response from one request to drive subsequent requests, such as when simulating login or more complex user flows.
Maybe the question is what combination of load testing and functional testing do you need?
--
Born in Roswell... married an alien...
http://emptyhammock.com/
Re: [users@httpd] http testing question
Posted by Stefan Magnus Landrø <st...@gmail.com>.
I highly recommend Gatling for load testing. It features non blocking io and can easily drive 50.000 connections from one machine.
Nice DSL too.
Stefan
Sendt fra min iPhone
> Den 17. nov. 2014 kl. 22.29 skrev Rose, John B <jb...@utk.edu>:
>
> | Maybe the question is what combination of load testing and functional testing do you need?
>
> I would like to test the values/settings we have had on our Apache 2.2 webserver with as close as
> possible to the equivalent settings in our new Apache 2.4 server.
>
> We would like to experiment with each value and run a test for each value change we make and see the im
> pact on the response perfromance and load on the server resources, RAM, CPU, Network capacity, various
> states shown in server-status. Over the years things have been put in place just to get it running or q
> uickly fixed. We want to thoroughly vette our web server configuration settings with this opportunity.
>
> We would like to take the historical log of accesses to the current production 2.2 server and run
> those against our 2.4 server. Experiement with existing directives and their values. Try other
> directives that look potentially useful we havent tried before. See how they help or hinder
> performance.
>
> We want to take each of the new features of 2.4 and one by one try them and test each one with
> various values.
>
> We want to stress test the server to see the maximum load it can take from a RAM perspective,
> then a CPU perspective, and still have acceptable perfromance. Find the tipping point of
> simultaneous connectivity that begins to render the server unusable and the approx number of
> connections when that is likely occur.
>
> We want to know from the server perspective that we have everything as optimally tuned
> for performance and capacity as we can possibly get it. Up to the point where only the
> developers web site development techniques would be the only thing that could slow it down.
>
> And we are open to any suggestions from you experts.
>
> From: Jeff Trawick <tr...@gmail.com>
> Reply-To: "users@httpd.apache.org" <us...@httpd.apache.org>
> Date: Friday, November 7, 2014 7:08 AM
> To: "users@httpd.apache.org" <us...@httpd.apache.org>
> Subject: Re: [users@httpd] http testing question
>
>> On Thu, Nov 6, 2014 at 2:37 PM, Rose, John B <jb...@utk.edu> wrote:
>> Any pro/con views, personal experience, insights on these tools, or any not listed here you think are good?
>>
>> Ab - Apache
>> jmeter
>> Webpagetest
>> Httperf
>> Apache Flood
>>
>> We have tried ab, webpagetest, and jmeter a bit but still learning.
>>
>> Thanks
>
>
> What aspects do you plan to test?
>
> Most, maybe all, can create arbitrary load on your server, at least when you use multiple clients.
> Most, maybe all, can check the HTTP response code.
> Some can perform detailed checks of response content.
> Some can use the response from one request to drive subsequent requests, such as when simulating login or more complex user flows.
>
> Maybe the question is what combination of load testing and functional testing do you need?
>
> --
> Born in Roswell... married an alien...
> http://emptyhammock.com/
>
Re: [users@httpd] http testing question
Posted by "Rose, John B" <jb...@utk.edu>.
| Maybe the question is what combination of load testing and functional testing do you need?
I would like to test the values/settings we have had on our Apache 2.2 webserver with as close as
possible to the equivalent settings in our new Apache 2.4 server.
We would like to experiment with each value and run a test for each value change we make and see the im
pact on the response perfromance and load on the server resources, RAM, CPU, Network capacity, various
states shown in server-status. Over the years things have been put in place just to get it running or q
uickly fixed. We want to thoroughly vette our web server configuration settings with this opportunity.
We would like to take the historical log of accesses to the current production 2.2 server and run
those against our 2.4 server. Experiement with existing directives and their values. Try other
directives that look potentially useful we havent tried before. See how they help or hinder
performance.
We want to take each of the new features of 2.4 and one by one try them and test each one with
various values.
We want to stress test the server to see the maximum load it can take from a RAM perspective,
then a CPU perspective, and still have acceptable perfromance. Find the tipping point of
simultaneous connectivity that begins to render the server unusable and the approx number of
connections when that is likely occur.
We want to know from the server perspective that we have everything as optimally tuned
for performance and capacity as we can possibly get it. Up to the point where only the
developers web site development techniques would be the only thing that could slow it down.
And we are open to any suggestions from you experts.
From: Jeff Trawick <tr...@gmail.com>>
Reply-To: "users@httpd.apache.org<ma...@httpd.apache.org>" <us...@httpd.apache.org>>
Date: Friday, November 7, 2014 7:08 AM
To: "users@httpd.apache.org<ma...@httpd.apache.org>" <us...@httpd.apache.org>>
Subject: Re: [users@httpd] http testing question
On Thu, Nov 6, 2014 at 2:37 PM, Rose, John B <jb...@utk.edu>> wrote:
Any pro/con views, personal experience, insights on these tools, or any not listed here you think are good?
Ab - Apache
jmeter
Webpagetest
Httperf
Apache Flood
We have tried ab, webpagetest, and jmeter a bit but still learning.
Thanks
What aspects do you plan to test?
Most, maybe all, can create arbitrary load on your server, at least when you use multiple clients.
Most, maybe all, can check the HTTP response code.
Some can perform detailed checks of response content.
Some can use the response from one request to drive subsequent requests, such as when simulating login or more complex user flows.
Maybe the question is what combination of load testing and functional testing do you need?
--
Born in Roswell... married an alien...
http://emptyhammock.com/
Re: [users@httpd] http testing question
Posted by Jeff Trawick <tr...@gmail.com>.
On Thu, Nov 6, 2014 at 2:37 PM, Rose, John B <jb...@utk.edu> wrote:
> Any pro/con views, personal experience, insights on these tools, or any
> not listed here you think are good?
>
> Ab - Apache
> jmeter
> Webpagetest
> Httperf
> Apache Flood
>
> We have tried ab, webpagetest, and jmeter a bit but still learning.
>
> Thanks
>
What aspects do you plan to test?
Most, maybe all, can create arbitrary load on your server, at least when
you use multiple clients.
Most, maybe all, can check the HTTP response code.
Some can perform detailed checks of response content.
Some can use the response from one request to drive subsequent requests,
such as when simulating login or more complex user flows.
Maybe the question is what combination of load testing and functional
testing do you need?
--
Born in Roswell... married an alien...
http://emptyhammock.com/
Re: [users@httpd] http testing question
Posted by KARTHIK SHIVAKUMAR <ns...@gmail.com>.
Try Burp (not free .... paid version)
URL : http://portswigger.net/burp/
It is good for http/https/rest/soap ..... testing
with regards
karthik
On Fri, Nov 7, 2014 at 1:07 AM, Rose, John B <jb...@utk.edu> wrote:
> Any pro/con views, personal experience, insights on these tools, or any
> not listed here you think are good?
>
> Ab - Apache
> jmeter
> Webpagetest
> Httperf
> Apache Flood
>
> We have tried ab, webpagetest, and jmeter a bit but still learning.
>
> Thanks
>
--
*N.S.KARTHIKR.M.S.COLONYBEHIND BANK OF INDIAR.M.V 2ND STAGEBANGALORE560094*