Re: Gatling Vs Jmeter

[Adrian Speteanu]

Hi,

While I'm not necessarily happy with the methodology used for comparison,
the point remains that there aren't significant differences when it comes
to the technical capabilities. So, in the end, it should come down to the
implementation team's preferences. Are they going to feel more comfortable
scripting in scala when the scenarios get more complex? Or do they prefer
using the GUI of JMeter?

Implementing scenarios would be simpler in JMeter in my view, but it would
be easier to maintain scripts over time if they were coded (in any
language).

-A



On Tue, Feb 4, 2020 at 7:14 AM Soumya Nayak  wrote:

> Hi Team,
>
> I came across a link about Gatling Vs JMeter .
>
> https://dzone.com/articles/gatling-vs-jmeter
>
> From the above link I could see some of the advantages of Gatling over
> JMeter with respect to memory and CPU usage.
>
> So anybody having more details can help me out this to create a line of
> difference to plan for the specific tool usage better.
>
> Regards,
> Soumya
>
>
>
> **
> This message may contain confidential or proprietary information intended
> only for the use of the
> addressee(s) named above or may contain information that is legally
> privileged. If you are
> not the intended addressee, or the person responsible for delivering it to
> the intended addressee,
> you are hereby notified that reading, disseminating, distributing or
> copying this message is strictly
> prohibited. If you have received this message by mistake, please
> immediately notify us by
> replying to the message and delete the original message and any copies
> immediately thereafter.
>
> If you received this email as a commercial message and would like to opt
> out of future commercial
> messages, please let us know and we will remove you from our distribution
> list.
>
> Thank you.~
>
> **
> FAFLD
>

Re: jmeter behind proxy server

[Adrian Speteanu]

Hi,

Haven't considered opening that path. It was indeed a function and returned
among other things the proxy that jmeter could use.

Thank you!

-A

On Tue, Dec 3, 2019 at 7:58 PM Felix Schumacher <
felix.schumac...@internetallee.de> wrote:

>
> Am 03.12.19 um 16:55 schrieb Adrian Speteanu:
> > Hi,
> >
> > Not sure how to configure JMeter to use a proxy server when it has a
> > slightly different format (/).
> > I've read
> https://jmeter.apache.org/usermanual/get-started.html#proxy_server
> > And by the looks of the UI, it seems like the user can only set a server,
> > port, username and password in order to configure that proxy (I don't
> even
> > need a username).
> >
> > The browsers, firefox / chrome / edge, are set to autoconfigure the proxy
> > from http:/// and it just works.
> >
> > Any suggestions on how to deal with these types of situations?
>
> What you are describing it the auto configuration protocol for browsers.
> You point your browser configuration to a special file that contains
> rules on which way a browser should reach a resource (by proxy or direct).
>
> JMeter is not a browser and currently has - to my knowledge - no builtin
> support for such a rules file (which is basically a JavaScript function
> that gets called with the url and the host to be visited and returns the
> means of reaching the destination).
>
> What you can do, is to get the config file by hand and look at the
> proxies that are mentioned in there. Than you configure JMeter using
> those values.
>
> Say, your config url returns a file with a following content:
>
> function FindProxyForURL(url, host) {
> if (isInNet(host, "10.0.0.0", "255.0.0.0")) {
>return "DIRECT";
> }
> return "PROXY proxy.example.invalid:8080";
> }
>
> then you would use proxy.example.invalid as the proxy host and 8080 as
> the proxy port, if you want to reach hosts outside of 10.0.0.0/8.
>
> Hope this helps
>
>  Felix
>
> >
> > For the record, I've already tried pasting it all in the server name
> field
> > and it didn't work.
> >
> > Thanks in advance,
> > -Adrian
> >
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

jmeter behind proxy server

[Adrian Speteanu]

Hi,

Not sure how to configure JMeter to use a proxy server when it has a
slightly different format (/).
I've read https://jmeter.apache.org/usermanual/get-started.html#proxy_server
And by the looks of the UI, it seems like the user can only set a server,
port, username and password in order to configure that proxy (I don't even
need a username).

The browsers, firefox / chrome / edge, are set to autoconfigure the proxy
from http:/// and it just works.

Any suggestions on how to deal with these types of situations?

For the record, I've already tried pasting it all in the server name field
and it didn't work.

Thanks in advance,
-Adrian

Re: AEM load testing - Apache JMeter

[Adrian Speteanu]

On the JMeter's machine you shouldn't need anything other than what your
browser needs in order to connect to that remote instance (proxy
configuration, if required). So you need to do a functional validation with
very light load at first and make sure that your test plan behaves similar
to what you see in the browser. Remote instances often come with different
configurations than what you'd get locally (tightened security
configurations, actual passwords set for default users, integrated
authentication systems etc) and the network where you do the load/stress
test needs to support the load you intend to generate.

Cheers,
Adrian

On Thu, Apr 26, 2018 at 11:59 PM, Pankaj Rohira 
wrote:

> I need help in doing load testing for AEM (Adobe Experience Manager) using
> JMeter. I wrote my scripts and it's working fine on my local. Using JMeter,
> I can login, perform Omni Search and able to test other custom
> functionalities also on my local AEM instance. When I try to do the same
> thing on remote instances, JMeter is error'ing out.
>
>
>
> First step is to login to remote instance - I am doing a POST call to
> /libs/granite/core/content/login.html/j_security_check and passing all the
> parameters but I get a 403 Forbidden error. As part of the POST call, I am
> passing necessary Request Headers as well such as: Host, User-Agent,
> Referer. (I actually traced the POST call using Firefox when I normally
> login to AEM using browser and grabbed all the Request Headers from there).
>
>
>
> Just wondering if I need to enable RMI or something on remote instance to
> connect or am I missing anything?
>
>
>
> Thanks.
>

Re: Opinion on this potential new look for JMeter UI

[Adrian Speteanu]

Hi,

Looks nice and it will be useful to some people. I personally use a
dark(-ish) colour theme in most of my editors and the console setup, so
this change is very much welcomed, it is difficult to adjust to apps with a
white interface. So I would really appreciate this improvement.

Cheers,
-A

On Tue, Oct 31, 2017 at 4:02 PM, Philippe Mouawad 
wrote:

> Hello,
> Your opinion is welcome on this new Look proposal for JMeter UI:
>
> - https://twitter.com/ApacheJMeter/status/925357635066572800
>
> Thanks for expressing yourselves at this link or directly here.
>
> Regards
> Philippe M. on behalf of Apache JMeter Team
> https://twitter.com/ApacheJMeter/status/925357635066572800
>

Re: JMeter is not calculating the time when the tread is getting timeout

[Adrian Speteanu]

If you can obtain the entire time interval by adding the intended request
and pre/post actions into a single transaction controller, than any
duration assertion you add as child of the controller should do what you
say. But, again, I'm still fuzzy on what exactly is that you're trying to
do.

Cheers,
A

On Tue, Aug 16, 2016 at 4:28 PM, Sankar Das <007bhabanisan...@gmail.com>
wrote:

> Add to this.
>
> If the "Response Assertion" can verify the response text irrespective of
> the sampler connection time than why not the "Duration Assertion" wont
> calculate the timing if the thread is timing out.
>
> PS: Do know that "Duration Assertion" will calculate the time of the
> sampler first than compare with the duration given. If the thread is
> waiting then it will also wait.
>
> Thanks!
>
> On Tue, Aug 16, 2016 at 4:07 PM, Sankar Das <007bhabanisan...@gmail.com>
> wrote:
>
> > Hi,
> >
> > Thanks a lot for the reply.
> >
> > Trying check the accessibility of some APIs on some intervals by help of
> > Jenkins, Ant and JMeter and do some actions if they are not
> accessible.Not
> > the server response time. I guess from JMeter we can do it.Yes, we can
> also
> > do functional testing using JMeter although it acts like a browser.
> >
> >
> > Thanks!
> >
> >
> > On Tue, Aug 16, 2016 at 3:40 PM, Adrian Speteanu 
> > wrote:
> >
> >> I don't understand the use-case or why anyone would want to include any
> >> timeouts in the response times if they're added by anything else other
> >> than
> >> the system under test. The purpose of the tool is to measure server side
> >> response times (from just before starting a new connection to the point
> >> just after receiving the last bit of the response), not the measurement
> >> tool's own processing/time-wait periods.
> >>
> >> That been said, I've seen use-cases where one might need to calculate
> the
> >> total duration of an action that requires multiple/nested requests and
> in
> >> order to get the total value for this group, one can use the transaction
> >> controller and adding a listener that has this controller in its scope.
> I
> >> think it might have some options that might prove useful:
> >> http://jmeter.apache.org/usermanual/component_reference.
> >> html#Transaction_Controller
> >>
> >> I must insist though that a bigger problem would be the requirements and
> >> making sure they are justified.
> >>
> >> Cheers,
> >> A
> >>
> >> On Tue, Aug 16, 2016 at 3:11 PM, Sankar Das <007bhabanisan...@gmail.com
> >
> >> wrote:
> >>
> >> > Hi,
> >> >
> >> > *Scenario*: My requirement is to do some actions when the assertion
> >> fails.
> >> >
> >> > *Issue*:When the thread is timing out (have provided the timeout
> details
> >> > in" HTTP Request Defaults", JMeter "Duration Assertion" is not
> >> calculating
> >> > the timing of the timed out sampler.
> >> >
> >> > Hence not able to do the action.
> >> >
> >> > *Expected*.
> >> > Response duration should be able to calculate the timeout time of the
> >> > sampler ,even if the sampler is timed-out.
> >> >
> >>
> >
> >
>

Re: JMeter is not calculating the time when the tread is getting timeout

[Adrian Speteanu]

I don't understand the use-case or why anyone would want to include any
timeouts in the response times if they're added by anything else other than
the system under test. The purpose of the tool is to measure server side
response times (from just before starting a new connection to the point
just after receiving the last bit of the response), not the measurement
tool's own processing/time-wait periods.

That been said, I've seen use-cases where one might need to calculate the
total duration of an action that requires multiple/nested requests and in
order to get the total value for this group, one can use the transaction
controller and adding a listener that has this controller in its scope. I
think it might have some options that might prove useful:
http://jmeter.apache.org/usermanual/component_reference.html#Transaction_Controller

I must insist though that a bigger problem would be the requirements and
making sure they are justified.

Cheers,
A

On Tue, Aug 16, 2016 at 3:11 PM, Sankar Das <007bhabanisan...@gmail.com>
wrote:

> Hi,
>
> *Scenario*: My requirement is to do some actions when the assertion fails.
>
> *Issue*:When the thread is timing out (have provided the timeout details
> in" HTTP Request Defaults", JMeter "Duration Assertion" is not calculating
> the timing of the timed out sampler.
>
> Hence not able to do the action.
>
> *Expected*.
> Response duration should be able to calculate the timeout time of the
> sampler ,even if the sampler is timed-out.
>

Re: Load testing of hybrid app.

[Adrian Speteanu]

Hi,

If the app uses HTTP or any supported protocol to communicate to its
server, then you can use jmeter to test the performance of the server
(through the api it exposes for the client apps). However, you won't be
able to test the actual performance of the client apps, only the response
times they will experience when making requests to the app server, in
various load scenarios.

Cheers,
Adrian

On Tue, Feb 16, 2016 at 3:31 PM, Pravesh Prajapati <
prajapati.prav...@gmail.com> wrote:

> Hi Team, please let me know, if it is possible to doing performance testing
> on hybrid apps in ios with the help of jmeter, of yes how?.
> Please repley
> Thanks in Adance.
>
> --
> Regards,
> Pravesh prajapati.
> Mob:-9702600170
>

Re: Performance of Jmeter itself

[Adrian Speteanu]

How JMeter sends requests is explained in the online documentation and you
have a correct understanding of that. Without any timers and controllers
impacting throughput and wait periods, each thread will attempt another
request as soon as it receives the response from it's previous request!

This becomes a problem only in the case of resource starvation, a large
thread pool for example will add to the operations that a CPU has to
perform and threads that don't receive processor time will obviously have
to wait until they do. It's also a false positive, sometimes, in the case
of applications with large response times per request. For example, we've
ran into a known issue of the Apache server at one point and every
hundredth request or so was experiencing large response times (from 0.5s of
the 90% percentile to 40+ seconds). Obviously, we first noticed that "not
all threads were active" - a false assumption given by the fact that we
were performing less requests per second than we were expecting. With a low
number of active threads in a test, this becomes an issue that keeps a
thread apparently "inactive" (in terms of requests sent, as the thread is
very much alive and functioning properly) for as long as it takes for the
server to respond. But requests are generated sequentially and that thread
needs to wait for the response to arrive completely before proceeding. This
is covered in the response times though, so you should be able to see when
something unusual is going on.

So try experimenting without that plugin and see if you have issues that
are related to the plugin, behaviour should be similar. What I try to do
when I'm experiencing bad behaviour from the server, I increase the thread
pool available and use the Constant Throughput Timer to generate a constant
rate of requests per minute.

Cheers,
Adrian S

On Tue, Jul 14, 2015 at 5:28 PM, George  wrote:

> Hi,
> well yea for sure JMeter need "time" to create the socket the connection
> and sent the data - clearly.I do also accept the "think times" of
> JMeter.What i mean is if we can say that JMeter is basically sending the
> request out as fast as it is possible for JMeter - right?So there is no
> "speed up" value.If i want to stress my server i just need to multiple the
> load by putting more threads or seting up more jmeter instances and
> multiple the load with every instance.This is also clear for me.
>
> Of course i noticed that if i set the "connection timeout" values to e.g.
> 1ms than the requests goes out very fast but they are all red (fail)
> because telling jmeter to close the connection after 1ms is very stupid :)
> BrGeorge
>
>
>
>  Adrian Speteanu  schrieb am 16:14 Dienstag,
> 14.Juli 2015:
>
>
>  Hi,
>
> I had a test script of a very small server requests (pinging the webapp
> basically). JMeter recorded 5ms average response times, monitoring on the
> server side showed 2ms average response times. The difference is
> justifiable due to the load-balancer in between the test client machine and
> the targeted server + jmeter's own "think times". Anything at the level of
> the millisecond is easily considerable acceptable.  What do you mean by
> speeding up?
>
> Cheers,
> Adrian
>
> On Tue, Jul 14, 2015 at 3:55 PM, George  wrote:
>
> > Hello,
> > i have a question about JMeter's performance.
> >
> > I have a simple test plan with one thread looping 1000 times
> > I have one SMTP sampler targeting my smtp server on port 25 and sending a
> > file of couple Kb.I have no connection timeout or connection close
> defined.
> > Technically everything works perfect.
> > I know i could put some "timer" or "sleep" to slow down sending my smtp
> > request e.g. i put a "sleep 10" and jmeter waits 10 sec between every
> smtp
> > sampler, but what about to "speed up" the sending process?I mean how
> "fast"
> > do JMeter sent out the smtp request? Is it related to my environement?
> > Can i "speed up" jmeter somehow?
> > I have the same question for http as well.
> > BrGeorge
> >
> >
> >
>
>
>
>

Re: Performance of Jmeter itself

[Adrian Speteanu]

Hi,

I had a test script of a very small server requests (pinging the webapp
basically). JMeter recorded 5ms average response times, monitoring on the
server side showed 2ms average response times. The difference is
justifiable due to the load-balancer in between the test client machine and
the targeted server + jmeter's own "think times". Anything at the level of
the millisecond is easily considerable acceptable.  What do you mean by
speeding up?

Cheers,
Adrian

On Tue, Jul 14, 2015 at 3:55 PM, George  wrote:

> Hello,
> i have a question about JMeter's performance.
>
> I have a simple test plan with one thread looping 1000 times
> I have one SMTP sampler targeting my smtp server on port 25 and sending a
> file of couple Kb.I have no connection timeout or connection close defined.
> Technically everything works perfect.
> I know i could put some "timer" or "sleep" to slow down sending my smtp
> request e.g. i put a "sleep 10" and jmeter waits 10 sec between every smtp
> sampler, but what about to "speed up" the sending process?I mean how "fast"
> do JMeter sent out the smtp request? Is it related to my environement?
> Can i "speed up" jmeter somehow?
> I have the same question for http as well.
> BrGeorge
>
>
>

Re: ELK

[Adrian Speteanu]

It requires you to make logstash working properly on the jmeter client
machine. You have to check that it sends the requests with the expected
format. Check the other thread and the blog post for details, is there
something specific not clear?

Regards,
Adrian S

On Mon, Jul 13, 2015 at 4:52 PM, Erez Naim  wrote:

> Can you share how did you connect it with Jmeter since I have succeded in
> setting kibana and elasticsearch on my browser I wonder how to make it work
> with Jmeter...
>
> Thanks!
>
> -Original Message-
> From: Adrian Speteanu [mailto:asp.ad...@gmail.com]
> Sent: Monday, July 13, 2015 4:50 PM
> To: JMeter Users List
> Subject: Re: ELK
>
> Hi,
>
> We have such a setup, it implied quite some tweaking and maintenance time
> allocated, but was required because we have to run tests from multiple
> different location and storing results over time is necessary.
>
> Cheers,
> Adrian S
>
> On Mon, Jul 13, 2015 at 1:30 PM, Erez Naim  wrote:
>
> > Hey guys,
> >
> >
> >
> > Any of you tried using Elasticsearh + Logstash + Kibana in order to
> > visualize jmeter's data and reports?
> >
> >
> >
> > Thanks!
> >
> >
> >
> > Erez Naim  |  QA Lead  |   [image: LinkedIn]
> > <http://www.linkedin.com/profile/view?id=34323263>  [image: vCard]
> > <http://www.theneura.com/vCards/gilad.vcf>  [image: Twitter]
> > <http://twitter.com/#!/giladmeiri/>
> >
> > [image: cid:image004.png@01D01E1B.5F153500]
> >
> > EL AL 2 Street | Herzliya
> > mobile (+972) 50 9 555 686 | fax (408) 689.1366 | skype erez.neura
> >
> >
> >
> > website <http://www.theneura.com/>  |  blog
> > <http://www.startupbitz.com/>  |  twitter
> > <https://twitter.com/intent/user?screen_name=theneura>  |  map
> > <http://goo.gl/maps/ZRkg5>
> >
> >
> >
> >
> >
>
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

Re: Jmeter + realtime Elasticsearch +Kibana

[Adrian Speteanu]

Very nice post.

Thank you,
Adrian S

On Fri, Jul 10, 2015 at 12:01 AM, Deepak Shetty  wrote:

> Hi
> It looks like 1 person found this useful so just sharing the link- Here is
> one more way to draw graphs for JMeter tests
>
>
> http://theworkaholic.blogspot.com/2015/05/graphs-for-jmeter-using-elasticsearch.html
>
> regards
> deepak
>

Re: ELK

[Adrian Speteanu]

Hi,

We have such a setup, it implied quite some tweaking and maintenance time
allocated, but was required because we have to run tests from multiple
different location and storing results over time is necessary.

Cheers,
Adrian S

On Mon, Jul 13, 2015 at 1:30 PM, Erez Naim  wrote:

> Hey guys,
>
>
>
> Any of you tried using Elasticsearh + Logstash + Kibana in order to
> visualize jmeter's data and reports?
>
>
>
> Thanks!
>
>
>
> Erez Naim  |  QA Lead  |   [image: LinkedIn]
>   [image: vCard]
>   [image: Twitter]
> 
>
> [image: cid:image004.png@01D01E1B.5F153500]
>
> EL AL 2 Street | Herzliya
> mobile (+972) 50 9 555 686 | fax (408) 689.1366 | skype erez.neura
>
>
>
> website   |  blog 
>  |  twitter   |  map
> 
>
>
>
>
>

Re: Can jmeter run the ajax request?

[Adrian Speteanu]

Hi,

Jmeter doesn't run client side code out of the box. However, you can
properly implement all HTTP calls with it, regardless of how are they
composed in the browser.

>From my experience with login implementations, it's usually a header or
dynamically generated param or cookie that is not properly implemented in
the HTTP requests sent out by jmeter (given that you are already making all
the proper requests required by the login). Ajax web-rich pages do nothing
special, so in itself has, most probably, no bearing on the issue you've
encountered.

Adrian S

On Thu, Jun 25, 2015 at 10:38 AM, Jesse Zhang 
wrote:

> Hi,all.
> Recently I use the jmeter record a website with actions of login.But when
> replaying the requests recorded,i find that the login is not correctly
> run,then many requests are redirected to the login page.The login request
> are with the header "X-Request-With:XMLHttpRequest",so i guess whether it's
> the problem.Can Jmeter run the ajax request or not?Does anyone has the same
> problems?And how can i fix it?
> Thanks.
>
> Jesse
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

Re: Non HTTP response code: java.net.SocketException,Non HTTP response message: Connection reset by peer: socket write error jmeter

[Adrian Speteanu]

Hi,

This is not a JMeter exception per say, it's a network issue. You must make
sure that your network configuration on the test client machine and on the
test-target server allow for the amount of traffic you intend to generate
with JMeter and that no firewall is interfering with your test.

Regards,
Adrian

On Thu, Feb 26, 2015 at 2:22 PM, Brydges, Jamie 
wrote:

> Hi,
> We're running some load tests with 1000 threads split between 2 JMeter
> instances run through the CMD line, not via the GUI.  We're getting the
> above error on some of our HTTP calls.  This is also causing the call times
> for the calls that come directly below this to become very large (normally
> about 200 - 500 ms, this goes up after the above message to around 5000ms).
>
> Does anyone know what this relates to and how this can be fixed?
>
> (I've tried running 1000 threads split between 4 instances and this hasn't
> worked)
>
>
> Thanks,
> Jamie.
>
>
>
> Jamie Brydges
> Software Developer Tester
> First Databank
> Tel: +(0) 1392 440 154
> Email:  jbryd...@fdbhealth.com
> Website:  www.fdbhealth.co.uk
>
> Follow us on Twitter and LinkedIn<
> http://www.linkedin.com/company/first-databank-fdb-uk>
> Read KnowHow: The FDB Blog
>
> First Databank(tm)
> Drug knowledge that makes a difference
>
>  --
>   This e-mail message is intended only for the personal use of the
>  recipient(s) named above. If you are not an intended recipient, you
>  may not review, copy or distribute this message. If you have received
>  this communication in error, please notify FDB Support
>  (helpd...@fdbhealth.com) immediately by e-mail and delete the
>  original message.
>  --
>

Re: java.net.SocketException: Socket closed

[Adrian Speteanu]

Socket exceptions are usually caused by OS specific configurations (or
better said, lack of required configurations). In this case, JMeter is
trying to re-use a socket which was just closed by the system.

--Adrian S

On Fri, Jan 30, 2015 at 9:49 AM, Jeff Ohrstrom 
wrote:

> Is there a reason you're using Java implementation?  I quickly looked at
> the code and the two (HC3/4 and Java) seem to be reading off the wire in
> different ways.  Perhaps the H4 implementation is much better at dealing
> with redirects?
>
> That, off the top, would be my first guess, that the Java implenation
> when being redirected is doing something funny with the connection. I'm
> not sure if HttpJavaImpl is widely used (I certainly don't ever use
> it).
>
> On Thu, 2015-01-29 at 17:00 -0700, rmil0987 wrote:
> >  I'm developing a script using JMeter 2.12, executing the script against
> > Tomcat 7. I've reached a point in the script development where there are
> ten
> > samplers the send POST requests to a REST web service. After the last of
> the
> > requests the script goe to a logout transaction that executes three
> > samplers. The script stalls on the second sampler in the logout
> transaction.
> > In the View Results Tree you can't see the logout transaction until you
> > click the stop button. When this happens the first logout request
> returns an
> > HTTP 302 response where under normal operation it returns and HTTP 200
> > response. Here's the stack trace as seen in the second logout sampler:
> >
> > java.net.SocketException: Socket closed
> >   at sun.reflect.GeneratedConstructorAccessor42.newInstance(Unknown
> Source)
> >   at
> >
> sun.reflect.DelegatingConstructorAccessorImpl.newInstance(DelegatingConstructorAccessorImpl.java:45)
> >   at java.lang.reflect.Constructor.newInstance(Constructor.java:526)
> >   at
> >
> sun.net.www.protocol.http.HttpURLConnection$6.run(HttpURLConnection.java:1675)
> >   at
> >
> sun.net.www.protocol.http.HttpURLConnection$6.run(HttpURLConnection.java:1673)
> >   at java.security.AccessController.doPrivileged(Native Method)
> >   at
> >
> sun.net.www.protocol.http.HttpURLConnection.getChainedException(HttpURLConnection.java:1671)
> >   at
> >
> sun.net.www.protocol.http.HttpURLConnection.getInputStream(HttpURLConnection.java:1244)
> >   at
> java.net.HttpURLConnection.getResponseCode(HttpURLConnection.java:468)
> >   at
> >
> org.apache.jmeter.protocol.http.sampler.HTTPJavaImpl.readResponse(HTTPJavaImpl.java:258)
> >   at
> >
> org.apache.jmeter.protocol.http.sampler.HTTPJavaImpl.sample(HTTPJavaImpl.java:514)
> >   at
> >
> org.apache.jmeter.protocol.http.sampler.HTTPSamplerProxy.sample(HTTPSamplerProxy.java:74)
> >   at
> >
> org.apache.jmeter.protocol.http.sampler.HTTPSamplerBase.followRedirects(HTTPSamplerBase.java:1486)
> >   at
> >
> org.apache.jmeter.protocol.http.sampler.HTTPSamplerBase.resultProcessing(HTTPSamplerBase.java:1561)
> >   at
> >
> org.apache.jmeter.protocol.http.sampler.HTTPAbstractImpl.resultProcessing(HTTPAbstractImpl.java:338)
> >   at
> >
> org.apache.jmeter.protocol.http.sampler.HTTPJavaImpl.sample(HTTPJavaImpl.java:588)
> >   at
> >
> org.apache.jmeter.protocol.http.sampler.HTTPSamplerProxy.sample(HTTPSamplerProxy.java:74)
> >   at
> >
> org.apache.jmeter.protocol.http.sampler.HTTPSamplerBase.sample(HTTPSamplerBase.java:1141)
> >   at
> >
> org.apache.jmeter.protocol.http.sampler.HTTPSamplerBase.sample(HTTPSamplerBase.java:1130)
> >   at
> >
> org.apache.jmeter.threads.JMeterThread.process_sampler(JMeterThread.java:431)
> >   at
> org.apache.jmeter.threads.JMeterThread.run(JMeterThread.java:258)
> >   at java.lang.Thread.run(Thread.java:745)
> > Caused by: java.net.SocketException: Socket closed
> >   at java.net.SocketInputStream.socketRead0(Native Method)
> >   at java.net.SocketInputStream.read(SocketInputStream.java:152)
> >   at java.net.SocketInputStream.read(SocketInputStream.java:122)
> >   at java.io.BufferedInputStream.fill(BufferedInputStream.java:235)
> >   at java.io.BufferedInputStream.read1(BufferedInputStream.java:275)
> >   at java.io.BufferedInputStream.read(BufferedInputStream.java:334)
> >   at sun.net.www.http.HttpClient.parseHTTPHeader(HttpClient.java:687)
> >   at sun.net.www.http.HttpClient.parseHTTP(HttpClient.java:633)
> >   at
> >
> sun.net.www.protocol.http.HttpURLConnection.getInputStream(HttpURLConnection.java:1323)
> >   at
> >
> sun.net.www.protocol.http.HttpURLConnection.getHeaderField(HttpURLConnection.java:2678)
> >   at
> java.net.URLConnection.getHeaderFieldLong(URLConnection.java:639)
> >   at
> java.net.URLConnection.getContentLengthLong(URLConnection.java:511)
> >   at java.net.URLConnection.getContentLength(URLConnection.java:495)
> >   at
> >
> org.apache.jmeter.protocol.http.sampler.HTTPJavaImpl.readResponse(HTTPJavaImpl.java:224)
> >   ... 12 more
> >
> > There is some int

Re: how to pass 200 reqs/2 millisecond in thread group

[Adrian Speteanu]

Hi,

I'm surprised by those requirements myself. But for applications that
respond to very small requests, it might be possible.

If your system supports 20 concurrent requests, how do you plan on
achieving 200/ms? At 120ms response times, 1 thread will be able to make at
best 8 requests per second. You have to plan your test depending on this.
You can generate a high spike of traffic with JMeter, even if your server
isn't able to handle all the traffic. You will simply need enough threads
to generate that much traffic as you require. The test client machines must
also be able to allow that many network connections as you need in order to
generate that kind of traffic.

Cheers,
A





On Tue, Jan 27, 2015 at 3:25 PM, Mahadev Karpe 
wrote:

> Hi Flavio,
>
> Thanks for quick response.
>
> I would like to share more on the our business scenario.
>
> In our application we are using kinvey for user authentication and its
> passing only 20 reqs/ms by default.
>
> At present Kinvey Supports 20 outgoing requests to be open at any given
> moment. our requests are taking approximately 80 - 120ms to complete (84 -
> 114 to be more exact which averages out to roughly 100ms).
>
> But we need to achieve 500 user load. We have around 10-12
> transactions.(69 request  per thread).
>
> Please let me know how to staggered  requests over the course of a second.
>
> Regards,
> Mahadev
>
> -Original Message-
> From: UBIK LOAD PACK Support [mailto:supp...@ubikloadpack.com]
> Sent: Tuesday, January 27, 2015 6:08 PM
> To: JMeter Users List
> Subject: Re: how to pass 200 reqs/2 millisecond in thread group
>
> Hi,
> Agree with what Flavio advised except for slaves.
> Achieving this throughput is possible with 1 single machine (not even a big
> one) provided you follow best-practices.
>
> Regards
> @ubikloadpack
>
>
>
> On Tuesday, January 27, 2015, Flavio Cysne  wrote:
>
> > Throughput is directly dependent of response times. If a request is
> > responding in 10ms, you'll need 1000 requests to achieve targeted
> > throughput (considering that the response times are 10ms constant).
> > You'll need a bunch of JMeter slaves to achieve 100 requests/ms, or
> equivalent.
> >
> > Try the Throughput Shapping Timer, from JMeter-Plugins, and increase
> > the number of slaves if VM resources aren't enough to achieve the
> > targeted load.
> >
> > If you don't have enough resources for this, you could try BlazeMeter.
> > Em 27/01/2015 09:20, "Mahadev Karpe"  > > escreveu:
> >
> > >  Hi Team,
> > >
> > >
> > >
> > > I am facing issues with our business scenario for 200 requests/2 ms.
> > >
> > >
> > >
> > > I have checked with all throughput timer and tried.
> > >
> > >
> > >
> > > Requests are  staggered over the course of a second.
> > >
> > >
> > >
> > > Please let me know how to pass 200 requests/2ms in one thread group
> > > in Jmeter.
> > >
> > >
> > >
> > > Thanks & Regards,
> > >
> > > *Mahadev Karpe*
> > >
> > > *[image: Description: cid:image001.png@01CF2E27.BC33B020]*
> > >
> > > #1, Kshema Dhama, Global Village, Mysore Road| Bangalore | 560 059|
> > > Karnataka | India |
> > >
> > >  CUG #5773609| Cellphone +91-9902028620|
> > >
> > > mahadev.ka...@mphasis.com  | www.mphasis.com
> > >
> > >
> > >
> > >
> > > Information transmitted by this e-mail is proprietary to Mphasis,
> > > its associated companies and/ or its customers and is intended for
> > > use only by the individual or entity to which it is addressed, and
> > may
> > > contain information that is privileged, confidential or exempt from
> > > disclosure under applicable law. If you are not the intended
> > > recipient or it appears that this mail has been forwarded to you
> > > without proper authority, you are notified that any use or
> > > dissemination of this information in any manner is strictly
> > > prohibited. In such cases, please notify us immediately at
> > > mailmas...@mphasis.com  and delete this mail from your
> > records.
> > >
> >
>
>
> --
>
> Regards
> Ubik Load Pack  Team Follow us on Twitter <
> http://twitter.com/ubikloadpack>
>
>
> Cordialement
> L'équipe Ubik Load Pack  Suivez-nous sur Twitter
> 
> Information transmitted by this e-mail is proprietary to Mphasis, its
> associated companies and/ or its customers and is intended
> for use only by the individual or entity to which it is addressed, and may
> contain information that is privileged, confidential or
> exempt from disclosure under applicable law. If you are not the intended
> recipient or it appears that this mail has been forwarded
> to you without proper authority, you are notified that any use or
> dissemination of this information in any manner is strictly
> prohibited. In such cases, please notify us immediately at
> mailmas...@mphasis.com and delete this mail from your records.
>
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jm

Re: how to control jmeter threads during run time.

[Adrian Speteanu]

Hi,

One more approach is to focus on controlling the throughput generated by
your test, instead of changing the number of threads. Have a larger number
of threads, available, within safe limits of the machine's resources - of
course - and control the throughput generated by the test using timers
(including the constant throughput timer).

Cheers,
Adrian S

On Mon, Dec 22, 2014 at 3:56 PM, Flavio Cysne  wrote:

> Take a look at my blog http://flaviocysne.blogspot.com . there's an
> example
> of to achieve dynamics loading with JMeter. As sebb said you could use
> BeanShell server to chance properties values and togheter with BeanShell
> Sampler make it a powerful feature.
>
> You could also use the Jmeter-Plugins listener that controles throughput,
> the throughput shape listener, and vary its target throughput to achieve
> your requirements.
> Em 22/12/2014 06:54, "mahesh bhasme"  escreveu:
>
> > Hi,
> >
> > I am new to jmeter and using jmeter for load testing. Through jmeter we
> are
> > verifying maximum concurrent session for our web product. Here while
> > running load test ,we want to control threads as per user input in run
> > time. As there any way to do this in jmeter feature or plugin?
> >
> > Thanks for help
> >
> > --
> > Thanks,
> > MAhesh
> >
>

Re: How to test if I am subscribed to Apache Mailing lists without posting a mail?

[Adrian Speteanu]

You might find this useful:
http://www.apache.org/foundation/mailinglists.html

On Thu, Dec 18, 2014 at 6:01 AM, Ben Stover  wrote:
>
> How to test if I am subscribed to Apache Mailing lists without posting a
> mail?
>
> On Apache website there is no such login
>
>
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

Re: Duplicated thread groups all write results into intial group's listeners

[Adrian Speteanu]

Hi,

The listener should record all samples that are present in it's own parent
element or at a lower hierarchical level. If this does not apply than it's
probably a bug.

You should report it and mention the tree structure / elements.

Adrian S


On Mon, Nov 3, 2014 at 11:44 PM, Kevin Farmer  wrote:

> Hi all,
>
> In JMeter v 2.11 r1554548 I have created a test plan with a single thread
> group that after a few database queries has Summary Report listener as a
> child element.  I then right click -> Duplicate that thread group several
> times, and set the test plan to run thread groups consecutively.
>
> When I run the test, all of the results for all of the thread groups show
> up under the original thread group's Summary Report rather than in their
> own.  If I go through each of the new thread groups and do something like
> change the name of the Summary Report and then save it and re-run, the
> results now show up in the Summary Report of the correct thread group.  Is
> this expected behavior?  Or should I be setting this up differently?
>
> Thank you,
> Kevin Farmer
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

Re: Coding JMeter tests

[Adrian Speteanu]

Hi,

+1 for starting this topic.
In my experience, developers split into 2 categories: either they don't
mind using JMeter (but they might still struggle with the interface) or
they feel uncomfortable that they can't "code" the test scenario (not the
actual execution of jmeter, most teams I've worked with managed to embed
JMeter in whatever solution they preferred prior to adopting JMeter for
performance tests). The example itself is pretty illustrative of what can
be accomplished from within an IDE and seems self sufficient to me. Finding
[more of] these examples would definitely help. But to this day, I've never
seen it used in the teams I've worked or collaborated with.

Regards,
Adrian S

On Wed, Nov 12, 2014 at 3:29 PM, Shmuel Krakower  wrote:

> Hi all,
> For a long time I'm struggling to push JMeter to the hands of developers
> within the different development teams I'm working with. Today some of
> those developer know and use JMeter, because it is part of our processes
> (to performance test as much as possible), but they are not happy to take
> such JMeter tasks.
>
> I was discussing this today with one of my colleagues, our pain is that
> JMeter fail to be considered 'sexy', by most developers.
> This is true especially when you look at the recent years changes where
> developers are also taking care for end user / functional testing
> development using frameworks such as cucumber.
>
> It feels like there's a lot of hype and 'cool'-ness (and fun) about
> functional testing as part of agile methods and continuous deployment
> efforts, but not once it comes to performance testing.
>
> I believe that the most immediate blocker is that JMeter takes the
> developer outside of his comfort zone, outside of the IDE. JMeter
> considered as a 'tool'.
>
> If we scope the discussion (for simplicity) to Java developers, I believe
> that if they take JMeter as a library or a framework, which will live
> inside the IDE, so it allows them to write tests in Java code and some
> mechanism to run those tests, it will increase the attraction of JMeter in
> the eyes of developers.
>
> Technically speaking, I guess it should be relative easy to utilize JMeter
> from Java code, so a developer could nicely 'develop' tests and run them as
> ANT / MVN task. Mainly because there are already such integrations to run
> JMeter tests from within the IDE and because JMeter is written in Java -
> thus why not using the Java API to create scripts via code?
>
> I just googled for that and found this post by blazemener:
> http://blazemeter
> .com/blog/5-ways-launch-jmeter-test-without-using-jmeter-gui
> Where point 4 talks about exactly the same thing.
> Does anyone has experience with this approach? Pros and Cons? Ideas?
>
> I don't mean that load tests should run from the IDE, but developers should
> be able to develop scripts and run them (for testing) in their comfort
> zone. The load tests execution should surely run later from a dedicated
> environment / machines.
>
> This is mostly for discussion and as it is based on my own experience and
> my own thoughts, I might be completely wrong here with my feelings and
> assumptions, but would be nice to hear other thoughts on that topic.
>
> Best,
>
> Shmuel Krakower.
> www.Beatsoo.org - re-use your jmeter scripts for application performance
> monitoring from worldwide locations for free.
>

Re: counting threads

[Adrian Speteanu]

Hi,

The execution percentages exceed 100%. I have no idea how JMeter handles
such an edge case as I've always avoided it and it can be accomplished in
multiple different ways.

If you tightly want to control the workload: x% of users execute use-case 1
(i.e. browse), y% of users execute use-case 2 (i.e. shop), etc, etc... then
I would recommend configuring 1 thread group per use case and balance the
number of requests by changing the number of threads each group has and the
timers accordingly (or maybe consider the throughput timer and configure
actual number of requests per minute per thread group).

Back to the question you asked, yes it is possible even with your setup,
just increment a variable per thread, if it reaches max no. of workflows,
then increment a property (variables are local, the prop is global;
concurrency will be an issue you'd have to workaround). You can consider
jmeter's plugins or the beanshell server (
http://jmeter.apache.org/usermanual/best-practices.html). At the end of the
test you can run a debug sampler to get the value, or you can maybe write a
log entry.

But first find a different way to trigger the probability that a single
user executes a part of the scenario you planned (again, my recommendation
would be to avoid it altogether and instead focus on defining user groups
with different behaviours that together create the total load you'd see
once the application runs with real users, both in terms of total users and
in terms of number of hits per page/component).

--Adrian S

On Fri, Oct 17, 2014 at 10:19 AM, Léon Keijser 
wrote:

> Hi,
>
> I've created a testplan with among other things has several Throughput
> Controllers. I've configured about 100 threads and each controller has a
> different percentage configured. Like such:
>
> - Thread group Main
> `- TC 50%
> `- TC 15%
> `- TC 20%
> `- TC 35%
>
> Now what I would like is to count how many threads have completed the
> whole test after Jmeter is done running. Is such a thing possible?
>
> --
> Kind Regards, Met vriendelijke groet,
>
> *Léon Keijser*
> Linux Technical Consultant
>
>
> *SMILE BV*
> Goeman Borgesiuslaan 77 | 3515 ET Utrecht
> T +31 (0)30 22.87.995 | F +31 (0)30 22.87.997
> nl.smile.eu 
>
>

Re: JMeter and NewRelic

[Adrian Speteanu]

Hi Philippe,

On Fri, Aug 22, 2014 at 11:07 PM, Philippe Mouawad <
philippe.moua...@gmail.com> wrote:

> Hello Oliver,
> What exactly do you mean by connecting JMeter output to New Relic ?
> Is it about putting the results from JMeter into NewRelic to be able to see
> all jmeter indicators (response time, errors, response code ...) within New
> Relic dashboard ?
> If so, in my understanding NewRelic proposes Api for different languages
> (Java is one of them) to add custom metrics, I suppose it can be done like
> this using a Listener to do it.
> But it would be interesting to have the official answer from NewRelic and
> why not share it with us.
>
> @Adrian, were you able to have those infos ? I don't think so, but I
> suppose you used the JMeter Graphs and compared with New Relice dashboard
> to give results ?
>

In my case, the projects that required testing have big tech / it ops
support. This is why I'm not familiar with the specific/techie details of
the actual setup. But we did require support from newrelic people for the
initial setup.

In large tests we do very minimal client side monitoring (i.e. jmeter logs,
graphs are out of the question), in either jmeter or the system itself: no
logging apart from console output and minimal machine resources monitoring.
This is needed only to double check test health and results stored in other
tools (ganglia, open tsdb, newrelic, load balancer stats, etc). As such
there is very little to compare. We did double check before running the
test at full speed that results on both sides coincide, but after that
point, monitoring was done only in those tools, whatever was used for each
project, and only on the server under test side! I recommend this setup to
detailed client side monitoring because the first setup allows you more
detailed debugging of the actuall application under test and having both in
place makes little sense when you're confident in your test setup.

I can confirm the results were similar if that's what you need, but I'm not
really sure if the data can be shared publicly... the data is also very
specific to each project (like response times of a particular method, which
is possible in newrelic & open tsdb).

Cheers,
Adrian

>
> A description or blog about this experience would be great ?
>
> Thanks
> Regards
> @philmdot
>
>
> On Fri, Aug 22, 2014 at 5:52 PM, Adrian Speteanu 
> wrote:
>
> > Hi,
> >
> > If you don't get any specific responses from others, I'd recommend
> > contacting their tech support with confidence, if you're using the
> > pro/enterprise version.
> >
> > We used New Relic only to monitor the applications under test and it
> proved
> > a useful way to report performance KPIs, especially since the same tool
> was
> > used in production for monitoring. I'd recommend this setup, unless you
> > need something very specific out of the JMeter logs / graphs.
> >
> > Cheers,
> > Adrian S
> >
> >
> > On Wed, Aug 20, 2014 at 7:01 AM, Oliver Erlewein 
> > wrote:
> >
> > > Hi everyone,
> > >
> > > Has anyone connected JMeter output up to NewRelic? I know BlazeMeter &
> Co
> > > are doing it so it is possible. What I am looking for is to suck in
> > > response time data and error data into NewRelic.
> > >
> > > I thought I'd ask before I go on and try and develop something myself
> > (and
> > > it doesn't look as simple as I first thought). Any help would be
> > > appreciated.
> > >
> > > Cheers Oliver
> > >
> >
>
>
>
> --
> Cordialement.
> Philippe Mouawad.
>

Re: JMeter and NewRelic

[Adrian Speteanu]

Hi,

If you don't get any specific responses from others, I'd recommend
contacting their tech support with confidence, if you're using the
pro/enterprise version.

We used New Relic only to monitor the applications under test and it proved
a useful way to report performance KPIs, especially since the same tool was
used in production for monitoring. I'd recommend this setup, unless you
need something very specific out of the JMeter logs / graphs.

Cheers,
Adrian S


On Wed, Aug 20, 2014 at 7:01 AM, Oliver Erlewein 
wrote:

> Hi everyone,
>
> Has anyone connected JMeter output up to NewRelic? I know BlazeMeter & Co
> are doing it so it is possible. What I am looking for is to suck in
> response time data and error data into NewRelic.
>
> I thought I'd ask before I go on and try and develop something myself (and
> it doesn't look as simple as I first thought). Any help would be
> appreciated.
>
> Cheers Oliver
>

Re: How to increase the JMeter heap size without modifying the jmeter.bat file?

[Adrian Speteanu]

I think on windows you can create a shortcut and specify command line
parameters that the app should take. The shortcut wouldn't depend on
version upgrades, if they use the same location.

However, any "automation" setup should be able to handle a basic
requirement such as passing runtime parameters to the applications they are
executing. So, maybe you should find a way to send those parameters from
whatever tool you are using to automate the start of JMeter.

Adrian


On Sat, Aug 16, 2014 at 12:51 AM, Deepak Shetty  wrote:

> you could probably also use something like
> http://www.programmerplanet.org/projects/jmeter-ant-task/
>
>
> On Fri, Aug 15, 2014 at 2:18 PM, Deepak Shetty  wrote:
>
> > you probably need to raise a bugzilla (you might try adding it to
> JVM_ARGS
> > and see if the last option wins i.e. the option will come twice , but
> this
> > would likely be fragile)
> >
> >
> > On Fri, Aug 15, 2014 at 1:53 PM, albert_newton 
> > wrote:
> >
> >> I am trying to increase the JMeter heap size to get rid of the error
> >> 'java.lang.OutOfMemoryError: PermGen space'. I know that it can be
> >> increased
> >> in the jmeter.bat file by setting the following properties to a higher
> >> desired number:
> >>
> >> set HEAP=-Xms256m -Xmx512m
> >> set PERM=-XX:PermSize=64m -XX:MaxPermSize=128m
> >>
> >> But we are wanting to be able to set it differently ( maybe through
> >> system.properties or jmeter.properties or something else?), so that we
> >> don't
> >> have to modify the jmeter.bat file everytime a new version comes out (we
> >> are
> >> basically automating the installation of JMeter in our application), and
> >> we
> >> are trying to just override the heap size property a different way
> without
> >> tinkering with the jmeter.bat file.
> >>
> >> Does anyone have an idea of how this can be achieved?
> >>
> >>
> >>
> >>
> >>
> >> --
> >> View this message in context:
> >>
> http://jmeter.512774.n5.nabble.com/How-to-increase-the-JMeter-heap-size-without-modifying-the-jmeter-bat-file-tp5720849.html
> >> Sent from the JMeter - User mailing list archive at Nabble.com.
> >>
> >> -
> >> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> >> For additional commands, e-mail: user-h...@jmeter.apache.org
> >>
> >>
> >
>

Re: jmeter - thread count limit

[Adrian Speteanu]

Hi,

+1 to the previous responses, in short you're going to have to experiment
with things and decide what works best for you.

I can only share my experience with rest APIs, though my example is on the
extreme. In the case I will present, the roundtrip of a single request
takes 2-10-20ms and the added size of the request + it's response is
smaller than 1kb. But the app needs to handle many concurrent active users.

To generate and handle the kind of load that was required for such a test,
we used actual physical machines (2x quad core, with 24Gb RAM servers) or
EC2 instances (the extra large memory type of instance). And we were able
to generate a maximum throughput of 30 thousands requests per second, on a
single test client (using the physical machines). The test was able to
sustain this value for long periods of time (but it required supervision).
Assertions and logging done by the test script was reduced to a bare
minimum, enough to make sure that the script is actually executing what we
think it does.

After experimentations, the test script was configured to run smoothly in a
2Gb Max Heap memory instance of jmeter. It spawned 1000-3000 threads and it
ran smoothly (was able to generate constant throughput, when the test
target was stable, for long periods of time, 12-48h and the amount of
errors that were related to the test client was negligible). To generate
30k requests, 10-12 instances of jmeter were used and the machine CPU usage
was 70-90%, but there were many tweaks on the OS side we had to make,
mostly related to the network configuration (after we ran into various
issues). The idea of a 2Gb instance was to be able to scale horizontally
though. In this case, scaling vertically meant using more jmeter instances
and this kind of setup proved more stable and easier to maintain.

So that was the long version of: jmeter can handle as many threads as the
JVM can on that particular machine, which is quite a lot.

But with more complex applications, things will get complicated. The bigger
the response, the more processing you require within jmeter, the more
resources it will require and the number of threads you'll be able to spawn
in a stable setup will decrease dramatically. For something like a website,
you can overwhelm 1 frontend instance of the target server with 50-100
threads (but they will consume a lot of CPU on your test client).

--Adrian


On Tue, Aug 12, 2014 at 6:24 PM, Deepak Shetty  wrote:

> Hi
> >"docs mention there is no hard & fast rule, it depends on system
> hardware/software config & how soon the server responds for APIs under
> test."
> This is the technically correct answer
>
> >" lists a number between 200-300 for number of threads"
> This is most likely a rule of thumb or a starting point (e.g. I will rarely
> exceed 40 threads on my client PC)
>
> >We now need to put more load on it
> It depends on how you define "load" - *Usually* you mean the concurrent
> usage that the server sees in which case option B is the correct one.(It's
> possible you might be running a long running test , just to see whether
> there are memory leaks or something like that and you are using "load" in
> that context in which case you probably only need option A. - happens , but
> rare)
> However lets say you want to test out 10 users accessing your system at the
> same time, for an hour , then you might want to use different users (rather
> than using the same user or  10 users which means you might get better than
> expected results due to server caching user details or something like that)
> - in which case you do also need Option A to make Option B realistic.
> However you cannot just increase the number of threads without limit ,
> because at some point of time , either the JMeter machine or the network
> might start slowing down your tests which would not happen in a real life
> scenario - in which case you now need to add more JMeter machines (either
> distributed or standalone) - What number of threads your system supports is
> dependent on multiple factors and there is no way out of testing and
> validating your results - This has probably been discussed numerous times
> in the archives of these mailing lists.
>
>
>
> regards
> deepak
>
>
>
>
>
>
>
>
> On Mon, Aug 11, 2014 at 6:06 PM, Shilpa Kulkarni 
> wrote:
>
> > Hi
> >
> > I have a question about thread count limit in jmeter. Some of the -
> > slightly older documentation on the internet - lists a number between
> > 200-300 for number of threads. Some docs mention there is no hard & fast
> > rule, it depends on system hardware/software config & how soon the server
> > responds for APIs under test.
> >
> > I have built a basic API test plan for a mobile/server type of
> application
> > & tested for 2 users in csv file & thread count = 10. We now need to put
> > more load on it & I was checking out for any guide lines here. What are
> my
> > options? A + B or just one of them? Are there any other options?
> > A. Increase number of user

Re: Number of requests per time period

[Adrian Speteanu]

The JMeter plugins project on google code is the easiest way to get
responses/throughput per time graphs, especially if you're running the test
in realtime from the GUI. In the plugins, the Hits per second will give you
an RPS graph, no matter how you structure your test script.

Transactions are arbitrarily defined (and throughput is transactions per
time unit), so it depends where you find this notion used and on the people
using it to properly explain it. RPS is what you get from jmeter's log
easiest, as it records/logs each request, but sometimes multiple requests
are needed to complete an action as it is understood by the end-user, so a
transaction is sometimes defined as the collection of requests that serve
that purpose. In your jmeter test you'd have to make sure that your script
hierarchy groups the requests in such a way that you can properly record
such transactions and there throughput. Example: if somebody wanted to know
how many edits per hour does a wiki support, they don't really care that it
takes 5 requests to open the editor and another 3 requests to post a new
version of the page (and the numbers are completely randomly picked, insert
anything there), they just wanted to know how many page versions can be
added to the wiki in one hour.

Note: processing large .jtl files after running big tests, in non-gui mode,
takes some time whatever method you're using to plot your test results
(xslt, any plugin, perl scripts, etc, etc). So I keep one separate instance
of jmeter that runs the same script but with less threads. It generate
significantly less traffic, I arbitrarily picked 1% of the total
throughput, and a smaller log file, so I can get results before the full
logs are parsed (and sometimes analysing the full logs is not needed
anymore).

--Adrian S


On Tue, Jul 15, 2014 at 10:09 PM, Deepak Shetty  wrote:

> Hi
> out of box , the listeners is what you get
> JMeter plugins have some additional graphs (you can check if any does what
> they want - they do have transactions per second etc)
> Finally JMeter results are always available in CSV/XML and you can load/
> and generate graph in any number of ways based on your requirements
>
>
> On Tue, Jul 15, 2014 at 11:14 AM,  wrote:
>
> > Hi,
> >
> > Is there an easy way for me to get number of requests per time period,
> > so I am able to retrospectivly graph time/throughput?
> >
> > Thanks
> >
> > Paul
> >
> > -
> > To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> > For additional commands, e-mail: user-h...@jmeter.apache.org
> >
> >
>

Re: Running JMeter unit tests (test errors)

[Adrian Speteanu]

Hi,

Couldn't help noticing that you're using an old version. Is there a
particular reason why you're not using the latest one?
(I doubt that a lot of people have ran tests on such an old version in the
past year or more...)

Cheers,
Adrian S


On Tue, Jul 8, 2014 at 1:46 AM, Sai Zhang  wrote:

> Hi all,
>
> I downloaded JMeter 2.9 and would like to run its unit test suite.
>
> I followed the instructions in its build.xml file to build JMeter from
> source.
>
> However, the following command fails (well not exactly fails, the test it
> supposes to run fails):
>
> ant junit  (by default: it runs the AllTest class)
>
> or
>
> ant  -Dtest.case=org.apache.jmeter.engine.TestTreeCloner.testCloning junit
>
> More specifically, both commands above result in "Tests run=1, Failures: 0,
> Errors: 1" which seems to be strange to me.
>
> I doubt that I may use the wrong way to execute the unit tests of JMeter,
> so could any one kindly share some tips on running its full test suite?
>
> Thanks a lot
>
> -Sai
>

Re: beanshell suggestion - adding a cookie to the manager

[Adrian Speteanu]

Thwarted by the extra digit. It's because the way I was using the sampler
object, though, that made me experiment with the values setup of the
cookie... However, I still can't make the samplers at a certain level to
use the new cookie, it's the existing cookie manager that I was hoping to
add to.

--Adrian S


On Tue, Jun 10, 2014 at 7:50 PM, Deepak Shetty  wrote:

> Hi
> long expiry = -1l;
> Cookie cookie = new Cookie("cookieAddedByPreprocessor", value, site, path,
> false, expiry);
>
> the expiry parameter is a long
>
>
> On Tue, Jun 10, 2014 at 8:43 AM, Adrian Speteanu 
> wrote:
>
> > Sorry, I've written the code snippet out of the top of my head, I had the
> > imports and the right getCookieCount method name as bellow. the complete
> > version looks like this:
> >
> > -Cookie Manager
> > -HTTP Sampler #1 - simple GET request
> >  \- BS PreProcessor:
> >import org.apache.jmeter.protocol.http.control.CookieManager;
> >import org.apache.jmeter.protocol.http.control.Cookie;
> >CookieManager manager = sampler.getCookieManager();
> >vars.putObject("CookieMananger",manager);
> >value = "test-value";
> >site = "localhost";
> >path = "/";
> >Cookie cookie = new Cookie("cookieAddedByPreprocessor", value, site,
> > path, false, 99);
> >int count = manager.getCookieCount();
> >vars.put("count", String.valueOf(count));
> >manager.add(cookie);
> > -Debug Sampler
> > -HTTP Sampler #2
> > -View Results Tree
> >
> > In the debug sampler I can see the site's cookies,
> >
> CookieManager=rg.apache.jmeter.protocol.http.control.CookieManager@30b49c1d
> > and no "count" variable. HTTP Sampler #1 didn't make the use of the
> cookie.
> >
> > --Adrian S
> >
> >
> > On Tue, Jun 10, 2014 at 6:15 PM, Deepak Shetty 
> wrote:
> >
> > > You have (atleast) two issues (Im assuming you have added a cookie
> > manager
> > > in your test)
> > >
> > > import org.apache.jmeter.protocol.http.control.CookieManager;
> > > CookieManager manager = sampler.getCookieManager();
> > > vars.putObject("CookieMananger",manager);
> > > int count = manager.getCookieCount(); //method is getCookieCount not
> > > getCookiesCount
> > > vars.put("count", String.valueOf(count)); //method accepts string as
> > > key/value , not integer
> > >
> > > ThreadGroup
> > > +Sampler
> > > ++BeanshellPreProcessor
> > > +CookieManager
> > >
> > >
> > >
> > > On Tue, Jun 10, 2014 at 8:05 AM, Adrian Speteanu 
> > > wrote:
> > >
> > > > Thanks for the input. First of all, I didn't realise that I can't use
> > it
> > > in
> > > > a beanshell sampler, but I moved the code to a sampler as opposed to
> a
> > > > preprocessor because it wasn't working. And I can't even do something
> > > like:
> > > >
> > > > CookieManager manager = sampler.getCookieManager();
> > > > vars.putObject("CookieMananger",manager);
> > > > int count = manager.getCookiesCount();
> > > > vars.put("count", count);
> > > >
> > > > CookieManager is null at the level of the pre-processor (just to make
> > > sure,
> > > > the pre-processor is a child to the HTTP Sampler).
> > > >
> > > > So, I'm still not sure that I'm not doing the correct thing, but it
> > > appears
> > > > to not work and I'm not aware of other restrictions.
> > > >
> > > > --Adrian S
> > > >
> > > >
> > > >
> > > > On Tue, Jun 10, 2014 at 5:53 PM, Deepak Shetty 
> > > wrote:
> > > >
> > > > > I dont think sampler HTTPSampler creates a CookieManager - the
> > > framework
> > > > > sets it on the sampler so CookieManager manager =
> > > > > sampler.getCookieManager(); should be returning it null
> > > > >
> > > > > If you are trying to manipulate cookies in places other than a
> > pre/post
> > > > > processor of an HTTPSampler , then on your first HTTPSampler you
> can
> > do
> > > > > something like (in a pre processor scoped to this sampler)
> > > > > CookieManager manager = sampler.getCookieManager();
> > > > > vars.p

Re: beanshell suggestion - adding a cookie to the manager

[Adrian Speteanu]

Sorry, I've written the code snippet out of the top of my head, I had the
imports and the right getCookieCount method name as bellow. the complete
version looks like this:

-Cookie Manager
-HTTP Sampler #1 - simple GET request
 \- BS PreProcessor:
   import org.apache.jmeter.protocol.http.control.CookieManager;
   import org.apache.jmeter.protocol.http.control.Cookie;
   CookieManager manager = sampler.getCookieManager();
   vars.putObject("CookieMananger",manager);
   value = "test-value";
   site = "localhost";
   path = "/";
   Cookie cookie = new Cookie("cookieAddedByPreprocessor", value, site,
path, false, 99);
   int count = manager.getCookieCount();
   vars.put("count", String.valueOf(count));
   manager.add(cookie);
-Debug Sampler
-HTTP Sampler #2
-View Results Tree

In the debug sampler I can see the site's cookies,
CookieManager=rg.apache.jmeter.protocol.http.control.CookieManager@30b49c1d
and no "count" variable. HTTP Sampler #1 didn't make the use of the cookie.

--Adrian S


On Tue, Jun 10, 2014 at 6:15 PM, Deepak Shetty  wrote:

> You have (atleast) two issues (Im assuming you have added a cookie manager
> in your test)
>
> import org.apache.jmeter.protocol.http.control.CookieManager;
> CookieManager manager = sampler.getCookieManager();
> vars.putObject("CookieMananger",manager);
> int count = manager.getCookieCount(); //method is getCookieCount not
> getCookiesCount
> vars.put("count", String.valueOf(count)); //method accepts string as
> key/value , not integer
>
> ThreadGroup
> +Sampler
> ++BeanshellPreProcessor
> +CookieManager
>
>
>
> On Tue, Jun 10, 2014 at 8:05 AM, Adrian Speteanu 
> wrote:
>
> > Thanks for the input. First of all, I didn't realise that I can't use it
> in
> > a beanshell sampler, but I moved the code to a sampler as opposed to a
> > preprocessor because it wasn't working. And I can't even do something
> like:
> >
> > CookieManager manager = sampler.getCookieManager();
> > vars.putObject("CookieMananger",manager);
> > int count = manager.getCookiesCount();
> > vars.put("count", count);
> >
> > CookieManager is null at the level of the pre-processor (just to make
> sure,
> > the pre-processor is a child to the HTTP Sampler).
> >
> > So, I'm still not sure that I'm not doing the correct thing, but it
> appears
> > to not work and I'm not aware of other restrictions.
> >
> > --Adrian S
> >
> >
> >
> > On Tue, Jun 10, 2014 at 5:53 PM, Deepak Shetty 
> wrote:
> >
> > > I dont think sampler HTTPSampler creates a CookieManager - the
> framework
> > > sets it on the sampler so CookieManager manager =
> > > sampler.getCookieManager(); should be returning it null
> > >
> > > If you are trying to manipulate cookies in places other than a pre/post
> > > processor of an HTTPSampler , then on your first HTTPSampler you can do
> > > something like (in a pre processor scoped to this sampler)
> > > CookieManager manager = sampler.getCookieManager();
> > > vars.putObject("CookieMananger",manager)
> > >
> > > Then (CookieManager)vars.getObject("CookieManager") will allow you to
> get
> > > it anywhere
> > >
> > >
> > > On Tue, Jun 10, 2014 at 1:56 AM, Adrian Speteanu 
> > > wrote:
> > >
> > > > Hi guys,
> > > >
> > > > Need a suggestion with a piece of code that is not compiling in
> > > beanshell,
> > > > and I have no idea why it doesn't work. The code is simple and it's
> in
> > a
> > > > beanshell sampler at the moment:
> > > > "
> > > > import org.apache.jmeter.protocol.http.sampler.HTTPSampler;
> > > > import org.apache.jmeter.protocol.http.control.CookieManager;
> > > > import org.apache.jmeter.protocol.http.control.Cookie;
> > > >
> > > > HTTPSampler sampler = new HTTPSampler();
> > > > CookieManager manager = sampler.getCookieManager();
> > > > Cookie cookie = new Cookie("SessionPersistence", "value",
> > > "localhost:8080",
> > > > "/", true, 0);
> > > > manager.add(cookie);
> > > > "
> > > > Fails with
> > > > "Response message: org.apache.jorphan.util.JMeterException: Error
> > > invoking
> > > > bsh method: eval"
> > > > because of the very last line.
> > > >
> > > > Any ideas what to do and why it fails?
> > > >
> > > > Thanks,
> > > > --Adrian S
> > > >
> > >
> >
>

Re: beanshell suggestion - adding a cookie to the manager

[Adrian Speteanu]

Thanks for the input. First of all, I didn't realise that I can't use it in
a beanshell sampler, but I moved the code to a sampler as opposed to a
preprocessor because it wasn't working. And I can't even do something like:

CookieManager manager = sampler.getCookieManager();
vars.putObject("CookieMananger",manager);
int count = manager.getCookiesCount();
vars.put("count", count);

CookieManager is null at the level of the pre-processor (just to make sure,
the pre-processor is a child to the HTTP Sampler).

So, I'm still not sure that I'm not doing the correct thing, but it appears
to not work and I'm not aware of other restrictions.

--Adrian S



On Tue, Jun 10, 2014 at 5:53 PM, Deepak Shetty  wrote:

> I dont think sampler HTTPSampler creates a CookieManager - the framework
> sets it on the sampler so CookieManager manager =
> sampler.getCookieManager(); should be returning it null
>
> If you are trying to manipulate cookies in places other than a pre/post
> processor of an HTTPSampler , then on your first HTTPSampler you can do
> something like (in a pre processor scoped to this sampler)
> CookieManager manager = sampler.getCookieManager();
> vars.putObject("CookieMananger",manager)
>
> Then (CookieManager)vars.getObject("CookieManager") will allow you to get
> it anywhere
>
>
> On Tue, Jun 10, 2014 at 1:56 AM, Adrian Speteanu 
> wrote:
>
> > Hi guys,
> >
> > Need a suggestion with a piece of code that is not compiling in
> beanshell,
> > and I have no idea why it doesn't work. The code is simple and it's in a
> > beanshell sampler at the moment:
> > "
> > import org.apache.jmeter.protocol.http.sampler.HTTPSampler;
> > import org.apache.jmeter.protocol.http.control.CookieManager;
> > import org.apache.jmeter.protocol.http.control.Cookie;
> >
> > HTTPSampler sampler = new HTTPSampler();
> > CookieManager manager = sampler.getCookieManager();
> > Cookie cookie = new Cookie("SessionPersistence", "value",
> "localhost:8080",
> > "/", true, 0);
> > manager.add(cookie);
> > "
> > Fails with
> > "Response message: org.apache.jorphan.util.JMeterException: Error
> invoking
> > bsh method: eval"
> > because of the very last line.
> >
> > Any ideas what to do and why it fails?
> >
> > Thanks,
> > --Adrian S
> >
>

beanshell suggestion - adding a cookie to the manager

[Adrian Speteanu]

Hi guys,

Need a suggestion with a piece of code that is not compiling in beanshell,
and I have no idea why it doesn't work. The code is simple and it's in a
beanshell sampler at the moment:
"
import org.apache.jmeter.protocol.http.sampler.HTTPSampler;
import org.apache.jmeter.protocol.http.control.CookieManager;
import org.apache.jmeter.protocol.http.control.Cookie;

HTTPSampler sampler = new HTTPSampler();
CookieManager manager = sampler.getCookieManager();
Cookie cookie = new Cookie("SessionPersistence", "value", "localhost:8080",
"/", true, 0);
manager.add(cookie);
"
Fails with
"Response message: org.apache.jorphan.util.JMeterException: Error invoking
bsh method: eval"
because of the very last line.

Any ideas what to do and why it fails?

Thanks,
--Adrian S

Re: Clarification regarding the Throughput Controller across thread groups

[Adrian Speteanu]

Hi,

I have used it several times, in a single thread group and I always assumed
(and validated) that it represents a fixed % of all requests, from all
threads (per user wasn't set). I.e.: the samplers under this controller
represented 2% of all transactions recorder, just like the controller was
configured.

However, I haven't got the chance to use it since version 2.9 or 2.10, I
think, as Constant throughput timer suits my test cases better and gives me
the chance to make other interesting configuration.

In addition, be careful of nesting of such timers/controllers, some bugs
were found in the past due to this and it causes problems with
understanding/designing the scenario covered by the script.

Cheers,
Adrian S


On Wed, Mar 12, 2014 at 3:48 PM, Shmuel Krakower  wrote:

> Hi
>
> I am using the Throughput Controller across several thread groups which
> build up my load test scenario.
> I'm having a strange issue where I believe is caused by the fact that the
> throughput controller is sharing the calculation across all thread groups
> if the "Per User" is not marked and I use the "percent executions".
>
> As it is not clear from the documentation, my question is what is the
> designed sharing mode of the Throughput Controller if the "Per User" is not
> set and calculation based on Percent Executions? Is it shared across all
> threads in the specific thread group or across all threads across all
> thread groups?
>
> Best,
>
> Shmuel Krakower.
> www.Beatsoo.org - re-use your jmeter scripts for application performance
> monitoring from worldwide locations for free.
>

Re: Is it possible to launch browser from Jmeter

[Adrian Speteanu]

Hi,

There are third party tools that might be able to help with that (they
literally integrate JMeter with Selenium). But, in my opinion, it defeats
the purpose. You should first consider if that's really the way to go
forward.

Adrian S


On Mon, Mar 3, 2014 at 8:10 AM, Srinivasan, Kavya <
kavya.sriniva...@netapp.com> wrote:

> Hello users,
>
> Please let me know if its possible to launch browsers using Jmeter and
> simulate a user action in the browser as in the way QTP or Selenium does.
>
> Thanks and Regards,
> Kavya Srinivasan
>
>

Re: A test application to test against

[Adrian Speteanu]

Hi,

Interesting task. I had a similar setup recently and instead of showing off
a particular app, I would recommend making a simple web server setup with
established tools. I would exemplify from what I did:

On a Linux/Mac OS machine
- install jdk
- install tomcat and its examples (so you have some pages to hit)
- install varnish or apache in front of tomcat
- setup tomcat to run on port 8080 and apache/varnish to run on 80
- install, setup jmeter and prepare a test suite good enough to compare the
effectiveness of each web server setup
- test tomcat examples standalone (localhost:8080/examples/..,)
- test tomcat/apache setup with the same script
- take down apache, put varnish in place and test tomcat/varnish setup
I've even reconfigured tomcat to read the apps from /var/www/html and
placed some larger htmls there.

It's a very interesting setup for a tutorial on performance testing. You
should mess around with the tomcat/apache/varnish settings to re-configure
the number of available threads (keep a low number to be able to test on
the same machine). It will allow you to test concurrency and denial of
service scenarios. Monitor with Java VisualVM, available if you install jdk
instead of jre. My purpose was to show off in the simplest way the
advantages of a memory cache as opposed to a setup that uses disk
intensively. It's a very simplistic approach, but works well as a tutorial.
If you don't do production specific tweaks, you'll run into minor
performance issues with the response times, even if you don't test with
more concurrent requests than available (recurrent spikes in response
times). I recommend this because it can serve for both very short tests,
and to benchmark the actual tools for production purposes, it allows for a
more in depth analysis (these tools are stable, mature, but require some
tweaking), so the tutorial can go to more advance topics. Show off the
differences in response times by jmeter, when running GUI mode versus
non-GUI mode.

Ultimately, if you really want to have an application that cracks under
pressure, write a very short java application, instead of using a more
complex app, and insert a memory leak kind of scenario (preferably one that
shows off after an intenser test), But I don't think its needed because you
can tweak the jvm parameters so that they don't handle performance tests
and you could cause OOM and heap dumps during a test (but I'm not 100%
sure, didn't need to go that far).

--Adrian S


On Mon, Feb 17, 2014 at 3:28 AM, Oliver Erlewein wrote:

> Hello all,
>
> So I want to train some testers on perf testing with JMeter. What I am
> looking for is a test app (http/https/soap) that has known perf issues that
> I can install and they test against. Has anyone got such a beast or know
> where to find one?
>
> Cheers Oliver
>

Re: JMeter 2.11 Regular Expression Extractor - Source Field - what is that for?

[Adrian Speteanu]

I'm working on 2.11 r1554548 (the released version), and I don't have that
field in the Regular Expression Extractor GUI. It matches the
documentation, in my case. Maybe you have a plugin that adds that?

--Adrian S



On Thu, Feb 13, 2014 at 12:58 AM, Robin D. Wilson wrote:

> No, I meant "Source Field"...
>
> The Regular Expression Extractor GUI interface has the following setup:
>
> Name:   [_]
> Comments: [_]
>
> 
> 
> Response Field to check
> Body O  Body (unescaped) O  Headers O  URL O  Response Code O  Response
> Message
> O
>
> 
> 
> Source Field  [_] <- NEW
> FIELD
> Reference Name:   [_]
> Regular Expression:   [_]
> Template: [_]
> Match No. (0 for Random): [_]
> Default Value:[_]
>
>
> ---END OF PAGE---
>
> (NOTE: the missing ":" on the "Source Field" is intentional - since on the
> interface, it is actually missing.)
>
> I am specifically referring to the "Source Field" which showed up in the
> JM2.11
> stuff (may have been earlier, but I just now noticed it).
>
> --
> Robin D. Wilson
> Sr. Director of Web Development
> KingsIsle Entertainment, Inc.
> VOICE: 512-777-1861
> http://www.kingsisle.com
>
>
> -Original Message-
> From: UBIK LOAD PACK Support [mailto:supp...@ubikloadpack.com]
> Sent: Wednesday, February 12, 2014 4:09 PM
> To: JMeter Users List
> Subject: Re: JMeter 2.11 Regular Expression Extractor - Source Field -
> what is
> that for?
>
> Hello,
> Do you mean Response Field to check field ?
>
> It is documented in
>
> http://jmeter.apache.org/usermanual/component_reference.html#Regular_Expression_
> Extractor
> :
> Response Field to check The following response fields can be checked:
>
>- Body - the body of the response, e.g. the content of a web-page
>(excluding headers)
>- Body (unescaped) - the body of the response, with all Html escape
>codes replaced. Note that Html escapes are processed without regard to
>context, so some incorrect substitutions may be made.
>
> Note that this option highly impacts performances, so use it only when
>absolutely necessary and be aware of its impacts
>- Body as a Document - the extract text from various type of documents
>via Apache Tika (see View Results
> Tree<
> http://jmeter.apache.org/usermanual/component_reference.html#View_Results_T
> ree>Document
> view section).
>
> Note that Body as a Document option can impact performances, so ensure
>it is Ok for your test
>- Headers - may not be present for non-HTTP samples
>- URL
>- Response Code - e.g. 200
>- Response Message - e.g. OK
>
> Headers can be useful for HTTP samples; it may not be present for other
> sample
> types.
>
>
> --
>
> Regards
> Ubik Load Pack  Team Follow us on Twitter
> 
>
>
> Cordialement
> L'équipe Ubik Load Pack  Suivez-nous sur Twitter
> 
>
>
> On Wed, Feb 12, 2014 at 11:03 PM, Robin D. Wilson  >wrote:
>
> > I was editing an old test case on my system, and I popped into the
> > Regular Expression Extractor to check some stuff, and noticed that
> > there is a new
> > field:
> > "Source Field" (above the "Reference Name" field that was already there).
> >
> > I hadn't seen this field before - so I tried to look it up. None of
> > the help pages refer to it, and I couldn't find in the "Changes" page.
> >
> > So, the question arises - what is that for, and how do I use it?
> >
> > --
> > Robin D. Wilson
> > Sr. Director of Web Development
> > KingsIsle Entertainment, Inc.
> > VOICE: 512-777-1861
> > http://www.kingsisle.com
> >
> >
> >
> > -
> > To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> > For additional commands, e-mail: user-h...@jmeter.apache.org
> >
> >
>
>
> 
>
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

Re: Thread group Loop count: How can I tell in which round am I?

[Adrian Speteanu]

Not just like that, there is a method exposed:
http://people.apache.org/~mkostrze/jmeter-docs/docs/api/org/apache/jmeter/functions/ThreadNumber.html

In the script, use this function (try the function helper for details):
${__threadNum}

The thread number is also exposed in test output (the jtl), several
listeners (View Results Table - don't use during high load because of its
high memory consumption). The information is displayed as (example:) tn="Thread
Group 1-1" (first digit group represents the thread group number, second
digit group represents the thread number inside that particular thread
group).

--Adrian S


On Wed, Jan 29, 2014 at 10:40 AM, ZK  wrote:

> Hi,
> just add a  counter
>   to
> your thread group and reference the variable of that counter to find the
> loop count
>
>
> ZK
>
>
>
> --
> View this message in context:
> http://jmeter.512774.n5.nabble.com/Thread-group-Loop-count-How-can-I-tell-in-which-round-am-I-tp5719210p5719218.html
> Sent from the JMeter - User mailing list archive at Nabble.com.
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

Re: Scenario: Login with 1000 Users. Is it required to create 1000 usernames and passwords?

[Adrian Speteanu]

Hi,

I would expect you would require more than 1,000. If your requirements say
that the app needs 1000 concurrent users, then you'd expect to have 1000
concurrent users at any given time out of a larger pool. The exact ratio,
live/total users, depends on specifics (how long is a single user session,
how many users do you expect to have at peak even if they're not active
etc).

You should consider other details, such as how often should you recycle the
users that are active. It doesn't need to be realistic, but the generated
session pool should not surpass real life values (consider the time that
sessions are kept on the server side, the more unique sessions you
generate, the memory requirements increases).

This affects how you configure the Cookie Manager (maybe you'd want to keep
cookies instead of deleting them after each thread run) or your test
script, maybe you should make fewer logins in your script to simulate
real-life scenarios accurately. You have to decide that considering the
specifics of the app under test.

Adrian S


On Fri, Dec 20, 2013 at 6:52 AM, Amit Kumar  wrote:

> Dear All:
>
> I have a query. If I want to perform load testing for 1000 users in JMeter,
> is it required to create 1000 usernames and passwords to get realistic
> result?
>
> --
> Thanks and Regards,
> Amit
>

Re: Heap memory setting in JMeter

[Adrian Speteanu]

Hi,

Make sure you leave memory available for system processes, it doesn't make
sense to allocate 8G of RAM to JMeter when the machine has 8Gb installed,
I'm surprised it checks for this, but its a good thing. Also make sure your
test setup scales vertically. Its advisable to scale horizontally, but if
you really must use a memory intensive script, validate that your setup
does what you expect of it.


Cheers,
Adrian Sp



On Thu, Dec 12, 2013 at 3:00 PM, Muthukrishnan S. <
muthukrishna...@sonata-software.com> wrote:

> Dear All,
>
> The machine where JMeter is installed has 8 GB of RAM with 64-bit
> processor. When I tried changing the setting for Heap memory, I'm able to
> open JMeter without any issues, until I set HEAP=-Xms8192m -Xmx8192m. I was
> just wondering if this can happen, as I never experienced this, usually
> JMeter will take until 1/2 space of the RAM. Just wanted to know if this is
> fine.
>
> Just a note - Also, in other machine it is taking until HEAP=-Xms8192m
> -Xmx8192m which has 9 GB RAM with 64-bit processor.
>
> Thank you in advance!
>
> Thank you,
> Muthu
> Disclaimer: "The materials contained in this email and any attachments may
> contain confidential or legally privileged information. The information
> contained in this communication is intended solely for the use of the
> individual or entity to whom it is addressed and others authorized to
> receive it. If you are not the intended recipient you are hereby notified
> that any disclosure, copying, distribution or taking any action in reliance
> on the contents of this information is strictly prohibited and may be
> unlawful. If you have received this communication in error, please notify
> us immediately by responding to this email and then delete it from your
> system. Sonata is neither liable for the proper and complete transmission
> of the information contained in this communication nor for any delay in its
> receipt"
>

Re: Is my load server causing results to be in-accurate?

[Adrian Speteanu]

Hi,

I disagree with the previous message. The principle is good, but our
approaches differ even if we start from the same general idea. For short I
would max out the test injector as much as possible. And by as much as
possible I mean that you don't want that the way you generate load is
affecting the final results. That's true. But in order to know that for
sure, you should validate your assumptions on how your script will actually
work prior to running an important test!! This is a very important idea
that I can't stress enough.

Just by not using the entire CPU, you're not in the safe zone, maybe
something in your scenario isn't valid and impedes the test from generating
load in a predictable fashion (and by predictable, I mean that it won't
surprise the person that created the script, not making a monotonous
script) or maybe it uses too much resources and so on.

Also, if you use a large portion of the machine's resources, it does NOT
imply that your test will be "unstable" and the results affected. IF the
resource consumption and the generated throughput is predictable and
stable, then you did a good job for one, and you can scale up the CPU
consumption (I would leave 10% available to the system, on an EC2 instance,
if the CPU usage doesn't spike during the test, so look for max average
values, not an absolute average usage). In 99% of the times you should be
fine. Especially when the test is 1,000 threads complex. When designing
tests with more threads, the CPU's ability to cycle them might become a
bottleneck, but again you can validate this a priori by gradually
increasing the traffic generated, by creating a mock script that doesn't
hit the servers and uses test client resources the same way as the actual
script will, by monitoring the client machine, by comparing client side
results with server side results, etc. Apart from the CPU, there's also
some variation added by the networking involved, the monitoring needs to be
broad enough (and it depends on what is the resource that the script will
stress more).

Does it affect results? It can, this is why you need to prepare the test
and to be able to compare results with a third party tool. So the second
part of making sure the test is ok is having a monitoring system in place
on the server side as well. All production servers use some level of
monitoring, so why not have it up and running by the time you test? Its a
must imho.

Once your test performs as you intend it, then you won't have to worry
about test injector machines' resource consumption, because you should know
exactly how it will behave. If resources are in check then, the test
results won't be affected either.  Unless it has a bug that we don't know
about :). For simple metrics, I don't think it has.

Cheers,
Adrian S



On Wed, Oct 9, 2013 at 7:54 AM, Shmuel Krakower  wrote:

> From my experience the CPU on the load generator should remain idle if you
> want to get stable results from the test, such results where you can
> actually compare with previous tests without the question of whether the
> load generator is overloaded.
>
> A gross number I go with is at the range of 30% CPU usage or less on the
> JMeter host.
>
> Of-course you always need to take into consideration the memory is not
> swapped in OS level (so it can be ~100% RAM usage).
> One other thing is JMeter JVM GC workloads, which should be minimal, you
> should make sure JMeter is not spending too much effort on GC, no matter
> what's the configured heap size.
>
> But that's a good question, I'd also like to know what others are doing.
>
>
>
> Shmuel Krakower.
> www.Beatsoo.org - re-use your jmeter scripts for application performance
> monitoring from worldwide locations for free.
>
>
> On Wed, Oct 9, 2013 at 5:08 AM, Tim Koopmans  wrote:
>
> > At flood.io we find a better measure of performance and impact on test
> > results is JVM heap utilization.
> >
> > For example, this benchmark https://flood.io/954b7d5d79f134 shows
> > degradation of response time over time as heap utilization increases
> >
> >
> https://github.com/flood-io/flood-loadtest/blob/master/benchmarks/results/954b7d5d79f134.md
> >
> > Having said that we were running 30K users on a single JVM. You can find
> > out more about our benchmarks here:
> > https://flood.io/blog/11-benchmarking-jmeter-and-gatling
> >
> > You can correlate increased CPU of course with heavy resource utilization
> > within the JVM, but looking at CPU alone is like trying to measure
> rainfall
> > by listening to it fall on the roof.
> >
> > Regards,
> > Tim
> >
> >
> >
> >
> > Tim Koopmans
> > +61 417 262 008
> >
> > 
> >
> > The Automation Company
> >
> >
> >
> > On Wed, Oct 9, 2013 at 1:00 PM, Ophir.Prusak <
> ophir.pru...@blazemeter.com
> > >wrote:
> >
> > > I'm running a JMeter test using JMeter on an amazon EC2 instance
> (large)
> > as
> > > the load server using 1,000 threads. The load server CPU is steady at
> > about
> > > 90% utilization and memory

Re: Maximum recommended number of ThreadGroups?

[Adrian Speteanu]

***Correction:
And by forum, I meant folder. Apologies!

On Wed, Oct 9, 2013 at 7:16 PM, Adrian Speteanu  wrote:

> I think its possible. But should you do that for functional tests? Nope.
>
> I use the jmeter maven plugin to automate functional tests when they're
> scripted with JMeter, in conjuction with the Jenkins' performance plugin.
> In this case, tests are easily run, Everything in the forum is executed in
> a single execution. Thus I keep one different script for each major test.
> Its easier to update and if a script is re-usable then I use the Include
> controller to re-use them. Its not the best and the cleanest method, but
> the scripts are maintained in a way that they are readable and easy to
> modify.
>
> A similar method would be more advisable.
>
> Adrian S
>
>
> On Fri, Oct 4, 2013 at 1:07 PM, Jean FX  wrote:
>
>> I meant
>>
>> The View Results Tree listener allows me to generate automated functional
>> test results when each functional test is within a ThreadGroup result,
>> because I am able to save the thread group's name in the results file.
>>
>>
>>
>> On Fri, Oct 4, 2013 at 11:56 AM, Jean FX  wrote:
>>
>> > Hello,
>> >
>> > Thanks for the reply.
>> >
>> > The View Results Tree listener allows me to generate automated
>> functional
>> > test results when each functional test is within a test result, because
>> I
>> > am able to save the thread name in the results file.
>> >
>> > I could not identify a result listener that saves the name of the
>> > controller within which a sampler is located with the sampler results.
>> >
>> > Jean
>> >
>> >
>> > On Fri, Oct 4, 2013 at 10:56 AM, bobMeliev  wrote:
>> >
>> >> Well, seems it will be hard for you to maintain and change Thread
>> >> properties.
>> >> If you want group functional tests why don't you use "Simple
>> Controller"?
>> >>
>> >>
>> >>
>> >> --
>> >> View this message in context:
>> >>
>> http://jmeter.512774.n5.nabble.com/Maximum-recommended-number-of-ThreadGroups-tp5718307p5718331.html
>> >> Sent from the JMeter - User mailing list archive at Nabble.com.
>> >>
>> >> -
>> >> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
>> >> For additional commands, e-mail: user-h...@jmeter.apache.org
>> >>
>> >>
>> >
>>
>
>

Re: Maximum recommended number of ThreadGroups?

[Adrian Speteanu]

I think its possible. But should you do that for functional tests? Nope.

I use the jmeter maven plugin to automate functional tests when they're
scripted with JMeter, in conjuction with the Jenkins' performance plugin.
In this case, tests are easily run, Everything in the forum is executed in
a single execution. Thus I keep one different script for each major test.
Its easier to update and if a script is re-usable then I use the Include
controller to re-use them. Its not the best and the cleanest method, but
the scripts are maintained in a way that they are readable and easy to
modify.

A similar method would be more advisable.

Adrian S


On Fri, Oct 4, 2013 at 1:07 PM, Jean FX  wrote:

> I meant
>
> The View Results Tree listener allows me to generate automated functional
> test results when each functional test is within a ThreadGroup result,
> because I am able to save the thread group's name in the results file.
>
>
>
> On Fri, Oct 4, 2013 at 11:56 AM, Jean FX  wrote:
>
> > Hello,
> >
> > Thanks for the reply.
> >
> > The View Results Tree listener allows me to generate automated functional
> > test results when each functional test is within a test result, because I
> > am able to save the thread name in the results file.
> >
> > I could not identify a result listener that saves the name of the
> > controller within which a sampler is located with the sampler results.
> >
> > Jean
> >
> >
> > On Fri, Oct 4, 2013 at 10:56 AM, bobMeliev  wrote:
> >
> >> Well, seems it will be hard for you to maintain and change Thread
> >> properties.
> >> If you want group functional tests why don't you use "Simple
> Controller"?
> >>
> >>
> >>
> >> --
> >> View this message in context:
> >>
> http://jmeter.512774.n5.nabble.com/Maximum-recommended-number-of-ThreadGroups-tp5718307p5718331.html
> >> Sent from the JMeter - User mailing list archive at Nabble.com.
> >>
> >> -
> >> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> >> For additional commands, e-mail: user-h...@jmeter.apache.org
> >>
> >>
> >
>

Re: Multiple Jmeter instance on same machine.

[Adrian Speteanu]

Hi Ivan,

For both reasons actually. In order to better organise and collaborate my
scripts, I keep different test sessions scripted in different test plans
instead of different thread groups, even if I intend to run them
simultaneously at some point. Its just a preference and not necessarily
relevant. The biggest reason is indeed memory/cpu and garbage collection
optimisation (but I run JMeter on very powerful machines lately, 32-64Gb
boxes, so its easy to get tempted not to optimise the test bed, I wouldn't
worry as much on 4Gb boxes for example). The scripts I've used in the last
year were simple and are perform similarly when using one single large
instance of JMeter or when using multiple instances. The performance
optimisations that devs implemented in versions 2.8 and 2.9 are very
helpful to this outcome. And I specifically needed to run a single instance
that synchronised 18,000 threads to query the DB at roughly the same time.
Accordingly to the monitoring system, most requests were received by the
server within 5ms (considering the network latency were I tested, that's
actually pretty good).

But in the past, I got into a lot of trouble by not validating my
assumptions about how my script, itself, will perform. Its easier to
predict how a small instance will perform even over very long periods of
time (I test, monitor the machines and make sure that its how I presume,
from time to time), and the behaviour is stable out of the box (very few
changes needed to JVM parameters). I limit a single instance at 2Gb and CPU
consumption is always predictable, 4-5% in my case, for each instance,
including system + idle times.

BTW, all the specifics mentioned were tested with non-GUI mode.

Adrian S

On Fri, Aug 23, 2013 at 1:01 PM,  wrote:

> If I may ask, are the multiple Jmeter instances used:
> - to run different test scripts
> or
> - because multiple VM use better the memory, with more frequent but less
> blocking garbage collection...
>
> Thanks
> Ivan
>
> -----Ursprüngliche Nachricht-
> Von: Adrian Speteanu [mailto:asp.ad...@gmail.com]
> Gesendet: Freitag, 23. August 2013 10:26
> An: JMeter Users List
> Betreff: Re: Multiple Jmeter instance on same machine.
>
> Hi,
>
> I always run multiple instances on the same machines, but never tried it
> with distributed mode.
>
> So, to the question of multiple instances: it is definitely possible and
> scales quite nicely, but do be careful on memory allocation (the total -Xmx
> in your configuration should not exceed the total memory that you have
> available on the machine; leave some for the OS as well).
>
> Adrian S
>
>
> On Fri, Aug 23, 2013 at 2:10 AM, Jomebrew  wrote:
>
> > I have used both but prefer GUI.  I can run more instances when I use
> > CLI mode.
> >
> > Cheers!
> >
> >
> > On Thu, Aug 22, 2013 at 3:57 PM, hame  wrote:
> >
> > > Hey thanks for you quick response. even I thought that it will work
> > > if I use GUI.
> > > What if I don't want to use GUI and run in command line mode? will
> > > it
> > work?
> > >
> > >
> > >
> > >
> > > On Thu, Aug 22, 2013 at 3:44 PM, Jomebrew 
> wrote:
> > >
> > > > I run about four client instances on my server.  Each one run is
> > > graphical
> > > > mode (against the advise of the message list) but my server has
> > > > enough processor cores (8) and memory to handle the load.  Memory
> > > > used to be a problem but with later releases,much more CPU is
> > > > being used. I have
> > been
> > > > running the same basic scripts since 2.6.
> > > >
> > > > Each instance runs two thread groups with 250 threads each for a
> > > > total
> > of
> > > > 500 threads per instance.  I use the aggregate report to monitor
> > > > the
> > load
> > > > status.
> > > >
> > > > This all works well for me.
> > > >
> > > > Cheers!
> > > >
> > > >
> > > > On Thu, Aug 22, 2013 at 3:40 PM, hame...@gmail.com
> > > >  > > > >wrote:
> > > >
> > > > > Hi,
> > > > >  I want to run different test-plan by running multiple instances
> > > > > of
> > > > JMeter
> > > > > on the same machine at the same time. Is it possible?
> > > > >  I know I can run multiple jmeter server instances on different
> > > machine.
> > > > > But
> > > > > I am not sure whether I can run different instances of JMeter on
> > > > > the
> > > same
> > > > > 

Re: Multiple Jmeter instance on same machine.

[Adrian Speteanu]

Hi,

I always run multiple instances on the same machines, but never tried it
with distributed mode.

So, to the question of multiple instances: it is definitely possible and
scales quite nicely, but do be careful on memory allocation (the total -Xmx
in your configuration should not exceed the total memory that you have
available on the machine; leave some for the OS as well).

Adrian S


On Fri, Aug 23, 2013 at 2:10 AM, Jomebrew  wrote:

> I have used both but prefer GUI.  I can run more instances when I use CLI
> mode.
>
> Cheers!
>
>
> On Thu, Aug 22, 2013 at 3:57 PM, hame  wrote:
>
> > Hey thanks for you quick response. even I thought that it will work if I
> > use GUI.
> > What if I don't want to use GUI and run in command line mode? will it
> work?
> >
> >
> >
> >
> > On Thu, Aug 22, 2013 at 3:44 PM, Jomebrew  wrote:
> >
> > > I run about four client instances on my server.  Each one run is
> > graphical
> > > mode (against the advise of the message list) but my server has enough
> > > processor cores (8) and memory to handle the load.  Memory used to be a
> > > problem but with later releases,much more CPU is being used. I have
> been
> > > running the same basic scripts since 2.6.
> > >
> > > Each instance runs two thread groups with 250 threads each for a total
> of
> > > 500 threads per instance.  I use the aggregate report to monitor the
> load
> > > status.
> > >
> > > This all works well for me.
> > >
> > > Cheers!
> > >
> > >
> > > On Thu, Aug 22, 2013 at 3:40 PM, hame...@gmail.com  > > >wrote:
> > >
> > > > Hi,
> > > >  I want to run different test-plan by running multiple instances of
> > > JMeter
> > > > on the same machine at the same time. Is it possible?
> > > >  I know I can run multiple jmeter server instances on different
> > machine.
> > > > But
> > > > I am not sure whether I can run different instances of JMeter on the
> > same
> > > > machine or not? Will it create any kind of interference for
> eachother?
> > > >
> > > >
> > > >
> > > > Thank you,
> > > > Rakesh
> > > >
> > > >
> > > >
> > > > -
> > > > Technical blog : code4reference
> > > > LinkedIn :  Rakesh Kumar
> > > > --
> > > > View this message in context:
> > > >
> > >
> >
> http://jmeter.512774.n5.nabble.com/Multiple-Jmeter-instance-on-same-machine-tp5717948.html
> > > > Sent from the JMeter - User mailing list archive at Nabble.com.
> > > >
> > > > -
> > > > To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> > > > For additional commands, e-mail: user-h...@jmeter.apache.org
> > > >
> > > >
> > >
> >
>

Re: Number of request per second.

[Adrian Speteanu]

Hi,

It is possible to load a jtl after the test, you will also be able to check
the throughput.

The reason the throughput might not be the expected one can vary. It might
be that the way you designed a single "virtual user session" is not ok for
this goal (requests in the thread) and then you cannot achieve this much
throughput with only 800 threads (so check that the script does what you
think it does first of all) and then, if the script is ok, if the server
response times increase to a value that is unexpected, than the capacity to
generate throughput will decrease. Each thread is a single line of
execution, it waits for the previous response to arrive in order to
continue its execution...

Quick example:
* simple script: 100 threads
   - 1 thread: is designed to make aprox. 10 requests per second (this
assumes response times are ~100ms in average)
   - therefore, the test (100 threads) can achieve 100x10=1,000 requests
per second
   - I add a throughput timer because I do not want to make more than 1,000
rps at any time (its a hard limit in my app, or whatever other reason)
What happens if my test bed doesn't work as expected? Lets say that network
setup brings a 100ms latency in addition to the apps expected 100ms, and
then recorded response times are 200 ms instead of 100ms. Than my
throughput will be 500 rps.

What I do in these cases, I design a test with 1,000 rps set in throughput
constant timer which I know is what I would expect out of 100 users. But,
the script will be configured to spawn 300 threads instead of 100. This
way, the script will have more threads that are on hold, and they can
sustain the rate of 1,000, even if some of the threads are waiting for
responses. This is what would happen in real life, if server responds
slower and slower, user requests keep arriving, so its another scenario to
test for.

Cheers,
Adrian S

On Sat, Aug 10, 2013 at 8:14 PM, hame  wrote:

> I did notice the throughput in the report. But looks like those number are
> way-off and I want to validate this.
>
> Currently I am running 800 concurrent thread to simulate users. I have also
> used the constant throughput timer to make sure I get 600 request per
> second. But when I check the throughput I get close to 350.
>
>
> Is it possible to calculate the number of request call from Jtl file? If I
> group the records based on the second.
>
>
> On Sat, Aug 10, 2013 at 3:12 AM, Deepak Goel  wrote:
>
> > Yep
> >
> > On 8/10/13, Shmuel Krakower  wrote:
> > > The report contains the throughput. Isn't that what you are looking
> for?
> > > On Aug 10, 2013 9:41 AM, "Deepak Goel"  wrote:
> > >
> > >> Hey
> > >>
> > >> Little's Law:
> > >>
> > >> Average number of request per second =  Number of Concurrent Users
> > >> divided by (Response Time and Think Time)
> > >>
> > >> Average number of request per second =  Number of Concurrent Users /
> > >> (Response Time + Think Time)
> > >>
> > >> Think Time is the time between two subsequent request.
> > >>
> > >> :)
> > >> Deepak
> > >>
> > >>
> > >> On 8/10/13, hame...@gmail.com  wrote:
> > >> > Currently I am using JMeter to generate load for webapp. I am able
> to
> > >> > simulate 800 simultaneous users. I store the test result in output
> > file
> > >> to
> > >> > get the aggregate report. I am able to get the avg response time for
> > >> > the
> > >> > request but report doesn't contain average number of
> request/response
> > >> > per
> > >> > second.  Is there any way to calculate the number of
> request/response
> > >> > per
> > >> > second?
> > >> >
> > >> >
> > >> >
> > >> > Thank you,
> > >> >
> > >> >
> > >> >
> > >> >
> > >> > -
> > >> > Technical blog : code4reference
> > >> > LinkedIn :  Rakesh Kumar
> > >> > --
> > >> > View this message in context:
> > >> >
> > >>
> >
> http://jmeter.512774.n5.nabble.com/Number-of-request-per-second-tp5717812.html
> > >> > Sent from the JMeter - User mailing list archive at Nabble.com.
> > >> >
> > >> >
> -
> > >> > To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> > >> > For additional commands, e-mail: user-h...@jmeter.apache.org
> > >> >
> > >> >
> > >>
> > >>
> > >> --
> > >> Namaskara~Nalama~Guten Tag~Bonjour
> > >>
> > >>
> > >>--
> > >> Keigu
> > >>
> > >> Deepak
> > >> 7350012833
> > >> deic...@gmail.com
> > >> http://www.simtree.net
> > >>
> > >> Skype: thumsupdeicool
> > >> Google talk: deicool
> > >> Blog: http://loveandfearless.wordpress.com
> > >> Facebook: http://www.facebook.com/deicool
> > >>
> > >> "Contribute to the world, environment and more :
> > >> http://www.gridrepublic.org
> > >> "
> > >>
> > >> -
> > >> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> > >> For additional commands, e-mail: user-h...@jmeter.apache.org
> > >>
> > >>
> > >
> >
> >
> > --
> > Namaskara~Nalama~Guten Tag~Bonjour
> >
> >
> >--
> > Keigu
> >
> > Deepak
> 

Re: Load testing, Continuous Integration, failing on build-over-build degradation

[Adrian Speteanu]

I understand now, never thought of it. I just look at the older graphs and
compare (because it plots all available graphs, and I keep some of the old
results for a while), but don't fail the build automatically, they're side
by side.

If I wanted to do that, there are various hacks that could be used.
Duration assertions, for once, but if you are looking for thresholds for
the entire data set, then maybe with beanshell. But it should be the only
sampler that fails if you want to be notified. I was thinking of
transaction controller too, but there are too few usecases where this is
useful.

No pretty way to do this, duration assertion should be ok, but might be
tricky. Let's explore this: you set a threshold which is quite close to the
values obtained at the moment when creating this regression test script. If
the difference between the threshold value and the expected/current
response time is smaller than the standard deviation, then the duration
assertion will fail requests that are in the normally distributed set.
That's bad. You also have to tweak the expected error percentage threshold
that we thought was the feature you were looking for. Doable, not pretty at
all.

Cheers,
Adrian S



On Tue, Jul 16, 2013 at 5:10 PM, Marc Esher  wrote:

> So to be clear: that's simply detecting errors that rise above a certain
> threshold. But currently, there's no way to track performance degradation
> over time, correct?
>
> What I want is an automatic way to spot degradation job-over-job, such that
> Jenkins would realize "Your tests are now 10% slower than they were a week
> ago".
>
> Or is that asking for too much, and perhaps for trouble?
>
>
> On Tue, Jul 16, 2013 at 10:05 AM, Cedrick Johnson <
> cjohn...@backstopsolutions.com> wrote:
>
> > This is contained in Jenkins. I don't know about Maven and setting that
> > up. We are using the standard JMeter plugin within Jenkins. If you
> activate
> > it that should work. Here's my Build step in Jenkins (Execute Shell and
> > yeah, we're still on 2.8)
> >
> > rm -f *.jtl
> > $HOME/apache-jmeter-2.8/bin/jmeter -n -t SomeTestPlan.jmx -JServerName=
> > wee.com -JServerPort=8080 -JUserThreads=50 -JUserLoopCount=1 -l
> > RhubarbTestResults.jtl
> >
> > That works for us, and has caught some pretty big design changes that
> > slowed things down.
> >
> > -c
> >
> > -Original Message-
> > From: Shmuel Krakower [mailto:shmul...@gmail.com]
> > Sent: Tuesday, July 16, 2013 1:29 AM
> > To: JMeter Users List
> > Subject: RE: Load testing, Continuous Integration, failing on
> > build-over-build degradation
> >
> > Hi Cedrick
> > Thanks for sharing but is this post build action is part of the maven
> > plugin or part of jenkins?
> >
> > I am looking for exactly this capability for couple of months now! Can
> you
> > point on any link to brief introduction of this as I couldn't find any.
> >  On Jul 15, 2013 10:01 PM, "Cedrick Johnson" <
> > cjohn...@backstopsolutions.com>
> > wrote:
> >
> > > When you configure your JMeter Jenkins job, in Post-Build actions you
> > > can have it publish the performance test result report which points to
> > > the Test Results .jtl file that is generated when running the test. In
> > > that report, there's a Performance Threshold section where you can set
> > > it to identify when the build is unstable (number of errors exceeds
> > > this percentage
> > > amount) or build Failed when the number of errors exceeds this set
> > amount.
> > >
> > > The errors are determined in your actual load test, i.e. if requests
> > > start timing out, or other conditions that you are checking in your
> > > tests begin failing they will count against this threshold and Jenkins
> > > will alert you to a degradation in performance once those errors are
> met.
> > >
> > > -c
> > >
> > > -Original Message-
> > > From: Shmuel Krakower [mailto:shmul...@gmail.com]
> > > Sent: Monday, July 15, 2013 1:54 PM
> > > To: JMeter Users List
> > > Subject: Re: Load testing, Continuous Integration, failing on
> > > build-over-build degradation
> > >
> > > Hi Adrian
> > > Thanks for sharing but how exactly u control the response times
> > > thresholds or error rates?
> > > I cannot find any control of this...
> > >  On Jul 15, 2013 4:26 PM, "Adrian Speteanu" 
> wrote:
> > >
> > > > Hi,
> > > >
> > > > Check my attempt of an answer bel

Re: Load testing, Continuous Integration, failing on build-over-build degradation

[Adrian Speteanu]

On Mon, Jul 15, 2013 at 11:01 PM, Cedrick Johnson <
cjohn...@backstopsolutions.com> wrote:

> When you configure your JMeter Jenkins job, in Post-Build actions you can
> have it publish the performance test result report which points to the Test
> Results .jtl file that is generated when running the test. In that report,
> there's a Performance Threshold section where you can set it to identify
> when the build is unstable (number of errors exceeds this percentage
> amount) or build Failed when the number of errors exceeds this set amount.
>
> The errors are determined in your actual load test, i.e. if requests start
> timing out, or other conditions that you are checking in your tests begin
> failing they will count against this threshold and Jenkins will alert you
> to a degradation in performance once those errors are met.
>
> -c
>
>
+1
We did this for functional tests, covered known bugs as well and only
wanted to mark as unstable, instead of completely failing the job.


> -Original Message-
> From: Shmuel Krakower [mailto:shmul...@gmail.com]
> Sent: Monday, July 15, 2013 1:54 PM
> To: JMeter Users List
> Subject: Re: Load testing, Continuous Integration, failing on
> build-over-build degradation
>
> Hi Adrian
> Thanks for sharing but how exactly u control the response times thresholds
> or error rates?
> I cannot find any control of this...
>  On Jul 15, 2013 4:26 PM, "Adrian Speteanu"  wrote:
>
> > Hi,
> >
> > Check my attempt of an answer bellow.
> >
> > Regards,
> > Adrian S
> >
> > On Mon, Jul 15, 2013 at 2:56 PM, Marc Esher 
> wrote:
> >
> > > Greetings all,
> > >
> > > I'm integrating our load tests into our CI environment, with the
> > > goal of identifying performance degradation as soon as possible. The
> > > idea is is
> > to
> > > use some kind of threshold, from one CI build to the next, to
> > > identify
> > when
> > > performance has dipped to an unacceptable level from one run to
> another.
> > >
> > > I'm using Jenkins, currently.
> > >
> > > Anyone have any guidance, strategy, experience, wisdom here?
> > >
> > > The Jenkins Performance Plugin is decent for reporting trends, but
> > > it has no capabilities to automatically spot problems.
> > >
> >
> > What is your exact expectation regarding to this last phrase?
> >
> > I'm currently using the maven plugin, and it integrates nicely with
> > the jenkins plugin that you mentioned. The tests fail when expected.
> > Here are the configurations made to the pom.xml (I followed the
> > tutorial from the jenkins plugin project when first setting up this
> > test project). The threshold for failures are set in the jenkins plugin
> and they work.
> >
> > com.lazerycode.jmeter
> > jmeter-maven-plugin
> > ...
> > 
> > 
> > jmeter-tests
> > verify
> > 
> > jmeter
> > 
> > 
> > 
> >
> > execution: #mvn clean verify
> >
> >
> > > Thanks!
> > >
> > > Marc
> > >
> >
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>

Re: Load testing, Continuous Integration, failing on build-over-build degradation

[Adrian Speteanu]

Hi,

Check my attempt of an answer bellow.

Regards,
Adrian S

On Mon, Jul 15, 2013 at 2:56 PM, Marc Esher  wrote:

> Greetings all,
>
> I'm integrating our load tests into our CI environment, with the goal of
> identifying performance degradation as soon as possible. The idea is is to
> use some kind of threshold, from one CI build to the next, to identify when
> performance has dipped to an unacceptable level from one run to another.
>
> I'm using Jenkins, currently.
>
> Anyone have any guidance, strategy, experience, wisdom here?
>
> The Jenkins Performance Plugin is decent for reporting trends, but it has
> no capabilities to automatically spot problems.
>

What is your exact expectation regarding to this last phrase?

I'm currently using the maven plugin, and it integrates nicely with the
jenkins plugin that you mentioned. The tests fail when expected. Here are
the configurations made to the pom.xml (I followed the tutorial from the
jenkins plugin project when first setting up this test project). The
threshold for failures are set in the jenkins plugin and they work.

com.lazerycode.jmeter
jmeter-maven-plugin
...


jmeter-tests
verify

jmeter




execution: #mvn clean verify


> Thanks!
>
> Marc
>

Re: (Exponentially Distributed) CycleTime instead of ThinkTime in JMeter

[Adrian Speteanu]

Hi,

The concept is simple enough and interesting. But I think you can already
achieve what you're asking for.

Why would it be needed in the case of random timers? Let's say you have
cycle timer, and you go for a random period. Isn't the period added by the
cycletime minus existing thinktime covered in the deviation? If I'd want to
simulate what a single user does in a cycle, and the response times of the
app get in my way, I'd just adjust the offset and the deviation, at the
moment and I obtain random cycle times in the same range as if I could
achieve if I had the cycle timer option. It would add extra calculation and
little benefit. I fail to understand why its needed, but would support the
idea, because a lot of people struggle to understand how JMeter generates
its throughput, why v-users's actions get out of sync after the first few
thread loops. With a cycle time, it would be easier to synchronise threads
if its really needed. But its not a usecase that should be recommendable.

This is how I understand your request (an example of why I think the
functionality is available in a different form), please correct me if my
understanding is wrong:

If resp times degrade over time (justifiable by new features added), you
wouldn't have to adjust test scripts, with the cycle timer option, you'd
get the same amount of actions that you would have expected when you first
designed the test. This is one case where I would like such control of a
single threads throughput. But we do have something similar at the moment,
 I use the Constant Throughput Timer for this, with the per thread option.
Is this not enough?

Cheers,
Adrian Speteanu

On Tue, Jul 9, 2013 at 7:40 PM, Fabian Brosig  wrote:

> Hi,
>
> In my JMeter script, I wanted to configure a timer with an exponential
> distributed delay.
> However, using the Poisson timer, I could configure only the think time,
> not the cycle time. (In short: cycleTime = responseTime + thinkTime)
>
> Here is a more detailed explanation, what "Cycle Time" is, and why it is
> useful in load testing scenarios... (citation from
> http://www.testnscale.com/docs/CycleTimesTutorial.html)
> "Cycle Time in Faban refers to the inter-arrival time between subsequent
> requests that arrive at a server. For internet-based applications where the
> user population is unknown, it makes sense to model cycle times (rather
> than think times).  Note that there is a subtle difference between using
> think times and cycle times. In a load test that uses think times, if the
> server slows down and response times increase, the think times are still
> generated in the same manner giving the server time to recover. But in a
> test that uses cycle times, if the response time is large, the cycle time
> still remains the same - the emulated user simply sleeps for a shorter time
> between requests (and in the worst case where the response time is larger
> than the cycle time, will not sleep at all) causing the load on the server
> to increase. This will quickly result in excessive load on the server
> causing increasingly larger response times.  This is not necessarily a bad
> thing - it can pin-point scalability problems in the server more quickly
> and help you fix them."
>
> If I want to achieve a certain throughput level, configuring the ThinkTime
> is not useful, since the actual throughput depends on the response times. I
> now implemented an own PoissonTimer that times the cycle time, not the
> think time.
> (The implementation extends the standard PoissonTimer, and uses code
> fragments of the ConstantThroughputTimer.)
>
> My question:
> From my point of view, it would be definitely useful to provide at least
> all RandomTimers with an option "control think time or cycle time?". The
> implementation is rather straight-forward, and I really would like to help
> with that. What do you think?
>
> Note: Using the constant throughput timer is not an option, because the
> cycle times are then constant (see implementation of
> ConstantThroughputTimer), which is not really realistic/representative.
>
> Kind Regards,
> Fabian
>

Re: FW: Cannot assign requested address: JVM_Bind exception

[Adrian Speteanu]

The error is specific to the environment, not the tool. It has tried to
open a port already in use. Most likely cause: you are out of sockets, or
file descriptors needed to open more sockets. You will have to investigate
that the OS and/or other services/application are not restricting your
process.

BTW, I can't see the images so I can't be 100% sure (the list doesn't
support attachments, they need to be uploaded to hosting sites). So also
make sure that your http sampler configuration is correct.

Cheers,
Adrian S



On Fri, Jun 21, 2013 at 12:26 PM, Ankita Nair wrote:

>  Hi,
>
> ** **
>
> I am trying IP spoofing with JMeter. I have a text file with IP addresses
> to be used and they are replaced at runtime as in the following image. ***
> *
>
>  
>
> But when I run this, it gives me the following error.
>
> *org.apache.commons.httpclient.HttpMethodDirector: I/O exception
> (java.net.BindException) caught when processing request: Cannot assign
> requested address: JVM_Bind*
>
> ** **
>
> JMeter version:  2.9
>
> Implementation Type: HttpClient3.1
>
> ** **
>
> Can anyone please tell how can I solve this error.
>
> ** **
>
> Regards,
>
> Ankita
>

Re: Sharing logic between thread groups

[Adrian Speteanu]

You can add a header manager to use the same header for all your requests.
Use the header manager in a parent of all the samples that need to be in
its scope. For those requests that need to over-write it, use the HM as
child to those samplers.

For the general question: can logic be shared between threads? It has been
discussed recently, check those discussions first, maybe you find some
useful tips there, or share with us what those threads don't cover.

Cheers,
Adrian

On Thu, Jun 13, 2013 at 3:54 PM, Drach, Noa  wrote:

> Hello all,
>
> In my test I have 5 thread groups - I was wondering if it is possible to
> share common logic between the thread groups.
>
> Basically all of my tests have user id and organization id in the headers.
> For each thread that is running I am randomly selecting a user id in the
> range of 1-50 and organization id in the range of 1-100.
>
> All this is achieved by using a Bean Shell Preprocessor that is attached
> to each of the samplers.
>
> For each of the samplers this random editing of the header might also be
> accompanied by some editing of the url as a result of the random user ID.
>
> By attaching a preprocessor to each sample I manage to do all this - but I
> was wondering if there is some way to define some utility methods that I
> can call from each Bean Shell processor and avoid repeating the logic.
>
> For example:
> ...
> HeaderManager hm = sampler.getHeaderManager();
> callMySetHeaders(hm);
> // Do some specific code
>
> I know that it is possible to have custom functions - but my understanding
> is that they only return a string - I am asking if there is a way to write
> more complex code.
>
> Thanks,
> Noa Drach
>
>

Re: How to install and use Jmeter plugins?

[Adrian Speteanu]

And its advisable to use their site, wiki for instructions. They even have
their own user mailing list.

Adrian

On Thu, Jun 13, 2013 at 10:19 AM, Ronny Roscher  wrote:

> There is a README inside the zip archive you've downloaded and there is
> also a detailed installation instruction on the page where you got the
> file...
>
> Just extract the files to your JMeter/lib/ext folder
>
> Ronny
>
> 2013/6/13 Amit Kumar 
>
> > Dear All:
> >
> > Could you please guide me how in JMeter plugins installation? I
> downloaded
> > a plugin - 'Server hits per second'  but I know how to install and use
> it.
> >
> > --
> > Thanks and Regards,
> > Amit
> >
>

Re: Measuring page load / rendering time

[Adrian Speteanu]

Hi,

I have a different approach to this. But there's a lot of background to it,
which can't be covered answering a specific question (how to measure X),
all of it resumes to: you should not look for shortcuts and instead should
do things the right way. Measuring rendering times is the complete opposite
of doing things that way. Its a dead-end, because it is too hard to track
and fully cover. Are you going to test on a large enough number of
PC/Mac/Linux hardware configurations in conjunction with a large number of
software versions (OS, browsers, other plugins that might affect
rendering)? Is your test matrix going to be comprehensive enough? Usually
its not.

The approach to front-end should be different because UI has different
specific problems. I use YSlow!, a plugin for Firebug that works on
Firefox. It shows "missing optimisations", and gives a good starting point
for a development team to obtain the best rendering time for their project.
With JMeter, you create the load on the server side and with one desktop
machine, you evaluate what will be the most probable user experience during
high traffic and then improve that. Its the best thing you can do, and the
only honest approach to this problem. You can still make measurement taking
several samples from tools like Firebug, Chrome's dev tools and so on, but
what's the point? Are you trying to benchmark the renderer or your server
application? If its the second, there are more questions you ask.

Regards,
Adrian S




On Thu, Jun 6, 2013 at 3:28 PM, nmq  wrote:

> Hi everyone
>
> I have been told that JMeter does not measure page load or rendering time.
> Does anyone know of a roundabout way of making approximations using JMeter,
> which would be fairly close to actual times.
>
> Or if anyone knows of a better tool that can be used to achieve this?
>
> The AUT is a secured web portal giving access to a limited number of users
> and is document intensive. I need to measure the page load time of the
> Documents page which displays the first 100 documents and as the user
> scrolls down, renders the next 100 and so forth.
>
> Any tips or help for load/performance testing would be appreciated.
>
> Thanks
> Sam
>

Re: How to acheive load

[Adrian Speteanu]

Hi,

Not being able to achieve target throughput is not always related to things
that are wrong on the test client (considering that 25 rps seems rather low
), but rather with applications deployed in a state that is not ready to
handle the generated traffic. The advise below is very good, so that you
eliminate common/obvious mistakes client-side. But I also suggest you keep
an eye for possible application & setup specific problems.

Cheers,
Adrian S

On Thu, May 23, 2013 at 12:30 AM, Shmuel Krakower wrote:

> Things you may try:
> 1. Remove the Throughput controller and see what happens.
> 2. Add more threads and see what happens.
> 3. Is your load machine is overkilled (CPU at 100%, etc...)?
> 4. If all above leaves you with same limit of about 25MPS it might be the
> target service which cannot handle anymore load. Try:
> a. Running at the same time, the same load from another machine and see if
> you get to 50MPS in total or still you get about 25...
> b. Look into the service under load and try to figure out what is the
> bottleneck
> c. Maybe something else between the two machines which runs JMeter and the
> Service is limiting you? (i.e. internet connection limits)
>
>
>
> Shmuel Krakower.
> www.Beatsoo.org - re-use your jmeter scripts for application performance
> monitoring from worldwide locations for free.
>
>
> On Wed, May 22, 2013 at 6:48 PM, Sameera Rao P  >wrote:
>
> > Hi
> >
> > I Am using soap sample xml / RPC and trying to post ebxml messages. I am
> > not able to achieve my load of 100 Messages Per second running from GUI.
> > I tried using constant throughput timer with 15 threads ,but still not
> > able to go beyond 25MPS. Can you please suggest the solution for
> > achieving the load
> >
> > Thanks & Regards
> > Sam
> >
> >
> >
> Disclaimer:
> >  This message and the information contained herein is proprietary and
> > confidential and subject to the Tech Mahindra policy statement, you may
> > review the policy at http://www.techmahindra.com/Disclaimer.html
> > ">http://www.techmahindra.com/Disclaimer.html externally and  href="
> > http://tim.techmahindra.com/tim/disclaimer.html";>
> > http://tim.techmahindra.com/tim/disclaimer.html internally within
> > Tech
> >
> Mahindra.
> >
> > -
> > To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> > For additional commands, e-mail: user-h...@jmeter.apache.org
> >
> >
>

Re: How to access different test-fragment present in external test plan.

[Adrian Speteanu]

Hi,

http://jmeter.apache.org/usermanual/component_reference.html#Include_Controller

Cheers,
ASp

On Fri, May 17, 2013 at 2:54 AM, Rakesh Kumar  wrote:

> Hi,
> I want to create a common test plan and define multiple test fragment
> in this file. I want to use some of the specific test fragment in the
> specific test plan. I used include controller to include the external test
> plan component. When I use the Module controller it just shows the include
> controller in the list. It doesn't show test fragment present in the
> external test plan. Is there any way I can specifically use few of the test
> fragment present in the external test plan?
>
>
>
> Any kind of help is really appreciated.
>
>
>
>
>
>
> --
> Best Regards
> *Rakesh kumar*,
> Linkedin  |
> Blog
>

Re: Test is stopping by its own even before complete the test time?

[Adrian Speteanu]

Hi,

It usually happens if the scheduler has a time period specified which is
not large enough or if the thread loop count is not big enough to cover the
entire period (I usually set it to forever when using the scheduler or a
specific duration).

I might be missing something though.

Cheers,
Adrian S

On Mon, May 20, 2013 at 1:55 PM, Madhu Sekhar wrote:

> Hi Team,
>
> We are running a test with 5 users and 1 hour time span in threadgroup
> settings.
>
> We are seeing that the test is going to stop even it is not completed the
> test time.
>
> We ran the same test with 2 users but observed no such above issues
> occurred.
>
>
> Could any one please suggest us to clear the same and any reasons for the
> same?
>
> Thanks & Regards,
> Madhu.
>

Re: Huge database

[Adrian Speteanu]

Hi,

To answer 2.:
http://jmeter.apache.org/usermanual/listeners.html
http://jmeter.apache.org/usermanual/component_reference.html#View_Results_Tree
the results tree has a response data tab that permits some very basic
parsing of the server's response, it is not meant to emulate the browser,
but help you debug tests. Do note that it is not recommended to use during
large tests because the listener is keeping requests & responses in memory.

As for question no. 1, it sounds more like what your study should evaluate
and decide. I think you might find useful information gathered here:
http://wiki.apache.org/jmeter/
Especially:
http://msdn.microsoft.com/en-us/library/bb924357.aspx
http://msdn.microsoft.com/en-us/library/bb924370.aspx
it would help you organise and decide what to test for.

Cheers and good luck!
--Adrian S

On Fri, May 10, 2013 at 1:28 PM, Newbie  wrote:

> Hi every one,
>
> I'm new to Jmeter and trying to use Jmeter for my thesis test. I have the
> following questions that I would be grateful if you could help me:
> I have some web applications that I want to feed them some data and see
> their reaction. I have a dataset that the data are in two forms:
> wget --post-data "var1=value1&var2=value2&..." http://.../page.jsp
> wget http://.../page.jsp?var1=value1&var2=value2&;...
>
> 1.The database is huge do I have to feed all the parameter values to the
> related pages or is there an easier way to do that?
>
> 2. I have heard that we can see the pages in Jmeter just as we able to see
> them in a browser. Is that right? How can this happen?
>
> Regards
>
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

Re: jmeter.properties cleanup

[Adrian Speteanu]

I've used this one, since JMeter 2.3, I think:
onload.expandtree

and its one of the few things I configure when installing a new version.
People who create large trees in scripts know why its needed. But I'm
probably in the minority and know the GUI keyboard shortcuts to achieve the
same, so I wouldn't miss it.


On Wed, May 8, 2013 at 11:59 PM, Philippe Mouawad <
philippe.moua...@gmail.com> wrote:

> Hello,
>
> jmeter.properties has grown with a lot of properties that maybe are not
> that useful.
> I find it a good thing that lot of things are configurable in JMeter but
> maybe it's too much and one of the issues is users may not find the really
> useful ones (recently for example with https.socket.protocols).
>
> I propose to remove the following:
>
>- jmeter.loggerpanel.display=false => It's so easy to just click it
>- jmeter.errorscounter.display=true => Why would someone not want this
>feature ?
>- jmeter.toolbar.display=true => Why would someone not want this cool
>feature ?
>- jmeter.toolbar => Will users really want to reorganize these icons ?
>- jmeter.toolbar.icons => Same as before
>- onload.expandtree => Current default behaviour seems fine no ?
>- jmeter.save.saveservice.autoflush => After some further thinking, why
>would users not need this one ? If JMeter crashes and some data is lost
> ,
>then there are big chances that the test was not that fine before the
> crash.
>
> I have doubts about those ones:
>
> # Netscape HTTP Cookie file
> cookies=cookies => What does it do ?
>
> We could try to remove them and if users want them, we would have some
> bugzilla request to get them back.
> --
> Regards
> Philippe
>

Re: Output from running Java in Beanshell | receiving only 'class ${classname}'

[Adrian Speteanu]

Hello,

No you shouldn't add every jar to the lib folder. Only those custom classes
that your code requires. And if those jars have similar dependencies of
their own. But other than this I can't be much of a help, can only tell you
that every time we encountered similar issues, we used the IDE to search
for the packages that we needed to build as jars and move them to JMeter's
lib. When this was done correctly, in beanshell, it worked just like it did
in eclipse, if not, we had to tediously debug it.

Adrian S

Re: Adding abandonment rate to my test plan help

[Adrian Speteanu]

People often reach pages, by following links, miss-clicking, trying to find
various resources, that don't relate to their needs. That also adds to
abandonment. That brings me back to Deepak's response. You need to know
their percentage and use that.

Adrian S.


On Fri, Apr 19, 2013 at 7:52 AM, Kirk Pepperdine
wrote:

> More often abandonment is related to response time. Getting to the
> response time of the previous request would allow you to compare that to
> abandonment distribution.
>
> Regards,
> Kirk
>
> On 2013-04-18, at 10:09 PM, Deepak Shetty  wrote:
>
> > Before attempting complicated solutions , does throughput controller not
> > work for you?
> > +Sampler1
> > +ThroughPutController (90 , percent executions , uncheck per user = 10%
> > abandon)
> > ++Sampler2
> > ++ThroughPutController (80 , percent executions , uncheck per user = 20%
> > more abandon at this step)
> > +++Sampler3
> > and so on
> >
> > regards
> > deepak
> >
> >
> > On Thu, Apr 18, 2013 at 12:55 PM, Marcelo Jara  >wrote:
> >
> >> Hmm. I can try that.
> >> Is there a way to stop the iteration in a beanshell post processor? I'm
> >> thinking of doing the following:
> >> 1) Have percentages set as user defined variablesa)
> AbandonAfterSearch
> >> = 10%   b) AbandonAfterAddingToCart = 10%   c) AbandonAfterCheckout =
> 20%
> >> These don't have to add up to 100% and are mutually exclusive.
> >> 2) Add a beanshell post processor in each sampler. In it, generate a
> >> random number and compare it to the Abandon rate  from step 1. If it's
> >> lower, then continue execution. Otherwise, stop the iteration.
> >>
> >> I read an old thread (
> >>
> http://jmeter.512774.n5.nabble.com/BeanShell-Assertion-Can-a-failed-assertion-force-the-next-iteration-of-a-loop-td533789.html
> )
> >> which says something like this may not work.
> >> Is this still the case?
> >>> From: aadrah...@bluestoneinternational.com
> >>> To: user@jmeter.apache.org
> >>> Subject: RE: Adding abandonment rate to my test plan help
> >>> Date: Thu, 18 Apr 2013 16:44:38 +
> >>>
> >>> One way you can do this is create a switch controller with 4 children
> as
> >> simple controllers. The first child (the first simple controller under
> the
> >> switch controller) contains all four steps of your test 1) Searching, 2)
> >> Adding to Cart, 3)Checking out 4) Purchase. The second child only
> contains
> >> steps 1) to 2). The third child contains steps 1) to 3). The fourth
> child
> >> contains only step 1). For the switch controller, set the switch value
> to a
> >> variable read from a CSV file. Have 100 entries on the file with numbers
> >> 0,1,2,3. Choose the proportion of 0,1,2,3's according to your needs and
> >> have jmeter recycle at EOF. This way you will have the exact percentage
> of
> >> users you need to abandon on each step. For instance, a 0 would mean the
> >> user completes all 4 steps, a 1 would mean the user abandons before
> >> Checking out, etc.
> >>>
> >>>
> >>>
> >>> Andreas Adrahtas - Analyst
> >>> Blue Stone International, LLC
> >>>
> >>> Mobile: +646-266-0238
> >>> www.bluestoneinternational.com
> >>>
> >>> E-MAIL CONFIDENTIALITY NOTICE: The information contained in this e-mail
> >> and any accompanying documents may contain information that is
> confidential
> >> or otherwise protected from disclosure. If you are not the intended
> >> recipient of this message, or if this message has been addressed to you
> in
> >> error, please immediately alert the sender by reply e-mail and delete
> this
> >> message, including any attachment. Any dissemination, distribution or
> other
> >> use of the contents of this message by anyone other than the intended
> >> recipient is strictly prohibited
> >>>
> >>>
> >>> -Original Message-
> >>> From: Marcelo Jara [mailto:marceloj...@hotmail.com]
> >>> Sent: Thursday, April 18, 2013 12:12 PM
> >>> To: user@jmeter.apache.org
> >>> Subject: Adding abandonment rate to my test plan help
> >>>
> >>> I have a test plan that includes a thread group with multiple http
> >> samplers in it. The flow is a user making a purchase on a web site. So
> it
> >> includes 1) Searching, 2) Adding to cart, 3) Checking out, 4) Purchase.
> >>> To be more realistic, I want to add abandonment rates. So I want to add
> >> something that would stop the iteration at either steps 1, 2, or 3. And
> >> this should be weighted so more people would abandon at step 3 vs step
> 1.
> >>> What's the best way to do this? Do I add an IF statement before each
> >> step and then based on a percentage, either perform the action or not?
> >>>
> >>> Thanks,
> >>> Marcelo
> >>>
> >>>
> >>> -
> >>> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> >>> For additional commands, e-mail: user-h...@jmeter.apache.org
> >>>
> >>
> >>
>
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additiona

Re: Output from running Java in Beanshell | receiving only 'class ${classname}'

[Adrian Speteanu]

Hi,

You can expect that if the dependencies of that code are imported into
JMeter's lib folder.

I've done that a few times and all errors were related to classpath and
libs missing. The problem with copy-pasting from an IDE appears when you
have to build the lib jars. If those libs have dependencies of their own,
and those are not in JMeter's classpath, it becomes difficult to debug. But
I don't know what to say about what you observed, it doesn't compute with
me.

Adrian S


On Fri, Apr 19, 2013 at 7:16 AM, stuntgirl  wrote:

> Hi All,
>
> I really hope you can help.
>
> I bashed out a piece of Java code to do what I need to do. In Eclipse, it
> runs like a dream.
> I copy and pasted the code from Eclipse into a Beanshell Sampler.
> However, everytime I run my JMeter script, all I get is:
>
> class ${classname} in the Response Data tab.
>
> I'm actually expecting some kind of output. For context, my Java should log
> into a GMail account, retrieve a specific email (based on subject), open
> it,
> and interpret the contents of the email.
>
> Can I expect the output from Eclipse to be the same in JMeter? Am I doing
> something wrong?
>
> Regards,
> stuntgirl
>
>
>
> --
> View this message in context:
> http://jmeter.512774.n5.nabble.com/Output-from-running-Java-in-Beanshell-receiving-only-class-classname-tp5716741.html
> Sent from the JMeter - User mailing list archive at Nabble.com.
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

Re: Win7 connection refused, address is invalid on local machine, -Djava.net.preferIPv4Stack=true does not help

[Adrian Speteanu]

In these situations, you should try using telnet.
telnet  
You use port in case its different than 80 (http default). Also its space
between ip and port, not a ":".

This will tell you if the connection should work in the first place. If it
doesn't, its not related to JMeter and you need to figure out why that
connection doesn't work (eventually talk to the IT if no one in the team
knows). If it works with telnet and doesn't in JMeter, we need to check
other things.

Adrian S


On Fri, Apr 12, 2013 at 1:15 PM, Manish Sapariya wrote:

> Even if I hardcode my local IP address in Source IP address, field,
> its failing. So I believe its not the case of wrong IP address being
> passed to the field.
>
> In fact there are some posts that discuss the same issue on Win8,
> but same fix, i.e. using -Djava.new.preferIPv4Stack=true,
> do not work on win7.
>
> Regards,
> Manish
>
>
> On Fri, Apr 12, 2013 at 3:39 PM, sebb  wrote:
>
> > On 12 April 2013 09:57, Manish Sapariya  wrote:
> >
> > > Thanks for reply Seb.
> > >
> > > It works fine if I do not use the Source IP address, as expected.
> > >
> > > jmeter version 2.9
> > >
> > > C:\>java -version
> > > java version "1.7.0_17"
> > > Java(TM) SE Runtime Environment (build 1.7.0_17-b02)
> > > Java HotSpot(TM) Client VM (build 23.7-b01, mixed mode, sharing)
> > >
> > > Sorry  I forgot to note that, I tried following things but the results
> > are
> > > same.
> > >  - I tried with Java 1.6 64 bit
> > >  - Java 1.7 32bit and 64 bit
> > >  - apache jmeter 2.6
> > >
> > >
> > > Same test plan works on my XP and Linux box
> > > with Source IP address.
> > >
> > >
> > In which case - are you sure that the source address really is valid on
> > your Win 7 box?
> > If JMeter works correctly on some hosts and not on others, then the
> primary
> > suspect must be the host config.
> >
> >
> > > Regards,
> > > Manish
> > >
> > >
> > >
> > >
> > > On Fri, Apr 12, 2013 at 2:16 PM, sebb  wrote:
> > >
> > > > On 12 April 2013 09:30, Manish Sapariya  wrote:
> > > >
> > > > > Hi All,
> > > > > I am using source IP address in my HTTP sampler and I get following
> > > > > exception.
> > > > >
> > > >
> > > > What happens if you don't use the source IP address?
> > > >
> > > > Version of JMeter?
> > > >
> > > > Java?
> > > >
> > > >
> > > > > Some of the posts suggest to use -Djava.net.preferIPv4Stack=true,
> but
> > > > that
> > > > > did not help fixing this.
> > > > >
> > > > > I also disabled the IPv6 Stack (so do I believe), by un-checking
> the
> > > IPv6
> > > > > in
> > > > > network connection properties dialog box for all network
> connections.
> > > > >
> > > > > Any pointer would be helpful.
> > > > >
> > > > > Here is the exact trace I get in the response.
> > > > >
> > > > > The test plan is very simple with one HTTP request and Src Ip
> address
> > > set
> > > > > to
> > > > > my local IP address.
> > > > >
> > > > > org.apache.http.conn.HttpHostConnectException: Connection to
> > > > > http://example.com refused
> > > > > at
> > > > >
> > > > >
> > > >
> > >
> >
> org.apache.http.impl.conn.DefaultClientConnectionOperator.openConnection(DefaultClientConnectionOperator.java:190)
> > > > > at
> > > > >
> > > > >
> > > >
> > >
> >
> org.apache.http.impl.conn.ManagedClientConnectionImpl.open(ManagedClientConnectionImpl.java:294)
> > > > > at
> > > > >
> > > > >
> > > >
> > >
> >
> org.apache.http.impl.client.DefaultRequestDirector.tryConnect(DefaultRequestDirector.java:645)
> > > > > at
> > > > >
> > > > >
> > > >
> > >
> >
> org.apache.http.impl.client.DefaultRequestDirector.execute(DefaultRequestDirector.java:480)
> > > > > at
> > > > >
> > > > >
> > > >
> > >
> >
> org.apache.http.impl.client.AbstractHttpClient.execute(AbstractHttpClient.java:906)
> > > > > at
> > > > >
> > > > >
> > > >
> > >
> >
> org.apache.http.impl.client.AbstractHttpClient.execute(AbstractHttpClient.java:805)
> > > > > at
> > > > >
> > > > >
> > > >
> > >
> >
> org.apache.jmeter.protocol.http.sampler.HTTPHC4Impl.sample(HTTPHC4Impl.java:286)
> > > > > at
> > > > >
> > > > >
> > > >
> > >
> >
> org.apache.jmeter.protocol.http.sampler.HTTPSamplerProxy.sample(HTTPSamplerProxy.java:62)
> > > > > at
> > > > >
> > > > >
> > > >
> > >
> >
> org.apache.jmeter.protocol.http.sampler.HTTPSamplerBase.sample(HTTPSamplerBase.java:1088)
> > > > > at
> > > > >
> > > > >
> > > >
> > >
> >
> org.apache.jmeter.protocol.http.sampler.HTTPSamplerBase.sample(HTTPSamplerBase.java:1077)
> > > > > at
> > > > >
> > > > >
> > > >
> > >
> >
> org.apache.jmeter.threads.JMeterThread.process_sampler(JMeterThread.java:428)
> > > > > at
> > > org.apache.jmeter.threads.JMeterThread.run(JMeterThread.java:256)
> > > > > at java.lang.Thread.run(Unknown Source)
> > > > > Caused by: java.net.ConnectException: connect: Address is invalid
> on
> > > > local
> > > > > machine, or port is not valid on remote machine
> > > > > at java.net.TwoStacksPlainSocketImpl.socketConnect(Native
> Method)

Re: Problem with Sample Times?

[Adrian Speteanu]

Thanks for the clarification. That explains some things.


On Fri, Apr 12, 2013 at 2:56 AM, sebb  wrote:

> On 11 April 2013 10:04, Adrian Speteanu  wrote:
>
> > Hi,
> >
> > Sample time/response time is the elapsed period from the moment the last
> > bit of the request was sent until the last bit from the response is
> > received. It measures each sample individually.
> >
>
> Sorry, but to be exact, JMeter starts measuring the time from just before
> the first byte is sent until the last byte is received.
>
>
> > Now, if your test setup is correct (the test machine isn't over-used, CPU
> > is within normal limits), the response time JMeter reports = application
> > response time + network latency. No other pre-processing &
> post-processing
> > can affect this.
> >
>
> Agreed.
>
>
> > The rest are only statistical information which summarise all individual
> > samples.
> >
> > Regards,
> > Adrian S
> >
> >
> > On Wed, Apr 10, 2013 at 6:31 PM, JPeter  wrote:
> >
> > > Hi,
> > >
> > > im testing an oracle application on my server. Via LAN.
> > > The application has some reports of tables.(HTTP Sites with a table)
> > >
> > > 1. More data in a table = Higher Sample Time...right?
> > > 2. Sample Time = Response Time?
> > > 3. For example in an /View Results in Table/ Listener, its/ Sample
> Time/.
> > > What are the results in an /Aggregate Graph/(Min,Max, etc.)?
> > > 4. Now my real problem(besides my English(Im sorry)) :
> > >
> > > I requested an http site of my application with a big table (900k "data
> > set
> > > entrys?") on it.
> > > No matter how many threads i start,
> > > i get some Sample Times above 20-30 Seconds (3 ms) and sometimes
> even
> > > my
> > > server "crashes" and is not available for about 1-2 minutes from
> then...
> > > (I was pretty happy about that because i thought that its just too much
> > for
> > > my server etc and i have the result i wanted...nearly)
> > > An hour later the same Request has Sample Times about 40ms and
> everything
> > > works perfect.()
> > > I tried it a lot of times and i dont have any idea.
> > > Empty Sites always have about 40ms. A site with a smaller table has
> > > sometimes about 300ms (in all following testruns too ) and an hour
> later
> > it
> > > has also 40ms...
> > > im really sorry my questions might be a bit confusing but I AM even
> more
> > > confused.
> > > I dont even know if its a problem with my JMeter or with my
> > Server/Oracle.
> > > Maybe someone had the same issues...
> > >
> > > Thanks a lot
> > > JPeter
> > >
> > >
> > >
> > >
> > >
> > > --
> > > View this message in context:
> > >
> >
> http://jmeter.512774.n5.nabble.com/Problem-with-Sample-Times-tp5716652.html
> > > Sent from the JMeter - User mailing list archive at Nabble.com.
> > >
> > > -
> > > To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> > > For additional commands, e-mail: user-h...@jmeter.apache.org
> > >
> > >
> >
>

Re: Problem with Sample Times?

[Adrian Speteanu]

Hi,

Sample time/response time is the elapsed period from the moment the last
bit of the request was sent until the last bit from the response is
received. It measures each sample individually.

Now, if your test setup is correct (the test machine isn't over-used, CPU
is within normal limits), the response time JMeter reports = application
response time + network latency. No other pre-processing & post-processing
can affect this.

The rest are only statistical information which summarise all individual
samples.

Regards,
Adrian S


On Wed, Apr 10, 2013 at 6:31 PM, JPeter  wrote:

> Hi,
>
> im testing an oracle application on my server. Via LAN.
> The application has some reports of tables.(HTTP Sites with a table)
>
> 1. More data in a table = Higher Sample Time...right?
> 2. Sample Time = Response Time?
> 3. For example in an /View Results in Table/ Listener, its/ Sample Time/.
> What are the results in an /Aggregate Graph/(Min,Max, etc.)?
> 4. Now my real problem(besides my English(Im sorry)) :
>
> I requested an http site of my application with a big table (900k "data set
> entrys?") on it.
> No matter how many threads i start,
> i get some Sample Times above 20-30 Seconds (3 ms) and sometimes even
> my
> server "crashes" and is not available for about 1-2 minutes from then...
> (I was pretty happy about that because i thought that its just too much for
> my server etc and i have the result i wanted...nearly)
> An hour later the same Request has Sample Times about 40ms and everything
> works perfect.()
> I tried it a lot of times and i dont have any idea.
> Empty Sites always have about 40ms. A site with a smaller table has
> sometimes about 300ms (in all following testruns too ) and an hour later it
> has also 40ms...
> im really sorry my questions might be a bit confusing but I AM even more
> confused.
> I dont even know if its a problem with my JMeter or with my Server/Oracle.
> Maybe someone had the same issues...
>
> Thanks a lot
> JPeter
>
>
>
>
>
> --
> View this message in context:
> http://jmeter.512774.n5.nabble.com/Problem-with-Sample-Times-tp5716652.html
> Sent from the JMeter - User mailing list archive at Nabble.com.
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

Re: Will jmeter pick up error 324?

[Adrian Speteanu]

Hi,

If you make a minimum validation on that sampler, it will pick up any
scenario that its not expected. Is 3xx an error code? Have no idea about
324 but the ones I've worked with are actually expected behaviour, so test
shouldn't fail. 4xx/5xx are handled correctly/automatically. WIth response
code validation, you can even tell your test plan to accept 4xx situations
as valid in a particular test (just check for the expected code and check
ignore status in the GUI of the Response Assertion).

So my assumption is that you need to add some validation in the test (i.e.
if you were expecting something in content and had an assertion, it should
have failed when response was null).

Cheers,
Adrian S


On Tue, Apr 9, 2013 at 10:27 AM, Mike Franon  wrote:

> We ran though a series of jmeter distributed tests against our web
> application on AWS, and it passed  under extreme heavy loads.
>
> When we went live we found we were getting error 324 empty err
> response on some requests. Our fix was to change our keepalive on the
> apache servers and work with Amazon and their ELB to match our
> keepalive,  Has anyone had a similar issue with jmeter?  Will it pick
> up 324 as an error or does it consider it a 302?
>
> Thanks
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

Re: when is a sample not a sample?

[Adrian Speteanu]

Hi,

Listeners apply to their parent and all samples that are childs from their
parent level down. This is the only way to reduce the scope of a listener:
to add it in a different controller which contains only the samplers you
want to monitor. It is not always possible to do because you still want to
respect a certain sequence of samples sometimes. So when this happens, but
I still want to exclude some of the requests, I add different listeners as
child to every sampler I am interested in.

The only selective conditions available to listeners is success/failure. So
if you fail a request that doesn't meet a condition you can choose to
monitor only those that meet your expectations - but that means increasing
the error rate artificially (which is not a problem, if you only monitor
the successes anyway). Its not elegant, I don't recommend it.

Maybe some plugin does the job, but never searched for it, as I never
needed this scenario to work before.

Cheers,
Adrian S



On Tue, Apr 9, 2013 at 9:25 AM, Stott, Charlie  wrote:

> Hi,
>
> I am wondering if and how it is possible to exclude the results of a
> specific sample (or at least disregard it) from the results.
>
> We have a dual stripe system which can be tested in online or offline
> mode.  To automate the decision we can request a status page, and act on
> the results.  Not all our test systems have dual stripes, so the request
> fails.  However, this is a valid situation.
>
> I always want to lookup the status page, but only if it exists and only if
> it responds a certain way do I want it to affect the test run.
>
> I am finding the situation where it is not available is being recorded as
> a test failure.
>
> Effectively, I want to perform the request as a pre-processor of another
> sample, but it seems overkill to write it in java/beanshell/someotherlang
> code.  Is there anything like a HTTP Request pre-processor, or a flag I can
> set somewhere that says ignore this sample on error?
>
> Thanks in advance,
> Charlie
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

Re: GUI getting slower?

[Adrian Speteanu]

Hi,


If you are willing to contribute with tests & results of your own, or if
you find a scenario that doesn't work, its safe to say that everyone in the
list is interested in reading about it.

Cheers.





On Wed, Apr 3, 2013 at 4:04 PM, Kostas Papadopoulos <
kos...@methodosit.com.cy> wrote:

>
> Hi Philippe,
>
> Thanks for the response. We have already changed our test practice to
> non-GUI
> testing so it is not an issue for us. But if anyone needs to investigate
> this
> further I'm willing to help.
>
> --
> Best Regards
>
> Kostas Papadopoulos
>
> K.E MethodosIT Ltd
>
> Tel: +357 22463610
> Fax: +357 22515197
> Website: http://www.methodosit.com.cy
>
>
> On Wed 03 Apr 2013 15:46:12 Philippe Mouawad wrote:
> > Hello,
> > @Kostas, there is a clear recommandation to do test in NON GUI mode.
> > http://jmeter.apache.org/usermanual/best-practices.html
> >
> > Since 2.5 there has been a lot of fixes on somes GUI issues related to
> > synchronization problems, they may have incurred some performance
> > degradations although I didn't notice them on the few tests I did. There
> > has been a Bugzilla opened a while ago and tests didn't show it clearly
> >
> > See:
> >
> >- http://jmeter.apache.org/changes.html
> >- http://jmeter.apache.org/changes_history.html
> >
> > @Kirk,
> > Since 2.5 there has been quite a lot improvements on Memory and
> > Performances.
> > Among this:
> >
> >- fixes on memory leaks
> >- fixes on memory usage by Regexp Extractor which consumed the double
> >of a response size
> >- fixes in core
> >
> > In 2.9, we introduced a performance improvement that may have increased
> > usage of memory which in summary caches the response so that it is not
> > converted/parsed as many times as there are post processors.
> >
> > My tests showed an important performance gain if post-processors were
> used
> > which in my opinion is the majority of cases.
> >
> >
> > So at first sight I doubt there is performance degradation between 2.5
> and
> > 2.9 but I may be wrong.
> >
> > Don' t hesitate to open an issue with a sample test plan and as many
> > informations as you have
> >
> >
> > Regards
> >
> > Philippe M.
> >
> > On Wed, Apr 3, 2013 at 1:44 PM, Kirk Pepperdine
> >
> > wrote:
> > > I've noted some significant differences in performance which I'll be
> > > investigating over the next few days. The first thing I've noted is a
> > > significant increase in memory requirements.
> > >
> > > -- Kirk
> > >
> > > On 2013-04-03, at 4:48 AM, Kostas Papadopoulos
> > > 
> > >
> > > wrote:
> > > > We have been using JMeter version 2.5 to do some performance testing.
> > > > I recently installed JMeter version 2.9 and I found that the test
> > > > results
> > >
> > > for
> > >
> > > > the same tests are very different when using the GUI. In one
> instance,
> > >
> > > the
> > >
> > > > same test gives throughput 4500/sec with version 2.5 and 3800/sec
> with
> > >
> > > 2.9.
> > >
> > > > Running non-gui on both versions gives comparable results (within
> 2%).
> > >
> > > I'm
> > >
> > > > running these tests on the same exact environment for both version
> > > > (same machines, application to be tested, java version, database,
> > > > data, etc.)
> > > >
> > > > Is this to be expected? Is the recommended practise to run only
> > > > non-gui?
> > >
> > > Or
> > >
> > > > is it something to be looked at more carefully?
> > > >
> > > > Please excuse me if this has this been addressed before.
> > > >
> > > > --
> > > > Best Regards
> > > >
> > > > Kostas Papadopoulos
> > > >
> > > > K.E MethodosIT Ltd
> > > >
> > > > Tel: +357 22463610
> > > > Fax: +357 22515197
> > > > Website: http://www.methodosit.com.cy
> > > >
> > > >
> > > > -
> > > > To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> > > > For additional commands, e-mail: user-h...@jmeter.apache.org
> > >
> > > -
> > > To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> > > For additional commands, e-mail: user-h...@jmeter.apache.org
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

Re: recording content-encoding header

[Adrian Speteanu]

Hi,

Although that sounds weird (I'm unable to reproduce this), you can always
add it in the test later using HTTP Header Manager configuration element.


Cheers,
Adrian S


On Wed, Apr 3, 2013 at 1:05 AM, Mark Lybarger  wrote:

> i'm trying to record some traffic, and the content-encoding header seems to
> get stripped out. my application is expecting gzip content encoding and i
> see that header when i don't go through the jmeter proxy, but when i turn
> it on, there's no content-encoding header present.
>
> -mark-
>

Re: Testing application operations cause by HTTP Request

[Adrian Speteanu]

Hi,

So let's get this straight: you want JMeter to receive requests from the
server.

In case that is it, there was a similar thread quite recently. I don't
remember how that ended. Out of the box, I don't think its supported
behaviour. You should also check the feature request lists, if its a
behaviour that will become more and more common.


---

On a sidenote, if you don't mind coding, you can add any functionality to
JMeter, because it is extendible.

Regards,
Adrian S


On Sat, Mar 30, 2013 at 7:54 PM, Jeffrey Williams <
jeffxor.willi...@gmail.com> wrote:

> I have previously used JMeter to load test web applications. My current
> application is a little but different to a normal web application. It is
> more of processing and routing application.
>
> Essentially it takes a HTTP request payload does some transformation and
> then puts the transformed message onto a JMS queue. Then another piece of
> the application will take this message and send it out from the application
> via a HTTP request.
>
> What I am attempting to do is one of the following test the response from
> the first HTTP response (I have that already) and then verify the second
> HTTP Request coming from the machine is correct.
>
> From what I have read of the JMeter documentation and also googling I can
> not see a way to configure JMeter to do this. Plus I am a little confused
> on how to tie the inbound request to the outbound request.
>
> Is this possible with JMeter?
> Would/Should I attempt to do this JMeter?
> Is there a better way?
>

Re: Jmeter stuck with bad application error management

[Adrian Speteanu]

Hi,

My understanding is that this stack trace is applicable to those
times/situations when you use stop test instead of shutdown. Jmeter cuts
ongoing connections and this error is logged. This error is also logged if,
for whatever reasons, the connection is cut on server side or all
midle-ware components (from routers to firewalls and balancers).

Now, if the problem is that the application doesn't respond for a long
time, jmeter shouldn't close that connection frankly. Tweak the server
side. The situations where you want to keep connection up an indefinite
period of time are very rare.

But all modern browsers timeout connections after 1-2 min. You can use the
connect / response timeouts fields from the HTTP Sampler to achieve this in
JMeter as well.

Log the timeout errors separately from other errors. These will increase in
frequency as soon as you start using the timeouts on JMeter side. A high
percentage should be a good motivator for the dev/infrastructure teams to
start looking for a solution. But I have to say, most of the times
timeouted connections mean poor configuration, not necessarily poor
application, so you should look into it ASAP. There is little reason to
continue testing until these problems are eliminated, one by one, or their
effect is reduced as much as possible. Your end-users will get this
behaviour as well. I doubt that 2 min of waiting (and for nothing) is
desirable for them.


Regards,
Adrian S



On Mon, Mar 25, 2013 at 3:08 PM, Sergio Boso wrote:

> Hi I'm experiencing a weird behaviour  in Jmeter.
>
> As the application has its own burden of problems, it returns sockets
> error fro time to time.
> The point is that Jmeter, instead of recording the error and going on the
> next sample, gets stuck there forever (no time out is applied).
> I can stop it manually, reset it and restart it.
>
> Here is the error stack after manually stopping.
> have you ever experienced such a behaviour?
> Is this already known?
>
> Java the error stack follows
>
> Thank you in advance
>
> Sergio
>
> Java.net.SocketException: socket closed
> at java.net.SocketInputStream.**socketRead0(Native Method)
> at java.net.SocketInputStream.**read(Unknown Source)
> at java.net.SocketInputStream.**read(Unknown Source)
> at sun.security.ssl.InputRecord.**readFully(Unknown Source)
> at sun.security.ssl.InputRecord.**read(Unknown Source)
> at sun.security.ssl.**SSLSocketImpl.readRecord(**Unknown Source)
> at sun.security.ssl.**SSLSocketImpl.readDataRecord(**Unknown Source)
> at sun.security.ssl.**AppInputStream.read(Unknown Source)
> at org.apache.http.impl.io.**AbstractSessionInputBuffer.**fillBuffer(*
> *AbstractSessionInputBuffer.**java:166)
> at org.apache.http.impl.io.**SocketInputBuffer.fillBuffer(**
> SocketInputBuffer.java:90)
> at org.apache.http.impl.io.**AbstractSessionInputBuffer.**readLine(**
> AbstractSessionInputBuffer.**java:281)
> at org.apache.http.impl.io.**ChunkedInputStream.**getChunkSize(**
> ChunkedInputStream.java:251)
> at org.apache.http.impl.io.**ChunkedInputStream.nextChunk(**
> ChunkedInputStream.java:209)
> at org.apache.http.impl.io.**ChunkedInputStream.read(**
> ChunkedInputStream.java:171)
> at org.apache.http.impl.io.**ChunkedInputStream.read(**
> ChunkedInputStream.java:201)
> at org.apache.http.conn.**EofSensorInputStream.read(**
> EofSensorInputStream.java:155)
> at org.apache.jmeter.protocol.**http.sampler.HTTPSamplerBase.**
> readResponse(HTTPSamplerBase.**java:1649)
> at org.apache.jmeter.protocol.**http.sampler.HTTPAbstractImpl.**
> readResponse(HTTPAbstractImpl.**java:235)
> at org.apache.jmeter.protocol.**http.sampler.HTTPHC4Impl.**
> sample(HTTPHC4Impl.java:300)
> at org.apache.jmeter.protocol.**http.sampler.HTTPSamplerProxy.**
> sample(HTTPSamplerProxy.java:**62)
> at org.apache.jmeter.protocol.**http.sampler.HTTPSamplerBase.**
> downloadPageResources(**HTTPSamplerBase.java:1207)
> at org.apache.jmeter.protocol.**http.sampler.HTTPSamplerBase.**
> resultProcessing(**HTTPSamplerBase.java:1500)
> at org.apache.jmeter.protocol.**http.sampler.HTTPAbstractImpl.**
> resultProcessing(**HTTPAbstractImpl.java:251)
> at org.apache.jmeter.protocol.**http.sampler.HTTPHC4Impl.**
> sample(HTTPHC4Impl.java:358)
> at org.apache.jmeter.protocol.**http.sampler.HTTPSamplerProxy.**
> sample(HTTPSamplerProxy.java:**62)
> at org.apache.jmeter.protocol.**http.sampler.HTTPSamplerBase.**
> sample(HTTPSamplerBase.java:**1088)
> at org.apache.jmeter.protocol.**http.sampler.HTTPSamplerBase.**
> sample(HTTPSamplerBase.java:**1077)
> at org.apache.jmeter.threads.**JMeterThread.process_sampler(**
> JMeterThread.java:428)
> at org.apache.jmeter.threads.**JMeterThread.run(JMeterThread.**
> java:256)
> at java.lang.Thread.run(Unknown Source)
>
> --
>
> Ing. Sergio Boso
>
>
>
>
> --**--**-
> To unsubscribe, e-mail: 
> use

Re: JMeter Plugin with Sonar

[Adrian Speteanu]

Oops, I shouldn't have assumed you're running it in/with Jenkins. I'm stuck
then also. Maybe the people from Sonar can help you.

Regards,
Adrian S

On Thu, Mar 21, 2013 at 8:59 AM, Niraj  wrote:

> Nope. I have setup Sonar and provided .JTL file in JMeter configurations.
>
> On Wed, Mar 20, 2013 at 11:01 PM, Adrian Speteanu  >wrote:
>
> > You do have Jenkins with Sonar installed? and have you followed the
> > instructions regarding the Report Server?
> >
> > It should work just like any other plugin if everything is in place. Of
> > course I can't be of much help with Sonar itself if its not properly
> setup
> > in Jenkins, since in my case it was already installed and used by the
> > development team for other measurements.
> >
> > Regards,
> > Adrian S
> >
> > On Wed, Mar 20, 2013 at 7:11 PM, Niraj  wrote:
> >
> > > Do I need to use Maven to get this work?
> > > On Mar 20, 2013 6:08 PM, "Niraj"  wrote:
> > >
> > > > I am sorry about not adding complete details.
> > > >
> > > > Steps which i have followed:
> > > >
> > > > Downloaded Sonar JMeter plugin and copied to
> > > > C:\sonar-3.5\extensions\plugins .
> > > > I ran the JMeter test and got the .jtl file.
> > > > Added path of .jtl file to Settings->JMeter->Remote Server host
> (Remote
> > > > jmeter report server host, used for getting remote results if no
> local
> > > jtl
> > > > file is defined.)
> > > > Restarted Sonar.
> > > > But i am not able see anything under projects.
> > > >
> > > > Am i missing something?
> > > >
> > > > On Wed, Mar 20, 2013 at 5:56 PM,  wrote:
> > > >
> > > >> Niraj  wrote on 03/20/2013 07:23:52 AM:
> > > >>
> > > >> > From: Niraj 
> > > >> > To: JMeter Users List 
> > > >> > Date: 03/20/2013 07:24 AM
> > > >> > Subject: JMeter Plugin with Sonar
> > > >> >
> > > >> > Hello All,
> > > >> >
> > > >> > Anyone tried JMeter plugin with Sonar?
> > > >> >
> > > >> > http://docs.codehaus.org/display/SONAR/JMeter+Plugin
> > > >> >
> > > >> > I am facing some issues. Some one please help me on this.
> > > >> >
> > > >>
> > > >> If you want help, you'll need to describe the problem, describe what
> > you
> > > >> tried to fix it etc.
> > > >>
> > > >> http://www.catb.org/esr/faqs/smart-questions.html
> > > >>
> > > >> > Thanks,
> > > >> > Niraj
> > > >> >
> > > >> > --
> > > >> > This message has been scanned for viruses and dangerous content,
> > > >> > and is believed to be clean.
> > > >> >   Message id: 5CED4600A39.A91D9
> > > >> >
> > > >> >
> > > >>
> > > >> This communication and any attachments are confidential, protected
> by
> > > >> Communications Privacy Act 18 USCS § 2510, solely for the use of the
> > > >> intended recipient, and may contain legally privileged material. If
> > you
> > > are
> > > >> not the intended recipient, please return or destroy it immediately.
> > > Thank
> > > >> you.
> > > >>
> > > >>
> -
> > > >> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> > > >> For additional commands, e-mail: user-h...@jmeter.apache.org
> > > >>
> > > >>
> > > >
> > >
> >
>

Re: JMeter Plugin with Sonar

[Adrian Speteanu]

You do have Jenkins with Sonar installed? and have you followed the
instructions regarding the Report Server?

It should work just like any other plugin if everything is in place. Of
course I can't be of much help with Sonar itself if its not properly setup
in Jenkins, since in my case it was already installed and used by the
development team for other measurements.

Regards,
Adrian S

On Wed, Mar 20, 2013 at 7:11 PM, Niraj  wrote:

> Do I need to use Maven to get this work?
> On Mar 20, 2013 6:08 PM, "Niraj"  wrote:
>
> > I am sorry about not adding complete details.
> >
> > Steps which i have followed:
> >
> > Downloaded Sonar JMeter plugin and copied to
> > C:\sonar-3.5\extensions\plugins .
> > I ran the JMeter test and got the .jtl file.
> > Added path of .jtl file to Settings->JMeter->Remote Server host (Remote
> > jmeter report server host, used for getting remote results if no local
> jtl
> > file is defined.)
> > Restarted Sonar.
> > But i am not able see anything under projects.
> >
> > Am i missing something?
> >
> > On Wed, Mar 20, 2013 at 5:56 PM,  wrote:
> >
> >> Niraj  wrote on 03/20/2013 07:23:52 AM:
> >>
> >> > From: Niraj 
> >> > To: JMeter Users List 
> >> > Date: 03/20/2013 07:24 AM
> >> > Subject: JMeter Plugin with Sonar
> >> >
> >> > Hello All,
> >> >
> >> > Anyone tried JMeter plugin with Sonar?
> >> >
> >> > http://docs.codehaus.org/display/SONAR/JMeter+Plugin
> >> >
> >> > I am facing some issues. Some one please help me on this.
> >> >
> >>
> >> If you want help, you'll need to describe the problem, describe what you
> >> tried to fix it etc.
> >>
> >> http://www.catb.org/esr/faqs/smart-questions.html
> >>
> >> > Thanks,
> >> > Niraj
> >> >
> >> > --
> >> > This message has been scanned for viruses and dangerous content,
> >> > and is believed to be clean.
> >> >   Message id: 5CED4600A39.A91D9
> >> >
> >> >
> >>
> >> This communication and any attachments are confidential, protected by
> >> Communications Privacy Act 18 USCS § 2510, solely for the use of the
> >> intended recipient, and may contain legally privileged material. If you
> are
> >> not the intended recipient, please return or destroy it immediately.
> Thank
> >> you.
> >>
> >> -
> >> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> >> For additional commands, e-mail: user-h...@jmeter.apache.org
> >>
> >>
> >
>

Re: About BeanShell

[Adrian Speteanu]

+1 for last part...

Its a great method to setup your script file in the same element with all
the other static variables (so you can make all changes in a single place.)

On Wed, Mar 20, 2013 at 2:34 AM, Stott, Charlie  wrote:

> Hi Jakob,
>
> > * suppose I created a set of functions in an external script file - how
> to call
> > these from my script
>
> a) Read the file into a variable, then 'run' the variable using
> __javaScript function.
>
> Or
>
> b) Use the beanshell pre-processor and enter the filename in the 'Script
> file' field.
>
> > * for this same file: when running in GUI-mode, how to ensure the JMeter
> > has seen these function definitions?
>
> Non-GUI uses the same test plan, not sure what problem you envisage here?
>
> > * Section 16.8 shortly has some 'best practices', where it is mentioned
> that I
> > can 'save a script as a variable' - what does that mean? How do I achieve
> > that? My first guess was to copy/paste the function definition into the
> 'Value'
> > column of a User Defined Variable in the Test Plan, but somehow that
> > completely confuses the interpreter
>
> Almost..
>
> For example, using User Defined Variables:
>
> Name:   Value:
> SCRIPT_VAR  if ("${SOME_VAR" != "") {
> vars.put("SOME_VAR", "value");}
>
> Then in another User Defined Variables element (scope means it won't run
> in the same element):
>
> Name:   Value:
> RUN_SCRIPT  ${__javaScript(${SCRIPT_VAR})}
>
> I tend to set the scripts early in my test plan, and then invoke them
> where necessary.
>
> Hope that helps.
> Charlie
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

Re: What is the preferred method to assert multiple headers?

[Adrian Speteanu]

On Mon, Mar 11, 2013 at 8:07 PM, sebb  wrote:

> On 11 March 2013 17:56, Adrian Speteanu  wrote:
> > For content, if your response assertion contains:
> >   text1
> >   text3
> >   text2
> >
> > on a different line each of the strings, but the placement of this
> strings
> > in the content is text1...text2...text3, then the assertion will fail.
> > Their order matters. If they're in different elements it won't matter. If
> > your app changes layout often, this brakes functional assertions
> > inevitably. I thought this might apply to header assertion also.
> >
>
> I still don't follow, but perhaps I am missing something.
>
you're not. I was wrong.

>
> If the assertion contains 3 separate pattern entries, then they will
> each match (or not) individually against the chosen target.
> At least that is how it is supposed to work - the patterns are independent.
> If you have a counter example, please raise a Bugzilla issue for it.
>
that is the actual behaviour. I was under the impression that the patterns
to test are asserted in order, maybe it was older behaviour or I just
remembered incorrectly. But, I tested and it works as you say.

please disregard this and forgive my bluntness.

>
> If there is a single entry containing all the text strings separated
> by new-lines, then of course it matters which order is used.
> I don't see how using separate assertions helps here.
>
> >
> > On Mon, Mar 11, 2013 at 7:48 PM, sebb  wrote:
> >
> >> On 11 March 2013 17:07, Adrian Speteanu  wrote:
> >> > Hi,
> >> >
> >> > I choose maintainability in these cases. Whatever is easier to you to
> >> > update and maintain is good enough.
> >> >
> >> > In the case of response headers, I'd keep different elements for each
> >> > assertion, because I have no clue whether they will always be
> returned in
> >> > the order that I have entered them in the assertion, when writing the
> >> test
> >> > script.
> >>
> >> Huh?
> >>
> >> The Response Assertion checks all its pattern entries.
> >> The order does not matter.
> >>
> >> > Cheers,
> >> > Adrian Sp
> >> >
> >> > On Mon, Mar 11, 2013 at 5:33 PM, Jakob van Bethlehem <
> >> ja...@jet-stream.nl>wrote:
> >> >
> >> >> Dear users,
> >> >>
> >> >> Currently I'm working on building some header assertions (using
> Response
> >> >> Assertion) on HTTP Requests. There is a set of 6 headers that our
> >> >> application must send, which I'd like to add to the Response
> Assertion,
> >> and
> >> >> I was wondering what is the preferred method to achieve this in
> JMeter:
> >> >> (1) create a single, huge pattern that has all headers (this can be
> >> done,
> >> >> because all 6 headers will come out in a given order), or
> >> >> (2) create a separate pattern for each header
> >> >>
> >> >> I don't know for sure whether the order in which headers are received
> >> >> could ever be mixed up (maybe when the server is stressed?). If that
> is
> >> the
> >> >> case, I guess method (2) would be preferable. If that is not the
> case,
> >> >> method (1) would result in a single test to run, which may be faster
> >> than
> >> >> method (2), although it is a huge test. I'm lacking experience here -
> >> >> hopefully someone can give some pointers.
> >> >>
> >> >> Sincerely,
> >> >> Jakob van Bethlehem
> >> >> -
> >> >> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> >> >> For additional commands, e-mail: user-h...@jmeter.apache.org
> >> >>
> >> >>
> >>
> >> -
> >> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> >> For additional commands, e-mail: user-h...@jmeter.apache.org
> >>
> >>
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

Re: What is the preferred method to assert multiple headers?

[Adrian Speteanu]

For content, if your response assertion contains:
  text1
  text3
  text2

on a different line each of the strings, but the placement of this strings
in the content is text1...text2...text3, then the assertion will fail.
Their order matters. If they're in different elements it won't matter. If
your app changes layout often, this brakes functional assertions
inevitably. I thought this might apply to header assertion also.



On Mon, Mar 11, 2013 at 7:48 PM, sebb  wrote:

> On 11 March 2013 17:07, Adrian Speteanu  wrote:
> > Hi,
> >
> > I choose maintainability in these cases. Whatever is easier to you to
> > update and maintain is good enough.
> >
> > In the case of response headers, I'd keep different elements for each
> > assertion, because I have no clue whether they will always be returned in
> > the order that I have entered them in the assertion, when writing the
> test
> > script.
>
> Huh?
>
> The Response Assertion checks all its pattern entries.
> The order does not matter.
>
> > Cheers,
> > Adrian Sp
> >
> > On Mon, Mar 11, 2013 at 5:33 PM, Jakob van Bethlehem <
> ja...@jet-stream.nl>wrote:
> >
> >> Dear users,
> >>
> >> Currently I'm working on building some header assertions (using Response
> >> Assertion) on HTTP Requests. There is a set of 6 headers that our
> >> application must send, which I'd like to add to the Response Assertion,
> and
> >> I was wondering what is the preferred method to achieve this in JMeter:
> >> (1) create a single, huge pattern that has all headers (this can be
> done,
> >> because all 6 headers will come out in a given order), or
> >> (2) create a separate pattern for each header
> >>
> >> I don't know for sure whether the order in which headers are received
> >> could ever be mixed up (maybe when the server is stressed?). If that is
> the
> >> case, I guess method (2) would be preferable. If that is not the case,
> >> method (1) would result in a single test to run, which may be faster
> than
> >> method (2), although it is a huge test. I'm lacking experience here -
> >> hopefully someone can give some pointers.
> >>
> >> Sincerely,
> >> Jakob van Bethlehem
> >> -
> >> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> >> For additional commands, e-mail: user-h...@jmeter.apache.org
> >>
> >>
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

Re: What is the preferred method to assert multiple headers?

[Adrian Speteanu]

Hi,

I choose maintainability in these cases. Whatever is easier to you to
update and maintain is good enough.

In the case of response headers, I'd keep different elements for each
assertion, because I have no clue whether they will always be returned in
the order that I have entered them in the assertion, when writing the test
script.

Cheers,
Adrian Sp

On Mon, Mar 11, 2013 at 5:33 PM, Jakob van Bethlehem wrote:

> Dear users,
>
> Currently I'm working on building some header assertions (using Response
> Assertion) on HTTP Requests. There is a set of 6 headers that our
> application must send, which I'd like to add to the Response Assertion, and
> I was wondering what is the preferred method to achieve this in JMeter:
> (1) create a single, huge pattern that has all headers (this can be done,
> because all 6 headers will come out in a given order), or
> (2) create a separate pattern for each header
>
> I don't know for sure whether the order in which headers are received
> could ever be mixed up (maybe when the server is stressed?). If that is the
> case, I guess method (2) would be preferable. If that is not the case,
> method (1) would result in a single test to run, which may be faster than
> method (2), although it is a huge test. I'm lacking experience here -
> hopefully someone can give some pointers.
>
> Sincerely,
> Jakob van Bethlehem
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

Re: Persist JSESSIONID throughout a TestPlan

[Adrian Speteanu]

This question has been asked before, but I don't remember a time it has
been solved.

You'll want to create a variable in one of the thread group, such as
"variable_" and then all other threads will be able to
reference it. Of course, you'll need to make thread groups with an equal
number of threads. Creating the variable is easy, but making it accessible
to the other threads was not possible in older version (? maybe I'm
mistaken about this, but I looked to implement this particular use-case 3
years ago and found no solution). However, with the versions released in
the last couple of years, you have the beanshell server that can help with
this. I haven't tested it yet, its in my to do list, but I think someone
explained how it can be used to update a variable from within multiple
threads, the same principle should apply in your use case. Also, there is a
plugin:
http://code.google.com/p/jmeter-plugins/wiki/InterThreadCommunication, that
might do the job, but again, haven't had the time to use it myself.

So you're use-case is doable, but you should be double-careful :) and test
that what you think you are doing is actually what is happening. With
thread sharing and concurrency you can easily run into weird situations
(i'm thinking other values than expected, but memory leaks aren't excluded
either).

Good luck with this.

Also, please be kind and let us know what solution you used in the end. I
think you should appreciate how useful this contribution might be for
someone else.

Thanks,
--Adrian S

On Mon, Mar 11, 2013 at 1:48 PM, Alaka P A  wrote:

> I have to test the webserver which has to maintain sessions.
> For this purpose I need to maintain session ids in jmeter.
> It has been working for ThreadGroup as I
> am using HTTP Cookie Manager for every Thread Group.
> But i want to use the same
> JSESSIONID across the test plan which includes several Thread Groups.
> Not able to achieve this.
> I tried using HTTP Cookie Manager globally for a Test Plan,
> But it didnt work.
> I am using jmeter 2.5 version. Can anybody help me with this?
> Thanks !
>
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

Re: I/O exception (java.net.SocketException) caught when processing request: Connection reset

[Adrian Speteanu]

Hi,

These errors are not from JMeter, but from the OS's sockets. The root cause
is related to network problems (bad configuration, poor equipment, firewall
not coping with the volume of requests are the top on my mind, but there
are other causes too; overloading is sometimes an issue, but since you
don't have too much traffic, its out of the question here).

You'll need to investigate this problem - see if your test setup is correct
and then talk to the people responsible for the network and see what can be
done about it.

Regards,
Adrian

On Fri, Mar 8, 2013 at 7:50 AM, nick  wrote:

>
>
> Hi I am seeing a lot of errors in while running the load as just 100
> threads
>
> we are using
>
> java version "1.6.0_26"
> Java(TM) SE Runtime Environment (build 1.6.0_26-b03)
> Java HotSpot(TM) 64-Bit Server VM (build 20.1-b02, mixed mode)
>
> &
>
> Jmeter 2.7
>
>
>
> 2013/03/07 20:52:01 INFO  -
> org.apache.jmeter.protocol.http.sampler.HTTPHC4Impl$4: I/O exception
> (java.net.SocketException) caught when processing request: Connection reset
> 2013/03/07 20:52:01 INFO  -
> org.apache.jmeter.protocol.http.sampler.HTTPHC4Impl$4: Retrying request
> 2013/03/07 20:52:03 INFO  -
> org.apache.jmeter.protocol.http.sampler.HTTPHC4Impl$4: I/O exception
> (java.net.SocketException) caught when processing request: Connection reset
> 2013/03/07 20:52:03 INFO  -
> org.apache.jmeter.protocol.http.sampler.HTTPHC4Impl$4: Retrying request
> 2013/03/07 20:52:05 INFO  -
> org.apache.jmeter.protocol.http.sampler.HTTPHC4Impl$4: I/O exception
> (java.net.SocketException) caught when processing request: Connection reset
> 2013/03/07 20:52:05 INFO  -
> org.apache.jmeter.protocol.http.sampler.HTTPHC4Impl$4: Retrying request
>
>
> Any help or pointer would be greatly appreciated
>
>
> Thank you
>
> Nick
> .
>
>
>
>
>
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

Re: Is Jmeter Scalable?

[Adrian Speteanu]

Hi,

6000 users is easy to do.

For insanely larger tests (how about 1 million calls per second?), you
might find that other tools scale better. But only slightly, the difference
is given maybe not by the tool itself, as it is given by the technology
behind it. And those tools that I know of and seen in use, they aren't as
extensible as JMeter. As it uses Java and  there are so many systems
working on Java, you can imagine its pretty performant.

Mind you that all large tests come with quirks. So do the tools you use, no
matter which they are. If your approach is correct, JMeter will scale both
vertically and horizontally... You don't really need clouding services to
do it if you have the patience to take the hard approach, but if you don't
have much time at your disposal, they might prove useful.

Regards,
Adrian Speteanu

On Tue, Mar 5, 2013 at 2:16 PM, Sarndeep Nijjar <
sarndeepnij...@googlemail.com> wrote:

> Thanks Guys,
>
> So are we safe to assume that scaling up jmeter tests will give us
> accurate results, and we will not actually end up testing the jvm that
> jmeter runs on?
>
>
>
> On 03/03/13 20:16, Philippe Mouawad wrote:
>
>> Hello,
>> Yes it is scalable provided you use it well as any other performance tool.
>>
>> We use it in our Test Campaigns and reach frequently 6000 VUs (distributed
>> testing) without reaching limits on Injector machines.
>> Performances are getting better and better with each new version.
>>
>> For best practices read:
>>
>> - 
>> http://jmeter.apache.org/**usermanual/best-practices.html<http://jmeter.apache.org/usermanual/best-practices.html>
>> - http://www.ubik-ingenierie.**com/blog/jmeter_performance_**
>> tuning_tips/<http://www.ubik-ingenierie.com/blog/jmeter_performance_tuning_tips/>
>> - Component reference notes on Performance
>>
>>
>>
>
> --**--**-
> To unsubscribe, e-mail: 
> user-unsubscribe@jmeter.**apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

Re: Couple of Queries. Please confirm. 1. Threads supported by single machine at a time? 2. What does it mean if warning appears in results?

[Adrian Speteanu]

Just for fun to see if its possible and how much I can squeeze out of it.

Somebody else created a similar test (with a little less test logic than
mine) using wrk, https://github.com/wg/wrk. I wanted to compare the
throughput that can be generated with jmeter, over that generated with more
lightweight tools.

The discussion is a little more complex than this. Its the result of
experimenting over the last two years with setups that give the best
results in the case of tests that need to generate huge amounts of dynamic
test data.

Also, I've avoided the distributed mode setup because the master is a
single point of failure (i'm not 100% sure), while having independent
jmeter instances is more flexible (if you prefer using the shell). But
considering the improvements made in 2.9 (or was it 2.8?) to it, I will
have to reconsider that in the future.

Adrian S

On Mon, Feb 25, 2013 at 3:16 PM, Oliver Lloyd wrote:

> Hi Adrian, out of interest, what made you opt for vertical scaling using
> one machine? Did you try scaling horizontally using distributed mode (or
> even just running isolated instances)?
>
>
> On 25 Feb 2013, at 12:57, Adrian Speteanu  wrote:
>
> > Sorry about that, pressed send by mistake, I updated bellow the rest of
> the
> > message.
> >
> > On Mon, Feb 25, 2013 at 2:52 PM, Adrian Speteanu  >wrote:
> >
> >> Hi.
> >>
> >> The two questions are incompatible. And also number 2. creates problems
> >> with number 1. Please read on the subject, it was already answered in
> the
> >> mailing list before.
> >>
> >> As for #1, let me give an example of a test I'm running at the moment
> (but
> >> its not standard and it depends on the setup and this particular test):
> >> * test system:
> >>  - machine used to generate traffic: dual processor (6 physical
> cores
> >> each, 24 threads total) with 64Gb RAM, 15k RPM disks, running CentOS
> linux
> >>  - jmeter 2.9 (also used 2.8 with comparable results)
> >> * test requirements:
> >>  - generate 30 thousands requests per second
> >>
> > In order to generate this much traffic with one single machine, I have
> > tested multiple setups and configs and eventually used 10-12 instances of
> > JMeter, each spawning 3500 threads. The throughput is limited per
> instance
> > using the Constant throughput controller, so it reaches the desired
> level.
> > This setup is required in order to generate a stable and predictable
> > throughput from JMeter.
> >
> > Obviously, monitoring client side (JMeter) is inexistent (apart from
> > summary in console), unless I suspect fowl play. Monitoring is done at
> > application level. JMeter just gives me the load the app under test
> > requires. In order to add monitoring from JMeter, I have to write a jtl,
> > which significantly reduces throughput that I can generate, so when I do
> > use it, only one out of 10-12 instances writes to disk.
> >
> > Cheers,
> > Adrian
> >
> >>
> >>
> >> On Sat, Feb 23, 2013 at 7:52 AM, Amit Kumar 
> wrote:
> >>
> >>> Dear All:
> >>>
> >>> I have couple of simple queries, please confirm:
> >>>
> >>> 1. How much threads are supported by a single machine at a time (I
> heard
> >>> 300, right?)?
> >>> 2. In Listeners(Table/Tree): Some requests appear with warning(red
> >>> symbol), what does it mean? Does it mean that those requests are
> failed?
> >>> [Please refer screen shot].
> >>>
> >>> --
> >>> Thanks and Regards,
> >>> Amit
> >>>
> >>>
> >>> -
> >>> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> >>> For additional commands, e-mail: user-h...@jmeter.apache.org
> >>>
> >>
> >>
>
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

Re: Couple of Queries. Please confirm. 1. Threads supported by single machine at a time? 2. What does it mean if warning appears in results?

[Adrian Speteanu]

Sorry about that, pressed send by mistake, I updated bellow the rest of the
message.

On Mon, Feb 25, 2013 at 2:52 PM, Adrian Speteanu wrote:

> Hi.
>
> The two questions are incompatible. And also number 2. creates problems
> with number 1. Please read on the subject, it was already answered in the
> mailing list before.
>
> As for #1, let me give an example of a test I'm running at the moment (but
> its not standard and it depends on the setup and this particular test):
>  * test system:
>   - machine used to generate traffic: dual processor (6 physical cores
> each, 24 threads total) with 64Gb RAM, 15k RPM disks, running CentOS linux
>   - jmeter 2.9 (also used 2.8 with comparable results)
>  * test requirements:
>   - generate 30 thousands requests per second
>
In order to generate this much traffic with one single machine, I have
tested multiple setups and configs and eventually used 10-12 instances of
JMeter, each spawning 3500 threads. The throughput is limited per instance
using the Constant throughput controller, so it reaches the desired level.
This setup is required in order to generate a stable and predictable
throughput from JMeter.

Obviously, monitoring client side (JMeter) is inexistent (apart from
summary in console), unless I suspect fowl play. Monitoring is done at
application level. JMeter just gives me the load the app under test
requires. In order to add monitoring from JMeter, I have to write a jtl,
which significantly reduces throughput that I can generate, so when I do
use it, only one out of 10-12 instances writes to disk.

Cheers,
Adrian

>
>
> On Sat, Feb 23, 2013 at 7:52 AM, Amit Kumar  wrote:
>
>> Dear All:
>>
>> I have couple of simple queries, please confirm:
>>
>> 1. How much threads are supported by a single machine at a time (I heard
>> 300, right?)?
>> 2. In Listeners(Table/Tree): Some requests appear with warning(red
>> symbol), what does it mean? Does it mean that those requests are failed?
>> [Please refer screen shot].
>>
>> --
>> Thanks and Regards,
>> Amit
>>
>>
>> -
>> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
>> For additional commands, e-mail: user-h...@jmeter.apache.org
>>
>
>

Re: Couple of Queries. Please confirm. 1. Threads supported by single machine at a time? 2. What does it mean if warning appears in results?

[Adrian Speteanu]

Hi.

The two questions are incompatible. And also number 2. creates problems
with number 1. Please read on the subject, it was already answered in the
mailing list before.

As for #1, let me give an example of a test I'm running at the moment (but
its not standard and it depends on the setup and this particular test):
 * test system:
  - machine used to generate traffic: dual processor (6 physical cores
each, 24 threads total) with 64Gb RAM, 15k RPM disks, running CentOS linux
  - jmeter 2.9 (also used 2.8 with comparable results)
 * test requirements:
  - generate 30 thousands requests per second


On Sat, Feb 23, 2013 at 7:52 AM, Amit Kumar  wrote:

> Dear All:
>
> I have couple of simple queries, please confirm:
>
> 1. How much threads are supported by a single machine at a time (I heard
> 300, right?)?
> 2. In Listeners(Table/Tree): Some requests appear with warning(red
> symbol), what does it mean? Does it mean that those requests are failed?
> [Please refer screen shot].
>
> --
> Thanks and Regards,
> Amit
>
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>

Re: Test Environment for Performance Testing

[Adrian Speteanu]

Hello,

I've tried before to compile a list of the issues that I always run into
while setting up the test environment. But the list is too long.

If you get machines from Amazon, without any particular service, you can
easily deploy jmeter there yourself if you have minimum Linux skills.
Depending on what machine you will acquire you'll be able to generate more
or less traffic. But 1000 users you can simulate with a single dual core
desktop station if you run the test in non-GUI mode.

Be aware that if you move tests to Amazon, your application needs to be
available to external traffic already.

Adrian

On Mon, Feb 18, 2013 at 9:59 AM, Anand Kumar Sharma <
mail4anandsha...@gmail.com> wrote:

> Hi Mahesh
>
> Thanks Mahesh fro your valuable suggestion. I would like to know more about
> Cloud based Load testing
>
> I search for Amazon load testing and found a link http://gcload.com but
> this link is not working.
>
> Can you suggest any FREE Cloud based Load Testing tool which will test for
> maximum 1000 concurrent user
>
> On Mon, Feb 18, 2013 at 12:26 PM, మహేష్™  wrote:
>
> > Anand,
> >
> > This is always depends on what type of application you are testing.
> > If your application is internal application/product then you can test the
> > same in the LAN else if it is exposed to internet like web site then
> > you have to test the same from WAN.
> >
> > Based on the application behavior we have to test. If the application is
> > internet application and if you have a user base from
> > different geographic locations then better to simulate the user load
> > from different locations (Amazon cloud can be useful here).
> >
> > Cheers,
> > Mahesh
> >
> > On Mon, Feb 18, 2013 at 12:19 PM, Anand Kumar Sharma <
> > mail4anandsha...@gmail.com> wrote:
> >
> > > Hi all
> > >
> > >
> > >
> > > Can anyone let me know how to create proper Testing environment  which
> > will
> > > determine better testing results. Wheter I should go for testing
> > > application on LAN or some external environment
> > >
> > >
> > > --
> > > This mail has came from desk of.
> > > Anand Kumar Sharma (आनंद कुमार शर्मा)
> > >
> >
>
>
>
> --
> This mail has came from desk of.
> Anand Kumar Sharma (आनंद कुमार शर्मा)
>

Re: I am beginner !! Please help me on- How to record HTTPS requests using proxy server in JMeter?

[Adrian Speteanu]

Hi,

By the way, Google Authentication system, that gives you access to gmail is
a complex app and unfriendly to testers (at least the live one does). Its
not a good demo tool as it will give you a lot of headaches, before you get
to the https stuff you want to practice on. I'd recommend starting with
more basic stuff.

Just saying, but except for the certificate thing, it should work the same
as http.

Adrian S

On Mon, Feb 4, 2013 at 5:37 PM, Amit Kumar  wrote:

> Hi Dear:
>
> Here is what I tried:
>
> I recorded a script using Proxy Server. I used JMeter 2.9.
> Test Case: Login to Gmal and Logout.
>
> After recording, it shown only couple of requests as below.
>
> [image: Inline image 1]
> Does this information help?
>
>
> On Mon, Feb 4, 2013 at 8:27 PM, Mark Miller wrote:
>
>> You're going to need to provide a better question with more information
>> and
>> evidence of what you have tried and where you are having difficulty.
>>
>>
>> Mark
>>
>>
>> On Mon, Feb 4, 2013 at 7:23 AM, Amit Kumar  wrote:
>>
>> > Dear All:
>> >
>> > Hope you all are doing great !!
>> >
>> > I am beginner !!
>> > Please help me on- How to record HTTPS requests using proxy server in
>> > JMeter?
>> > Is there any other way?
>> > I tried it in JMeter 2.4 and 2.9 as well. But it didn't work.
>> > Please help !!
>> >
>> > Thanks in Advanced !!
>> > --
>> > Thanks and Regards,
>> > Amit
>> >
>>
>
>
>
> --
> Thanks and Regards,
> Amit
>

Re: Testing asynchronous webservices

[Adrian Speteanu]

Hi,

This applies in general (unfortunately):
If you want to have some smart way of doing the same stuff your client app
does, then you'll have to extend JMeter, which is rather easy, especially
if you already have the jars that implement the logic you need. Otherwise,
once you've configured the web-service sampler properly, the issues of
testing web-services are the same as in the case of any other apps.

Except: "AWS callback with info -> APP", I wouldn't know how to test this
in JMeter without extending it. Its not a behaviour I've encountered
before. The timeout works for responses to outgoing calls, but we're not
there yet with this requirement.

And because you made me curious, I googled a bit, found this:
http://www.codeaffine.com/2011/11/28/stressload-testing-of-asynchronous-httprest-services-with-jmeter/Apparently
they didn't had that problem, their challenge was similar to
what I expected, client initiated transactions and large response times in
some cases (the blog has some interesting ideas for you). Make sure you
really have to wait for a callback initiated by the server first (very few
systems use it). But if it does do that, would be nice of you to let us
know what solution you guys eventually picked.

Adrian S

On Tue, Jan 29, 2013 at 4:19 PM, Stefan Pivoda  wrote:

> Hi All,
>
> Has anybody experience with testing asynchronous web-services with Jmeter?
>
> Could you please describe me how to configure and test:
>
> whether i receive a confirmation from the async web-service;
> after that how to set some daemon that will wait for the callback with
> the requested information from that async web-service;
> evaluate whether this received response from the callback.
>
> The communication between the APP and the async web-service (AWS) should
> look like this:
>
> APP request -> (AWS)
> AWS confirmation -> APP
> AWS callback with info -> APP
> APP confirmation -> AWS
>
> I dont want to set some specific long time-out and just wait for the
> callback, if it is not needed as it will be different in each case base on
> the method in the background that the async webservice will handle. that's
> the reason why i'd like to use some deamon or some more sophisticated way,
> if it is possible.
>
> thank you
> Stefan
> _
>
> Vsetko podstatne z vedy, pocitacov,  mobilov aj hier -
> http://www.TECHsme.sk
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

Re: How to open jmx file when there is a missing lib

[Adrian Speteanu]

I don't know of any such option. Sounds like a good feature request, even
if its small in priority IMO, its still a good approach to be able to load
only those elements that are ok (I think current behaviour is only needed
when core elements fail to load, not when its caused by some plugin).

Regards,
Adrian S

On Mon, Jan 14, 2013 at 4:40 PM, Cyril SANTUNE wrote:

> yes, I can find this. I delete the text between the hashtree tag. It
> works but I am just wondering if it's possible to do it in jmeter GUI
> (force the opening).
>
>
>
> On Mon, Jan 14, 2013 at 2:45 PM, Adrian Speteanu 
> wrote:
> > Hi,
> >
> > Can you find a reference to that lib by editing that testplan with a text
> > editor?
> >
> > I can't test this scenario at the moment, but you'll find something LIKE:
> >  testclass="HTTPSamplerProxy"
> > testname=" -04- check device" enabled="true">
> > ...
> >
> > where guiclass will reference something which is dependent of that lib.
> If
> > you remove the entire element which needs it, it should work. You will
> need
> > to make this edit in such a way to not affect the schema.
> >
> > Cheers,
> > Adrian S
> >
> >
> > On Mon, Jan 14, 2013 at 11:53 AM, Cyril SANTUNE  >wrote:
> >
> >> Hi,
> >>
> >> I have an old scenario/jmx file that use an external library. This
> >> library is not install so I can't open my scenario.
> >> In my case, I know this library is useless. I delete it in my jmx file
> >> with a text editor in xml but it is not very clean.
> >>
> >> is There a way to open a jmx file with missing library in jmeter to
> >> delete the unusable sampler ?
> >>
> >> Cyril
> >>
> >> -
> >> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> >> For additional commands, e-mail: user-h...@jmeter.apache.org
> >>
> >>
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

Re: JMeter Distributed Testing identifying bottlenecks

[Adrian Speteanu]

Hi,

All your questions are dependant on the particularities of your test and
application under test.
 1. It is not recommended to run a distributed test with many agents as
there are alternatives and they will all send the results to the master
which will create a too large amount of network traffic. If you still
choose this approach, at least read suggestions on how to best approach
such a test.

  2. there is no direct relation, it depends on your test and test machines
hardware. I wouldn't recommend running a distributed setup with that many
threads as you have better alternatives IMO.

 3.  both will work just fine on the same hardware. Its more dependant on
Java than OS capabilities at this point. use jmeter 2.8 as it fixes a
critical performance issue with large tests. As a personal preference, I
would choose Linux as it is easier to make scripts and configure multiple
Linux boxes than it is with Windows.

Cheers,
Adrian S

On Mon, Jan 14, 2013 at 3:31 PM, Ankita Nair wrote:

> Hi,
>
>
>
> Can anyone help me in the following scenarios.
>
> 1.   If we use 10 agent machines controlled by a single controller,
> does
> controller become a bottleneck in this case?
>
> 2.   If we have 10,000 users/threads to run a particular test-suite? If
> YES, how many agents were used?
>
> 3.   Which OS supports more number of active threads at a single point
> of time (Windows vs Linux).
>
>
>
> Regards,
>
> Ankita
>
>

Re: How to open jmx file when there is a missing lib

[Adrian Speteanu]

Hi,

Can you find a reference to that lib by editing that testplan with a text
editor?

I can't test this scenario at the moment, but you'll find something LIKE:

...

where guiclass will reference something which is dependent of that lib. If
you remove the entire element which needs it, it should work. You will need
to make this edit in such a way to not affect the schema.

Cheers,
Adrian S


On Mon, Jan 14, 2013 at 11:53 AM, Cyril SANTUNE wrote:

> Hi,
>
> I have an old scenario/jmx file that use an external library. This
> library is not install so I can't open my scenario.
> In my case, I know this library is useless. I delete it in my jmx file
> with a text editor in xml but it is not very clean.
>
> is There a way to open a jmx file with missing library in jmeter to
> delete the unusable sampler ?
>
> Cyril
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

Re: JMeter out of memory

[Adrian Speteanu]

That doesn't make any sense, honestly.

If you want an answer to your question, you need to repeat the test in the
conditions already mentioned to see if you don't have a real problem at
root to this. However if the "memory problem" is caused by you storing
thousands of requests in memory, than there is no problem and you need to
search for alternatives, as there are plenty to what you are doing right
now.

Regards,
Adrian

On Wed, Jan 9, 2013 at 11:04 AM, Srinivasa Tekkatte Shenoy <
tsrinivas.she...@mimos.my> wrote:

> I am already using distributed testing one master and 3 slaves.
>
> We need these two listeners for analysis.
>
> Thank You
>
> Regards
> Shenoy
>
> -Original Message-
> From: Niraj [mailto:niraj.khatm...@gmail.com]
> Sent: Wednesday, 9 January, 2013 5:02 PM
> To: JMeter Users List
> Cc: jmeter-u...@jakarta.apache.org
> Subject: Re: JMeter out of memory
>
>  View Results in Table and View Results in Tree listeners consumes lot's
> of memory. Remove this two listeners and re-run the test again.
>
> If you are still facing the same issue then use distributed testing.
>
> On Wed, Jan 9, 2013 at 2:24 PM, Srinivasa Tekkatte Shenoy <
> tsrinivas.she...@mimos.my> wrote:
>
> > Hi All,
> >
> > I am executing the JMeter for a script with 100 user threads for 2
> > hours duration and loop count set to Forever. Ramp up and ramp down in
> > of total
> > 100 users in one minute.
> >
> > The min and max Heap size of Jmeter.bat is set as 1GB. It cannot be
> > increase to more. Whether any other way to increase the heap size.
> >
> > We currently have listeners for Summary Report, View Results in Table
> > and View Results in Tree.
> >
> > I am getting out of memory error after 30minutes.  Could you please
> > let me know how to fix this.
> >
> > Thank You
> >
> > Regards
> > Shenoy
> >
> >
> >
> > --
> > -
> > -
> > DISCLAIMER:
> >
> > This e-mail (including any attachments) is for the addressee(s) only
> > and may contain confidential information. If you are not the intended
> > recipient, please note that any dealing, review, distribution,
> > printing, copying or use of this e-mail is strictly prohibited. If you
> > have received this email in error, please notify the sender
> > immediately and delete the original message.
> > MIMOS Berhad is a research and development institution under the
> > purview of the Malaysian Ministry of Science, Technology and
> > Innovation. Opinions, conclusions and other information in this e-
> > mail that do not relate to the official business of MIMOS Berhad
> > and/or its subsidiaries shall be understood as neither given nor
> > endorsed by MIMOS Berhad and/or its subsidiaries and neither MIMOS
> > Berhad nor its subsidiaries accepts responsibility for the same. All
> > liability arising from or in connection with computer viruses and/or
> > corrupted e-mails is excluded to the fullest extent permitted by law.
> >
> >
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

Re: CookieManager warnings

[Adrian Speteanu]

Hello,

Edit jmeter.properties (its in the /bin folder) and find the chapter on
"Logging levels".
Make sure that everything is set to ERROR instead of WARN and it will get
rid of the problem.

I wish I knew what that message is about to be more precise, but if nothing
is wrong / everything still works, then simply remove that from logs - its
not necessary.

Adrian S

On Fri, Dec 21, 2012 at 4:48 PM, Mitesh Patel wrote:

> Hi All,
>
> My logs are flooded with the following warnings:
>
> 2012/12/21 14:47:41 WARN  - jmeter.protocol.http.sampler.HTTPSamplerBase:
> Existing CookieManager HTTP Cookie Manager superseded by HTTP Cookie Manager
>
> Anybody come across this in the past and know how to resolve it?
>
> Thanks,
>
> Mitesh
>

Re: Theoretical question

[Adrian Speteanu]

Hi,

This sounds about right. The purpose is to simulate the amount of requests
and whatever method gets you there is the way to go. What is not clear to
me, on 5) - how will this ensure that not all users re-login? You'll also
have to check if you have a session id or not before making the login
request, right? In that case it seems more efficient your way.

Adrian S


On Thu, Dec 20, 2012 at 4:02 PM, Shmuel Krakower  wrote:

> Hi,
> Thanks for the collaboration.
>
> Didn't got your first comment regarding loops.
> Anyway I think it will be some kind of a merge between option 2 and my
> needs.
>
> This is what I'll do:
> 1. To promise specific throughput per each transaction, I have separate
> thread groups per each transaction/set of very few transactions which must
> go together (like login->view a message, or login->view a message->post
> reply).
> 2. Then I set a Constant Throughput Timer to get the needed traffic on a
> transaction level (based on production stats).
> 3. So like this - a user that logged in for posting replies will do that
> during the entire load test.
> 4. With your suggestion - to cause users to log out in some percent,
> another user will login after a while and post replies. So that's another
> benefit for me (having replies from many users instead of the same few
> which logged in at ramp up).
> 5. I don't need DB to save session, all sessions are cookie sessions, so
> what I'll do is just clearing the session cookie with a BSF sampler
> (instead of putting more load with sending requests to the application
> logout action, which will cause non realistic load in some manner).
>
> Thanks again,
>
>
> Shmuel Krakower.
> www.Beatsoo.org - re-use your jmeter scripts for application performance
> monitoring from worldwide locations for free.
>
>
> On Wed, Dec 19, 2012 at 5:19 PM, Adrian Speteanu  >wrote:
>
> > Hi,
> >
> > If you use thread loops, instead of using loop controllers inside the
> > thread, shouldn't that trigger more logins?
> > 
> > I approached this differently some time ago:
> >1. separate workload into two use-cases: with login and without login.
> > This gives you two different scripts or two different thread groups (I
> had
> > decent analytics, so I knew how much traffic each use-case should
> > generate)... I prefer using different scripts for performance reasons.
> >
> >2. instead of doing login once in a while, do logout once in a while
> ;)
> > This is a very common use case in real life. People will login if
> required
> > to accomplish whatever they need to do and can only do it if they are
> > signed in, but will very rarely logout. I used the throughput controller
> to
> > specify a number that represented a percentage of the total logins
> > performed. This will give you more than 500 sessions in the web
> > container...
> >
> >   3. for those threads that don't logout wright the session in a database
> > from JMeter, along with the timestamp when it was created. threads that
> > don't login from second script will use with a X probability a sessionID
> > stored in the database on their first request.
> > This covers those use-cases where your users return to the web page but
> > already have session. Make sure to detele old session IDs (this should
> > match the expiration configuration of the web container)...
> > 
> > This is also theoretical. I feel that 1) and 2) are enough, so test setup
> > is not too complicated -> you just have to monitor the number of sessions
> > (concurrent active and total sessions still valid) so they aren't too
> many
> > or too few. The focus is to have approximately the same amount of
> sessions
> > per web container and that the total number of logins in rapport to total
> > number of request matches production environment (or matches a realistic
> > number if you lack those analytics from production environments).
> >
> > Testing this as realistically as possible didn't give me more and
> improved
> > results, though, since tomcat was pretty well configured in my case and
> had
> > enough memory, while the database was a bigger bottleneck - so in real
> > life, you couldn't get into a situation where session management is
> > actually an issue. But that was true for that system. Nowadays, it
> > shouldn't be something you should worry about, unless you specifically
> > think that might be the problem.
> >
> > Cheers,
> > Adrian S
> >
> > On Wed, Dec 19, 2012 at 4:42 PM, Shmuel Krakower 
> > wrot

Re: Users list management

[Adrian Speteanu]

Sebb explained this some weeks ago that they can't do anything about it
because of some technical issue/limitation of the system. And that the only
thing we can do is delete those mails.

It is annoying.

Adrian S

On Thu, Dec 20, 2012 at 9:52 AM, Shay Ginsbourg wrote:

> Not just u.
> A bit annoying.
>  On Dec 20, 2012 9:41 AM, "Shmuel Krakower"  wrote:
>
> > Hi
> > I am not sure if that's only me but each time I send a mail to the users
> > list I get a delivery error to m...@home.mayanet.pl
> >
> > Is it only me or others get this error too?
> > If others get it, who can remove it from the users list?
> >
> > Best,
> > Shmuel.
> >
>

Re: Theoretical question

[Adrian Speteanu]

Hi,

If you use thread loops, instead of using loop controllers inside the
thread, shouldn't that trigger more logins?

I approached this differently some time ago:
   1. separate workload into two use-cases: with login and without login.
This gives you two different scripts or two different thread groups (I had
decent analytics, so I knew how much traffic each use-case should
generate)... I prefer using different scripts for performance reasons.

   2. instead of doing login once in a while, do logout once in a while ;)
This is a very common use case in real life. People will login if required
to accomplish whatever they need to do and can only do it if they are
signed in, but will very rarely logout. I used the throughput controller to
specify a number that represented a percentage of the total logins
performed. This will give you more than 500 sessions in the web container...

  3. for those threads that don't logout wright the session in a database
from JMeter, along with the timestamp when it was created. threads that
don't login from second script will use with a X probability a sessionID
stored in the database on their first request.
This covers those use-cases where your users return to the web page but
already have session. Make sure to detele old session IDs (this should
match the expiration configuration of the web container)...

This is also theoretical. I feel that 1) and 2) are enough, so test setup
is not too complicated -> you just have to monitor the number of sessions
(concurrent active and total sessions still valid) so they aren't too many
or too few. The focus is to have approximately the same amount of sessions
per web container and that the total number of logins in rapport to total
number of request matches production environment (or matches a realistic
number if you lack those analytics from production environments).

Testing this as realistically as possible didn't give me more and improved
results, though, since tomcat was pretty well configured in my case and had
enough memory, while the database was a bigger bottleneck - so in real
life, you couldn't get into a situation where session management is
actually an issue. But that was true for that system. Nowadays, it
shouldn't be something you should worry about, unless you specifically
think that might be the problem.

Cheers,
Adrian S

On Wed, Dec 19, 2012 at 4:42 PM, Shmuel Krakower  wrote:

> Hi,
> Not directly JMeter question, but I think that this community is the right
> place to ask.
>
> How do you implement your load test scenario (test plan) when simulating
> the Login requests?
> Until today, all of my load tests where built up with having the login
> action inside a Once Only Controller, which caused my load tests to
> simulate a login only once per user.
>
> The problems I see with my current approach are:
> 1. Logins are only executed during the initial "ramp up" phase of the load
> test and no login requests are handled by the app later.
> 2. Amount of sessions in the application is ramped up and then remains on
> the same level (i.e. 500 threads = 500 sessions) - which is ok, but
> actually it doesn't reflect real life.
> In real life - users are logging in, doing action or few and then may come
> back later with new session, so even when I have same throughput for most
> of the actions in the system as in the real life usage,
> I still get only 500 application sessions during load tests, while on
> production we have thousands of them.
>
> So again, my question is not about above problems, it is:
> How do you implement your load test scenario when simulating the Login
> requests?
> 1. Do you randomly re-login for some of the iterations?
> 2. Do you login for each iteration?
> 3. Other ideas?...
>
> Best,
>
> Shmuel Krakower.
> www.Beatsoo.org - re-use your jmeter scripts for application performance
> monitoring from worldwide locations for free.
>

Re: JMETER request processing in parallel

[Adrian Speteanu]

I support the idea. Recently I've studied the problem of parallelism vs
multi-threading and got to Locust (after getting multiple headaches, in
parallel too):
http://docs.locust.io/en/latest/what-is-locust.html

The idea is very simple: utilise those threads to the maximum. Which can be
done in JMeter as well if you have requests that respond in milliseconds or
tens of milliseconds (apps such as these usually don't depend on session
ID). So one thread can send tens - hundreds of requests per second. However
- its ugly! In the end, the person that implements the script shouldn't
have to worry how the threads are handled and organized - this needs to be
handled by a separate layer. Even  when testing the entire system, not just
a component - so you will have large response times and all, if your tester
implements a logic to the "VU" to wait 10-15s, do you really want to keep
an entire thread bounded to a wait() action? Its not efficient, not
[i]"pretty"[/i].

Because of the GUI, test logic usually tends to mimic virtual users actions
per thread - and no matter how much I advise against it, that's what people
do. And then they try to scale that logic to thousands of users per machine
and => ba da boom - Fatality!!!

Unfortunately, Locust is not mature enough and other tools just don't cut
it to that level. JMeter should grow, if possible, to cover these scenarios
as well.

Definitely will support whatever steps are made into this direction.

Cheers,
Adrian S

On Wed, Dec 19, 2012 at 4:14 PM, Shmuel Krakower  wrote:

> Hi Mitesh
> That's a good question - actually there is a request for a "Parallel
> Controller":
> https://issues.apache.org/bugzilla/show_bug.cgi?id=53159
>
> I may vote for it.
>
> Best,
>
> Shmuel Krakower.
> www.Beatsoo.org - re-use your jmeter scripts for application performance
> monitoring from worldwide locations for free.
>
>
>
> On Wed, Dec 19, 2012 at 4:08 PM, HUSSEY, SCOTT T  wrote:
>
> > You can concurrently request page resources. The setting is in the HTTP
> > Sampler and you can adjust the number of sub-threads each test plan
> thread
> > uses when requesting internal page resources.
> >
> > > -Original Message-
> > > From: Mitesh Patel [mailto:patel_m...@hotmail.com]
> > > Sent: Wednesday, December 19, 2012 7:49 AM
> > > To: user@jmeter.apache.org
> > > Subject: JMETER request processing in parallel
> > >
> > > Hi,
> > >
> > > This may be an obvious/stupid question, but can JMETER process "HTTP
> > > Request" elements in parallel? Currently it all requests (html, css,
> js,
> > images)
> > > are processed sequentially and I am getting high latencies. This is of
> > course
> > > not really the case, as the individual time for each request is less
> > than 1 sec.
> > >
> > > Thanks,
> > >
> >
> > -
> > To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> > For additional commands, e-mail: user-h...@jmeter.apache.org
> >
> >
>

Re: Help using Jmeter to upload files

[Adrian Speteanu]

Hi,

JMeter can test webservers (load/functional) using the common protocols on
the internet, most commonly used being HTTP. So if the app you're referring
to uses a method supported by JMeter, then YES you can use it to test your
application (your use case seems simple enough). To get a proper answer,
you have to understand how JMeter works and give us the proper information
to understand the issue.

For example in HTTP, you have to do a multi-part POST in order to send
multiple files in a single request. This usecase is supported by JMeter, I
use it on a constant basis and for tests that require dynamic files to be
sent, JMeter has became more flexible in the latest version, 2.8. But this
doesn't mean it won't be quite an effort to implement a large scale test
for this use-case. This should cover the application part of the problem,
unless you want to use the app's UI to make that test, in which case it
won't work, its not made for those kind of tests.

For the scan part, again you have to know what the application does in the
background. Scanning gives you a file, most probably, that is later sent
just like a normal file, so the test doesn't change that much - you just
have to figure out what's the format of those files and how to randomise
them during actual tests to obtain reliable results. If it uses different
protocols to send the information directly than those supported, then the
problem changes. But again, you have to know the specifics of how that is
accomplish in order to test this properly (refer to previous answers for
details).

--Adrian

On Tue, Dec 4, 2012 at 3:35 PM, Carolina Marques <
carolina.marq...@smlinfo.com.br> wrote:

> Hi Daniel,
>
> Thanks for your response. I think Jmeter doesn't do what we need. We did a
> lot of research, but we found nothing and no one knows how to help us to do
> this.
>
> The application is running through web services.
>
> Thanks for your help for now.
>
> Best regards,
>
> Carolina
>
> -Mensagem original-
> De: Daniel Corredor [mailto:dan...@gmail.com]
> Enviada em: terça-feira, 4 de dezembro de 2012 11:04
> Para: JMeter Users List
> Assunto: Re: Help using Jmeter to upload files
>
> Hi Carolina,
>
> Two scenarios:
>
> 1- I have read (feed) from static files. Not sure is that will be an option
> for you.
> 2- Once I had to upload a chemical structure file. I was unable to do that
> at the beginning but later after sniffing the whole transaction I was able
> to understand how the application decodes  the "special" file.
>
> How are you guys recording the test? How are you understanding "sniffing"
> the whole interaction from your application?
>
> Good luck,
> Dan
>
>
>
> On Mon, Dec 3, 2012 at 8:46 AM, Carolina Marques <
> carolina.marq...@smlinfo.com.br> wrote:
>
> > Hi Daniel,
> >
> > We are trying to upload any kind of file, since images, then txt
> > files, doc, audio files, video files, of different sizes.
> >
> > We did the Jmeter calls through the link of the document we want to view.
> > We
> > have into the environment a security service running too.
> >
> > We did all the analysis and this is still not working.
> >
> > Thanks a lot for your return and I hope you can help us, because we
> > really need this.
> >
> > Best regards,
> >
> > Carolina
> >
> > -Mensagem original-
> > De: Daniel Corredor [mailto:dan...@gmail.com] Enviada em:
> > segunda-feira, 3 de dezembro de 2012 11:40
> > Para: JMeter Users List
> > Assunto: Re: Help using Jmeter to upload files
> >
> > Hi Carlina,
> >
> > Did you analyze how the information is embedded in the application
> > once is uploaded?
> >
> > What kind of file are you trying to upload? Is it a txt file or an
> > special type?
> >
> >
> > Thanks,
> > Dan
> >
> >
> >
> > On Fri, Nov 30, 2012 at 8:03 AM, Carolina Marques <
> > carolina.marq...@smlinfo.com.br> wrote:
> >
> > > Hi Jmeter team,
> > >
> > > We are having some troubles when we try to set the configuration to
> > > upload a file through Jmeter. It is an ECM web application and is to
> > > save, manage and store many kind of documents and all these actions
> > > require lots of mouse clicks. The main application module is to
> > > import documents and to scan documents from a scanner connected to
> > > the computer and store this. But when we try to configure these
> > > modules into Jmeter, it is not working.
> > >
> > > The application works this way:
> > >
> > > - Import module: We fill the text fields with the required
> > > information, then, we press the selection button and an selection
> > > window is opened to select the file on the path chosen. After this
> > > the file is carried into the document that is being created and
> > > then, after this, the document is indexed.
> > >
> > > - Scan module: We fill the text fields with the required
> > > information, then we press the scan button and then, the image is
> > > scanned and is showed in the viewer, then we finish the document.
> > >
> > > In both modules are possible to i

Re: Analysing JMeter results

[Adrian Speteanu]

Overkill alert.

I looked at your test plan, you are testing every piece of static resurce!
Just tick the "download embeded resources" in the main SAMPLER if you
really really need it. Otherwise avoid it.  If you want to know if you have
problems with statics, just install YSlow for Firebug and use that to
generate a report and suggestion! It makes no sense to test them if your IT
ops have installed the proper content delivery infrastructure.

Oh man... I can't believe we still see this in tests. There are so many
threads about this.

--Adrian S


On Thu, Nov 29, 2012 at 5:36 PM, Mitesh Patel wrote:

> Thanks.
>
> I understand that only samplers get plotted and not controllers. However,
> I have 82 samplers (http requests) for my homepage. Some of these samplers
> are also used in the About us page, which itself has 60 samplers. So I
> potentially have around 100 samplers being plotted, but I have no idea
> which request was for which page.
>
> Thanks,
>
> > Date: Thu, 29 Nov 2012 17:07:28 +0200
> > Subject: Re: Analysing JMeter results
> > From: asp.ad...@gmail.com
> > To: user@jmeter.apache.org
> >
> > Hi,
> >
> > Controllers do not get plotted! Samplers do (its a java thingy, listeners
> > wait for info from samplers.)
> >
> > To view information specific to the whole test I keep a listener as child
> > to the thread group or even test plan. But, if I need to view the
> specifics
> > of a controller, than I add a new listener to the controller (this will
> > give me TOTAL that applies to everything from the controller's tree level
> > down, in an inclusive manner).
> >
> > To group samplers you will have to contact the owners and mailing lists
> of
> > the plugins though. AFAIK it works, but could be wrong and for me,
> > listeners in controllers seals the deal: I don't need more.
> >
> > --Adrian S
> >
> > On Thu, Nov 29, 2012 at 4:10 PM, Flavio Cysne 
> wrote:
> >
> > > Just modify the script sampler's label (not the URL) to distinct names,
> > > like h...@global.css and abou...@global.css
> > >
> > >
> > > 2012/11/29 Mitesh Patel 
> > >
> > > > Hi,
> > > >
> > > > I am new to JMeter and have managed to record my first test scenario
> via
> > > > the proxy. As expected, this records every single request. So, my
> plan
> > > > looks something like this:
> > > >
> > > > TEST PLAN
> > > > --THREAD GROUP
> > > > SCENARIO1 (RECORDING CONTROLLER)
> > > > --HOME (TRANSACTION CONTROLLER)
> > > > home.html (HTTP REQUEST)
> > > > global.css (HTTP REQUEST)
> > > > page.js (HTTP REQUEST)
> > > > banner.png (HTTP REQUEST)
> > > > etc
> > > > etc
> > > > .
> > > > .
> > > > --ABOUT US (TRANSACTION CONTROLLER)
> > > > aboutus.html (HTTP REQUEST)
> > > > global.css (HTTP REQUEST)
> > > > aboutus.css (HTTP REQUEST)
> > > > page.js (HTTP REQUEST)
> > > > image.png (HTTP REQUEST)
> > > >
> > > > Now, when I try and plot the results using the OOTB listeners or the
> > > > JMETER PLUGINS, I cannot seem to get an accurate view of the results
> for
> > > > HOME or ABOUT US. It seems to groups similar files i.e. global.css,
> > > > page.js. This doesn't really help me analyse the behaviour of the
> test.
> > > >
> > > > Any ideas how I can get around this?
> > > >
> > > > Thanks,
> > > >
> > >
>
>

Re: Analysing JMeter results

[Adrian Speteanu]

Hi,

Controllers do not get plotted! Samplers do (its a java thingy, listeners
wait for info from samplers.)

To view information specific to the whole test I keep a listener as child
to the thread group or even test plan. But, if I need to view the specifics
of a controller, than I add a new listener to the controller (this will
give me TOTAL that applies to everything from the controller's tree level
down, in an inclusive manner).

To group samplers you will have to contact the owners and mailing lists of
the plugins though. AFAIK it works, but could be wrong and for me,
listeners in controllers seals the deal: I don't need more.

--Adrian S

On Thu, Nov 29, 2012 at 4:10 PM, Flavio Cysne  wrote:

> Just modify the script sampler's label (not the URL) to distinct names,
> like h...@global.css and abou...@global.css
>
>
> 2012/11/29 Mitesh Patel 
>
> > Hi,
> >
> > I am new to JMeter and have managed to record my first test scenario via
> > the proxy. As expected, this records every single request. So, my plan
> > looks something like this:
> >
> > TEST PLAN
> > --THREAD GROUP
> > SCENARIO1 (RECORDING CONTROLLER)
> > --HOME (TRANSACTION CONTROLLER)
> > home.html (HTTP REQUEST)
> > global.css (HTTP REQUEST)
> > page.js (HTTP REQUEST)
> > banner.png (HTTP REQUEST)
> > etc
> > etc
> > .
> > .
> > --ABOUT US (TRANSACTION CONTROLLER)
> > aboutus.html (HTTP REQUEST)
> > global.css (HTTP REQUEST)
> > aboutus.css (HTTP REQUEST)
> > page.js (HTTP REQUEST)
> > image.png (HTTP REQUEST)
> >
> > Now, when I try and plot the results using the OOTB listeners or the
> > JMETER PLUGINS, I cannot seem to get an accurate view of the results for
> > HOME or ABOUT US. It seems to groups similar files i.e. global.css,
> > page.js. This doesn't really help me analyse the behaviour of the test.
> >
> > Any ideas how I can get around this?
> >
> > Thanks,
> >
>

Re: Server response: "browser does not support JavaScript"

[Adrian Speteanu]

Hi,

see bellow...

On Mon, Nov 19, 2012 at 4:22 PM, Shay Ginsbourg wrote:

> Hi,
>
> While running a Jmeter script, we receive a response from the server
> that the "browser does not support JavaScript".
>
> At this point, our test fails because the script can't continue.
>
> What is the recommended way of working around such an undesired server
> response ?
>

Well, black-boxed - I would assume that someone expects something from a
script that isn't running to make this assumption. So, I would review the
test and the web-page in firebug and see what was added since the script
was first created. Also, I wouldn't assume from start that this requires
javascript in jmeter, since is probably something that comes in
body/headers as false (or is inexistent) and is returned as it equals true
in headers/body, anything extremely simplistic.

Btw, some sites had this in the basic HTML -> if you were missing flash
player or blocked javascript with addons / browser settings - they had a
default text or picture saying you need to enable whatever was needed for
their site. If the site was slow enough you could see those - even if it
did eventually load ok. In jmeter, of course, you only got that in the
HTML. Just something I wanted to throw in there, not sure if it is the case.


>
> Please advise.
>
> regards,
> Shay
>
>
> www.ginsbourg.com
>
> -
> To unsubscribe, e-mail: user-unsubscr...@jmeter.apache.org
> For additional commands, e-mail: user-h...@jmeter.apache.org
>
>

Re: SoapSampler deprecation ?

[Adrian Speteanu]

Hi,

I guess if people who used don't speak up it means it's not needed anymore
:)

Don't see the need for it anymore, also.

Cheers,
Adrian S

On Wed, Nov 7, 2012 at 12:25 AM, Philippe Mouawad <
philippe.moua...@gmail.com> wrote:

> Hello,
> I was wondering if we shouldn't deprecate Soap Sampler .
>
> Here are the reasons for that:
>
>- Since introduction of Raw Post body in HTTP Sampler it has become
>rather useless except for SoapAction computation inside GUI
>- It suffers from a bug which seems hard to fix (not technically but
>because we don't know where the source code of soap.jar is) and which
> seems
>to me rather a big one:
>   - https://issues.apache.org/bugzilla/show_bug.cgi?id=54006
>- It relies on rather old library
>
>
> --
> Regards.
> Philippe
>

Re: Computing number of VU

[Adrian Speteanu]

Hi,

I can't decide between 1 or 2. IMO, they are not exclusive of each other,
they should be used in combination:

What I do:
   -1- have a clear breakdown of real-life traffic (or estimated traffic):
that many requests of each type (numbers, network traffic fluctuations,
frequency of requests and so on...)
-2- have a clear breakdown of number of users
-3- create a test script that reaches targets from #1 at the scale of
the test setup (which is not always as big as the production setup)
-4- fine tune the script until the number of active/inactive user
sessions (server side) coincides with production target...
--
Reasoning:
Yeah, its nice to know user session length and actual behaviour. But in
practice, due to a desire to protect your users' privacy or due to lack of
data, you only have a clue of what you actually need, but not a real grasp
of the numbers you need. More, often you do have the big numbers and the
detailed use-cases actually come from marketing/sales/client side - and the
second, I strongly recommend to "use with caution", consult them, but don't
trust them. In the end, the impact on the server-side, will be the same if
you make your homework correctly... IMO, with this method you cover
edge-cases too: the average user session might not have that request that
takes only 1% of the total traffic, but that request might be good to have
in the load test plan.

When lacking good statistics or when you have a new application with no
data from existing live environments, then you should simply push the
system to the limit, gradually, until performance would be unacceptable to
you and/or end-users.

Other:
Last, but not least, you should NOT focus solely on simulating VUs. I still
see this trend of mistrusting synthetic benchmarks and is ridiculous. Yes,
some tests do not reflect how the end-user will perceive performance, BUT
those kind of tests: 1) give better understanding to the dev team of what
the problem is, performance wise - which obviously is needed in order to
FIX IT; 2) can be used to monitor performance fluctuation of specific
functionalities and this can be monitored over time, with multiple app
versions; 3) they make it easier for you, the tester, to understand the
application architecture and performance particularities which are actually
essential in order to create a realistic test that tries to simulate VUs...
so start with this, end with what you had in mind before you sent the email.

Cheers,
Adrian S


On Fri, Nov 2, 2012 at 5:15 PM, Philippe Bossu  wrote:

> Hello,
>
> I have to load test a website that has a number of visits per days and a
> visit duration.
>
> What is the formula to compute the number of Virtual Users (Threads) that I
> need to put in my tests.
>
> I read that this number can be computed using 2 different informations:
>
>- Method1 => Peak visit rate (visits/hour) and Average visit length
>(minutes/visit), it would then be =>  visitRate/(60/visitLength)
>
>
>- Method 2 => Peak page rate (pages/hour), Testcase size (number of
>pages seen by User) and Testcase duration (in minutes),it would then be
>=> (Peak page rate*testcaseDuration)/(60*testcaseSize)
>
>
>
>
> My question is with method 1, must my test Iteration last the "average
> visit length" using Timers and what more to make it last that length ?
>
>
> My question with method 2, are we talking about my Test Plan duration ?
>
>
> Which method is the best one ?
>
>
> Thank you
>
> Regards
>