Mailing-List: contact jmeter-user-help@jakarta.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "JMeter Users List" <jmeter-user@jakarta.apache.org>
Received-SPF: pass (nike.apache.org: domain of asp.adieu@gmail.com designates
 209.85.218.228 as permitted sender)
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=gamma;
        h=mime-version:in-reply-to:references:date:message-id:subject:from:to
         :content-type:content-transfer-encoding;
        b=qpacc8FDzIx/AJdATSCu4dyloOYtaD7JlJ9OyY3jwjz6VDOfJbiaKFEEsYc4eRZnVn
         5Jq0vrtO3Hvea7fFXDRExHhp6QrNWQwccHdIzJt7wWoE0og5s3uvLjGsIJa6TMv/xvkY
         /6ruZ1cKCTHCLxkDPdvAQr/cM98XVRLDAUXaM=
MIME-Version: 1.0
In-Reply-To: <4483c26c0907161041r49a59d9bue1bfecdf5119eab6@mail.gmail.com>
References: <24493378.post@talk.nabble.com>
	 <9980821.21247645983774.JavaMail.nobrien@olympia>
	 <24496582.post@talk.nabble.com>
	 <17160317.191247664526801.JavaMail.nobrien@olympia>
	 <24510645.post@talk.nabble.com>
	 <29a3f080907160623y42badaf6m5b2c01150e2d086@mail.gmail.com>
	 <f6f4ead70907160729g684afbb1ie6f58553e7779931@mail.gmail.com>
	 <4483c26c0907160930g33181537ja4857a7d27534470@mail.gmail.com>
	 <29a3f080907161019w7cfe0430r1533d85791a38c2e@mail.gmail.com>
	 <4483c26c0907161041r49a59d9bue1bfecdf5119eab6@mail.gmail.com>
Date: Mon, 20 Jul 2009 16:02:06 +0300
Message-ID: <f6f4ead70907200602u1227451dm67767be949ba9e62@mail.gmail.com>
Subject: Re: Do Jmeter Support Java Script calls..?
From: Adrian Speteanu <asp.adieu@gmail.com>
To: JMeter Users List <jmeter-user@jakarta.apache.org>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

Guess I am going to have taking a serious look at BSH / Java in JMeter.

In Raj's case, if Jmeter scripts are difficult to write, and it is
hard to test for scalability, I would try running Selenium scripts on
more machines (if possible to organize such a test where he works),
during the evening, at workstations left open by colleagues.

When all tests are running simultaneously, from another test machine I
would verify from a user's point of view how the server is responding
to see if the tabs are the problem indeed.

With Firebug, for example, you or the developer can see during
realtime use of the application what sequences run slower than others
/ what needs optimizing so you could identify some bottlenecks (the
NET panel shows the load times on this machine).
BTW: Any objections to this rudimentary method? (*Note: I am only
looking for bottlenecks with this test and is my last resort).

On Thu, Jul 16, 2009 at 8:41 PM, Deepak Shetty<shettyd@gmail.com> wrote:
>>However what I think the op wants to do is performance
>>test loading each tab, which is implemented in client side code
>>(javascript).
> Yep , I was replying to adrian.
> I agree with the rest of your points.
>
> On Thu, Jul 16, 2009 at 10:19 AM, Tony Lotts <tljmeter@gmail.com> wrote:
>
>> Deepak is right. =A0However what I think the op wants to do is performan=
ce
>> test loading each tab, which is implemented in client side code
>> (javascript).
>> The performance of executing client side code is based upon the user's o=
wn
>> machine, and browser; which is irrelevant to web application scalability
>> and
>> thus not served by JMeter.
>>
>> In this case you may want to deliver a comparison of tab navigation acro=
ss
>> each of your supported browsers, reporting the baseline hardware that th=
e
>> test was conducted with.
>> However unless the tab navigation is a suspect of a poorly performing
>> implementation, it's not worth investigation.
>>
>> On Thu, Jul 16, 2009 at 12:30 PM, Deepak Shetty <shettyd@gmail.com> wrot=
e:
>>
>> > >how practical with you consider using the __javaScript() function in
>> > >this type of situations? (or any other way to process client side
>> > >actions when they are strictly needed)
>> > I ran into this problem recently and decided to write java code that
>> > simulated what the function does (generating complex XML from variable=
s ,
>> > dynamic field names etc..). If I'm interested in testing functionality
>> that
>> > is browser/javascript dependent I normally use selenium (or QTP). Its =
far
>> > more difficult to get any accuracy of browser render or script executi=
on
>> > times , considering the various hardware configurations that can exist=
.
>> >
>> > > I got the help of two developers and we found one variable expected
>> > >from the applet, no appropiate value was found in all the traffic
>> > >recorded, so they assumed that it is sent directly from the applet to
>> > >the server
>> > If it's using HTTP it must work (the applet however could do other
>> stuff),
>> > You'd probably need to use a network sniffer tool like Wireshark and
>> > verify.
>> > In any case what an applet can do(in terms of client-server
>> communication)
>> > ,
>> > you can do in Java/BSH.
>> > regards
>> > deepak
>> >
>> > On Thu, Jul 16, 2009 at 7:29 AM, Adrian Speteanu <asp.adieu@gmail.com
>> > >wrote:
>> >
>> > > well since this question was re-re-asked (again :) ), here goes
>> nothing:
>> > >
>> > > =A0 how practical with you consider using the __javaScript() functio=
n in
>> > > this type of situations? (or any other way to process client side
>> > > actions when they are strictly needed)
>> > > ( I am thinking of situations were the application might be so
>> > > obfuscated that without running a certain js you cannot generate a
>> > > certain variable value to pass back to the server, so whatever you d=
o,
>> > > your requests will never be well formatted or complete. would it be
>> > > recommended/possible to adapt that js to what JMeter recognises / ca=
n
>> > > do? )
>> > >
>> > > I encountered this problem with applets: all apparent requests
>> > > encountered in proxy server were replicated and I even managed to ge=
t
>> > > the id's right =A0for all applet elements and for repository resourc=
es.
>> > > however, the application returned null pointer exception for one of
>> > > the most simple request of all.
>> > > I got the help of two developers and we found one variable expected
>> > > from the applet, no appropiate value was found in all the traffic
>> > > recorded, so they assumed that it is sent directly from the applet t=
o
>> > > the server (I have my doubts on that but could not prove them wrong)=
.
>> > > so, the test simply didn't act like the application no matter what.
>> > >
>> > >
>> > > On Thu, Jul 16, 2009 at 4:23 PM, Tony Lotts <tljmeter@gmail.com>
>> wrote:
>> > > >
>> > > > What is it that you want to accomplish?
>> > > > If this is for a scalability test, =A0the performance of client si=
de
>> code
>> > > > execution is irrelevant.
>> > > >
>> > > > On Thu, Jul 16, 2009 at 2:10 AM, MITRAJ <rajesh.mittapalli@gs.com>
>> > > wrote:
>> > > >
>> > > > >
>> > > > > Tabs are loaded on demand and contents in the tab are pre-loaded=
..
>> > > > >
>> > > > > Thanks,
>> > > > > Raj
>> > > > >
>> > > > > Noel O'Brien wrote:
>> > > > > >
>> > > > > > How is the content of the tab gotten from the server? Is it
>> > > pre-loaded or
>> > > > > > loaded on demand.? Either way, it's probably obtained using a
>> http
>> > > call.
>> > > > > >
>> > > > > > Try using a tool like wireshark to to analyze the call flow or
>> you
>> > > could
>> > > > > > use JMeters HTTP Proxy to recore the call flows directly.
>> > > > > >
>> > > > > > Regards,
>> > > > > > Noel
>> > > > > >
>> > > > > > ----- "MITRAJ" <rajesh.mittapalli@gs.com> wrote:
>> > > > > >>
>> > > > > >> Thanks Noel
>> > > > > >>
>> > > > > >> In my application have 3 tabs.
>> > > > > >> If i click on the tab, A frame is loaded. Those tabs are base=
d
>> on
>> > > Java
>> > > > > >> script. (HTTP request url is not there for these tabs). Is th=
ere
>> > any
>> > > way
>> > > > > >> to
>> > > > > >> test tabs..?
>> > > > > >>
>> > > > > >> Thanks
>> > > > > >> Raj
>> > > > > >>
>> > > > > >>
>> > > > > >> Noel O'Brien wrote:
>> > > > > >> >
>> > > > > >> > Hi,
>> > > > > >> >
>> > > > > >> > See the "JMeter is not a browser" section in
>> > > > > >> > http://jakarta.apache.org/jmeter/index.html
>> > > > > >> >
>> > > > > >> > Regards,
>> > > > > >> > Noel
>> > > > > >> >
>> > > > > >> > ----- "MITRAJ" <rajesh.mittapalli@gs.com> wrote:
>> > > > > >> >>
>> > > > > >> >>
>> > > > > >> >> --
>> > > > > >> >> View this message in context:
>> > > > > >> >>
>> > > > > >>
>> > > > >
>> > >
>> >
>> http://www.nabble.com/Do-Jmeter-Support-Java-Script-calls..--tp24493378p=
24493378.html
>> > > > > >> >> Sent from the JMeter - User mailing list archive at
>> Nabble.com.
>> > > > > >> >>
>> > > > > >> >>
>> > > > > >> >>
>> > > --------------------------------------------------------------------=
-
>> > > > > >> >> To unsubscribe, e-mail:
>> > > jmeter-user-unsubscribe@jakarta.apache.org
>> > > > > >> >> For additional commands, e-mail:
>> > > jmeter-user-help@jakarta.apache.org
>> > > > > >> >>
>> > > > > >> >>
>> > > > > >> >
>> > > > > >> > --
>> > > > > >> > Regards,
>> > > > > >> > Noel
>> > > > > >> >
>> > > > > >> >
>> > > > > >>
>> > > > > >> --
>> > > > > >> View this message in context:
>> > > > > >>
>> > > > >
>> > >
>> >
>> http://www.nabble.com/Do-Jmeter-Support-Java-Script-calls..--tp24493378p=
24496582.html
>> > > > > >> Sent from the JMeter - User mailing list archive at Nabble.co=
m.
>> > > > > >>
>> > > > > >>
>> > > > > >>
>> > > --------------------------------------------------------------------=
-
>> > > > > >> To unsubscribe, e-mail:
>> > jmeter-user-unsubscribe@jakarta.apache.org
>> > > > > >> For additional commands, e-mail:
>> > > jmeter-user-help@jakarta.apache.org
>> > > > > >>
>> > > > > >>
>> > > > > >
>> > > > > > --
>> > > > > > Regards,
>> > > > > > Noel
>> > > > > >
>> > > > > >
>> > > > >
>> > > > > --
>> > > > > View this message in context:
>> > > > >
>> > >
>> >
>> http://www.nabble.com/Do-Jmeter-Support-Java-Script-calls..--tp24493378p=
24510645.html
>> > > > > Sent from the JMeter - User mailing list archive at Nabble.com.
>> > > > >
>> > > > >
>> > > > >
>> ---------------------------------------------------------------------
>> > > > > To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.o=
rg
>> > > > > 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
>> > >
>> > >
>> >
>>
>

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