You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@jmeter.apache.org by Andy Dawson <an...@iti.upv.es> on 2006/07/20 17:53:28 UTC

Monitor Results - Best practice

Hi All,

I have a test plan containing 2 thread groups
    One for the test (Thread count >1)
    One for monitoring the results (Thread count ==1)

The test plan is for a fixed duration with a permanent thread loop.

For the proof of concept type testing I am doing at the moment this is
fine - the load on the machine running Jmeter is very low, and the
resultant graph gives a good summary of the server state during the
test. For a coming project it will be necessary to use distributed
testing - At the moment I have defined the results monitor to save the
results and a simple data writer to save the raw data(Save as XML, Save
ResponseData, Save TimeStamp,Save ActiveThreadCount) should any further
analysis be required.

I would like to start as I mean to go on: what is the best way to make
use of a monitor? Is it considered bad practice to use the same machine
to load the server and monitor it´s response?

Thanks & Regards,

Andy


---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org

Re: Monitor Results - Best practice

Posted by Andy Dawson <an...@iti.upv.es>.
Hi Peter,

I replied to the other thread, but just for completeness:

Thanks for your reply.

I´m effectively using it for endurance testing of a troublesome app, 
which possibly crosses the boundry into smoke testing given the low 
duration/number of users sometimes needed to kill the server. I will 
look into using a profiler if it´s deemed more appropriate.

Thanks & Regards,

Andy

Peter Lin wrote:
> the monitor is used 2 ways.
>
> 1. to smoke test tomcat.  several tomcat developers use it for smoke 
> testing
> tomcat
> 2. to monitor several tomcat 5 servers
>
> peter
>
>
> On 7/20/06, Andy Dawson <an...@iti.upv.es> wrote:
>>
>> Hi All,
>>
>> I have a test plan containing 2 thread groups
>>     One for the test (Thread count >1)
>>     One for monitoring the results (Thread count ==1)
>>
>> The test plan is for a fixed duration with a permanent thread loop.
>>
>> For the proof of concept type testing I am doing at the moment this is
>> fine - the load on the machine running Jmeter is very low, and the
>> resultant graph gives a good summary of the server state during the
>> test. For a coming project it will be necessary to use distributed
>> testing - At the moment I have defined the results monitor to save the
>> results and a simple data writer to save the raw data(Save as XML, Save
>> ResponseData, Save TimeStamp,Save ActiveThreadCount) should any further
>> analysis be required.
>>
>> I would like to start as I mean to go on: what is the best way to make
>> use of a monitor? Is it considered bad practice to use the same machine
>> to load the server and monitor it´s response?
>>
>> Thanks & Regards,
>>
>> Andy
>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
>> For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
>>
>>
>


---------------------------------------------------------------------
To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: jmeter-user-help@jakarta.apache.org

Re: Monitor Results - Best practice

Posted by Peter Lin <wo...@gmail.com>.
the monitor is used 2 ways.

1. to smoke test tomcat.  several tomcat developers use it for smoke testing
tomcat
2. to monitor several tomcat 5 servers

peter


On 7/20/06, Andy Dawson <an...@iti.upv.es> wrote:
>
> Hi All,
>
> I have a test plan containing 2 thread groups
>     One for the test (Thread count >1)
>     One for monitoring the results (Thread count ==1)
>
> The test plan is for a fixed duration with a permanent thread loop.
>
> For the proof of concept type testing I am doing at the moment this is
> fine - the load on the machine running Jmeter is very low, and the
> resultant graph gives a good summary of the server state during the
> test. For a coming project it will be necessary to use distributed
> testing - At the moment I have defined the results monitor to save the
> results and a simple data writer to save the raw data(Save as XML, Save
> ResponseData, Save TimeStamp,Save ActiveThreadCount) should any further
> analysis be required.
>
> I would like to start as I mean to go on: what is the best way to make
> use of a monitor? Is it considered bad practice to use the same machine
> to load the server and monitor it´s response?
>
> Thanks & Regards,
>
> Andy
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: jmeter-user-help@jakarta.apache.org
>
>