sebb a écrit :

On 05/05/2009, Jean-Louis.Pasturel <[email protected]> wrote:
sebb a écrit :



On 04/05/2009, Jean-Louis.Pasturel
<[email protected]> wrote:
sebb a écrit :




On 01/05/2009, Jean-Louis.Pasturel


<[email protected]> wrote:


sebb a écrit :






On 01/05/2009, Jean-Louis.Pasturel




<[email protected]> wrote:




Hi All,
I am a begginner with JMeter and this is my first post to this


mailing


list, and  sorry for perhaps faq questions. If it is, let me
know.
Is it possible to do static initialisation in a Java Request
Sample
that


uses the parameters set in  the panel of the the Java Request
Sample
?


Or is




there a better way with other feature of JMeter ( Pre-processor
?)?




I'm not sure what you are trying to do.

The Java Request Sampler is a very basic sampler that allows one
to
generate a simple sample response; not sure what static
initialisation
has to do with that.

Please explain in a bit more detail what you are trying to
achieve,
and then maybe someone can provide suggestions as to how it might
be
done.







Regards
JL Pasturel







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








---------------------------------------------------------------------
To unsubscribe, e-mail:




[email protected]




For additional commands, e-mail:




[email protected]




---------------------------------------------------------------------------------------
Orange vous informe que cet  e-mail a ete controle par
l'anti-virus
mail.


Aucun virus connu a ce jour par nos services n'a ete detecte.







Sorry,  i agree that I was not very  clear.
Exactly i don't know how JMeter and Java  Request Sample
instanciates
threads.




The StandardJMeterEngine creates threads; each Sampler instance runs
in a single thread.





So i wondering if to use static variables ( not final)  has a sense
in a Java  Request Sample.




No.





My goal, is to share objects ( with sometimes monitored accesses )


between


threads ( for example :  a datapool, read by all the threads,
counters,
getting a local port not used by others threads ...).




But what are you trying to achieve?





Hope  i am more clear ...




Sorry, not really.





Regards
JL Pasturel






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






---------------------------------------------------------------------
To unsubscribe, e-mail:


[email protected]


For additional commands, e-mail:


[email protected]


---------------------------------------------------------------------------------------
Orange vous informe que cet  e-mail a ete controle par l'anti-virus
mail.
Aucun virus connu a ce jour par nos services n'a ete detecte.





Sebb, thanks for the answer about static parameters.
What i did, is a java sampler for a non-supported protocol by JMeter (
Radius ) with an open source library : Tiny Radius.
This protocol is use in Telco to permit authentification of  internet
home
boxes ( Internet, TV, VOIP ...).
I have to handle a datapool of 10 000 000 lines and to pick rows
randomly.
So each thread has it own file descriptor ( RandomAccessFile instance).
My
Why not use a file of 10 000 000 lines and read them using CSV DataSet?

Each thread will get the next entry. No need for multiple file
descriptors.

Java Sampler runs fine and i can use other features of JMeter (
Listener,
trees...).
The goal of my initial question was about the best practices to handle,
for
example, my shared datapool with a random access with JMeter.


See above.



Other feature I look for, is IP Spoofing. I have begun to write a TCP
proxy
"out of the box" that handles IP Spoofing.
For  now, I have to solve, in the best way, the problem of the affinity
of
the session in a transaction or in a conversation ( tracking for example
cookies). Certainly by using regexp. Have  you  advices about it ?


Regards
JL Pasturel



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




---------------------------------------------------------------------
To unsubscribe, e-mail:
[email protected]
For additional commands, e-mail:
[email protected]
---------------------------------------------------------------------------------------
Orange vous informe que cet  e-mail a ete controle par l'anti-virus mail.
Aucun virus connu a ce jour par nos services n'a ete detecte.



<sebb> Why not use a file of 10 000 000 lines and read them using CSV
DataSet?

Each thread will get the next entry. No need for multiple file
descriptors</sebb>

Yes i have look at this feature, but my needs is to avoid the cache effect
at Radius server level and Database level, so i have to randomize as more as
possible the datas that i inject to my application. Serial reads can allow
caching between each stress.
Thanks for your advices.

The data in the CSV data file can be randomised before starting the test.

You can also guarantee uniqueness of the random data more easily if
you pre-compute it.

For the present use, it doesn't matter that a same data was played by two threads in the same time.
If it was not the case, that you said above is a solution.


Regards
JL Pasturel






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



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

---------------------------------------------------------------------------------------
Orange vous informe que cet e-mail a ete controle par l'anti-virus mail. Aucun virus connu a ce jour par nos services n'a ete detecte.






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

Reply via email to