Nabble - Grinder - User

Re: Running Grinder console without UI

2009-12-02T12:38:41Z

No, see
http://www.google.com/search?hl=en&q=%22net.grinder.console.client%22

Jyothi Krishnan wrote:

> Hi Philip,
>
> I found the thread I think. Is it the XVFB option?
>
> Thank you
> Jo
> --- On *Tue, 12/1/09, Philip Aston /<philip.aston@...>/* wrote:
>
>
> From: Philip Aston <philip.aston@...>
> Subject: Re: [Grinder-use] Running Grinder console without UI
> To: "grinder-use" <grinder-use@...>
> Date: Tuesday, December 1, 2009, 10:42 PM
>
> josri wrote:
> > Hi,
> > Is there anyway to run Grinder console without the UI?
> >
> > Thank you
> > Jo
> >
> -headless
>
> You'll have to use the (undcoumented) console API to control it
> though.
> Search this list for more details.
>
> - Phil
>
>
> ------------------------------------------------------------------------------
> Join us December 9, 2009 for the Red Hat Virtual Experience,
> a free event focused on virtualization and cloud computing.
> Attend in-depth sessions from your desk. Your couch. Anywhere.
> http://p.sf.net/sfu/redhat-sfdev2dev
> _______________________________________________
> grinder-use mailing list
> grinder-use@...
> </mc/compose?to=grinder-use@...>
> https://lists.sourceforge.net/lists/listinfo/grinder-use
>
>
> ------------------------------------------------------------------------
>
> ------------------------------------------------------------------------------
> Join us December 9, 2009 for the Red Hat Virtual Experience,
> a free event focused on virtualization and cloud computing.
> Attend in-depth sessions from your desk. Your couch. Anywhere.
> http://p.sf.net/sfu/redhat-sfdev2dev
> ------------------------------------------------------------------------
>
> _______________________________________________
> grinder-use mailing list
> grinder-use@...
> https://lists.sourceforge.net/lists/listinfo/grinder-use
>

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
grinder-use mailing list
grinder-use@...
https://lists.sourceforge.net/lists/listinfo/grinder-use

Re: Grinder and bandwidth throttling

2009-12-02T12:23:39Z

Thanks for sharing those links on the user-list. I’m sure they’ll help many a folks in the future. And yes, help ppl like me learn more about Grinder too :]

Thanks again!

anirvan

From: Jyothi Krishnan [mailto:stopjo@...]
Sent: 02 December 2009 23:45
To: grinder-use
Subject: Re: [Grinder-use] Grinder and bandwidth throttling

Hi Anirvan,

I found the foll. 2 links. Thought might be useful.

http://blackanvil.blogspot.com/2006/11/grinder-addressing-warts.html

http://mail-archives.apache.org/mod_mbox/geronimo-scm/200809.mbox/%3C20080929201911.521972388A01@...%3E

Thank you
Jo
--- On Wed, 12/2/09, Anirvan Majumdar <anirvan.majumdar@...> wrote:

From: Anirvan Majumdar <anirvan.majumdar@...>
Subject: Re: [Grinder-use] Grinder and bandwidth throttling
To: "'grinder-use'" <grinder-use@...>
Date: Wednesday, December 2, 2009, 9:32 AM

I do not know whether Grinder comes with in-built support for throttling,
but you might want to try some bandwidth throttling/shaping software which
can simulate a throttled network connection.
Perhaps one of the tools mentioned here
[http://stackoverflow.com/questions/1094760/network-tools-that-simulate-slow
-network-connection] will come in handy.

anirvan

|-----Original Message-----
|From: josri [mailto:stopjo@...]
|Sent: 02 December 2009 22:39
|To: grinder-use@...
|Subject: [Grinder-use] Grinder and bandwidth throttling
|
|
|Hi,
|
|Does grinder support variable connection speed/bandwidth throttling? If
|so,
|how do I do that? Any pointers would be helpful
|
|Thank you
|Jo
|--
|View this message in context: http://old.nabble.com/Grinder-and-
|bandwidth-throttling-tp26612563p26612563.html
|Sent from the Grinder - User mailing list archive at Nabble.com.
|
|
|------------------------------------------------------------------------
|------
|Join us December 9, 2009 for the Red Hat Virtual Experience,
|a free event focused on virtualization and cloud computing.
|Attend in-depth sessions from your desk. Your couch. Anywhere.
|http://p.sf.net/sfu/redhat-sfdev2dev
|_______________________________________________
|grinder-use mailing list
|grinder-use@...
|https://lists.sourceforge.net/lists/listinfo/grinder-use

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
grinder-use mailing list
grinder-use@...
https://lists.sourceforge.net/lists/listinfo/grinder-use

Shutdown questions

2009-12-02T11:53:56Z

Hi,

First time poster. Loving The Grinder.

When I'm running a test from the console and I click on the "Reset the worker processes" button in the console, I get lots of ShutdownExceptions.

What's confusing to me it that I expected that when I clicked that button the __del__ method would get called immediately. But that's not what happens. Instead, the remaining runs are all called first, each failing with a ShutdownException, and then __del__ gets called.

Is there a way that I can reverse this behavior, or at least check the shutdown status at the top of my __call__ method and exit out of it quickly.

My goal here is to not have an error file when I click on that button if I'm just aborting early.

Thanks for any help,

Mark

Re: Grinder and bandwidth throttling

2009-12-02T10:14:52Z

From: Anirvan Majumdar <anirvan.majumdar@...>
Subject: Re: [Grinder-use] Grinder and bandwidth throttling
To: "'grinder-use'" <grinder-use@...>
Date: Wednesday, December 2, 2009, 9:32 AM

I do not know whether Grinder comes with in-built support for throttling,
but you might want to try some bandwidth throttling/shaping software which
can simulate a throttled network connection.
Perhaps one of the tools mentioned here
[http://stackoverflow.com/questions/1094760/network-tools-that-simulate-slow
-network-connection] will come in handy.

anirvan

|-----Original Message-----
|From: josri [mailto:stopjo@...]
|Sent: 02 December 2009 22:39
|To: grinder-use@...
|Subject: [Grinder-use] Grinder and bandwidth throttling
|
|
|Hi,
|
|Does grinder support variable connection speed/bandwidth throttling? If
|so,
|how do I do that? Any pointers would be helpful
|
|Thank you
|Jo
|--
|View this message in context: http://old.nabble.com/Grinder-and-
|bandwidth-throttling-tp26612563p26612563.html
|Sent from the Grinder - User mailing list archive at Nabble.com.
|
|
|------------------------------------------------------------------------
|------
|Join us December 9, 2009 for the Red Hat Virtual Experience,
|a free event focused on virtualization and cloud computing.
|Attend in-depth sessions from your desk. Your couch. Anywhere.
|http://p.sf.net/sfu/redhat-sfdev2dev
|_______________________________________________
|grinder-use mailing list
|grinder-use@...
|https://lists.sourceforge.net/lists/listinfo/grinder-use

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
grinder-use mailing list
grinder-use@...
https://lists.sourceforge.net/lists/listinfo/grinder-use

Re: Grinder and bandwidth throttling

2009-12-02T09:32:54Z

Re: Running Grinder console without UI

2009-12-02T09:23:01Z

Hi Philip,

I found the thread I think. Is it the XVFB option?

Thank you
Jo
--- On Tue, 12/1/09, Philip Aston <philip.aston@...> wrote:

From: Philip Aston <philip.aston@...>
Subject: Re: [Grinder-use] Running Grinder console without UI
To: "grinder-use" <grinder-use@...>
Date: Tuesday, December 1, 2009, 10:42 PM

josri wrote:
> Hi,
> Is there anyway to run Grinder console without the UI?
>
> Thank you
> Jo
>
-headless

You'll have to use the (undcoumented) console API to control it though.
Search this list for more details.

- Phil

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
grinder-use mailing list
grinder-use@...
https://lists.sourceforge.net/lists/listinfo/grinder-use

Grinder and bandwidth throttling

2009-12-02T09:09:25Z

Hi,

Does grinder support variable connection speed/bandwidth throttling? If so, how do I do that? Any pointers would be helpful

Thank you
Jo

Re: Grinder Agent log

2009-12-02T09:05:45Z

We are eliminating Hudson for now. So, we are ok. We can live with the results produced in each agent and the consolidated report from the Grinder Console.

Thank you Anirvan you answers were really helpful in understanding how Grinder reports
Jo

Anirvan wrote:

Are you interested in the "data" log file or the "out" log file?
The Grinder results at Console correspond to the "data" log file format.

>From your description I feel you want to get a handle on the individual
agent "out" log file. But do you really want to capture information at
individual thread level operating at every agent?

anirvan

On Wed, Dec 2, 2009 at 11:19 AM, josri <stopjo@yahoo.com> wrote:

>
> Thank you for the response.
> I am starting the Grinder console from the hudson (Continuous Integration
> tool) using Grinder Plugin. The plugin has a place to give the output
> filename which is picked up to produce a chart at the end of the run. So,
> if the log file is only generated in the agent, then what Grinder Plugin
> would work only if the agent is running on the same machine? Also, when I
> save the grinder results from the console, it only gives the summary :(
>
> Thank you
> Jo
>
>
> Anirvan wrote:
> >
> > The result is consolidated in the Console. The way it should be set up is
> > -
> > run your agents on multiple systems. Configure each of the agents to talk
> > to
> > a single Console running on another system. When the agents run, they log
> > their statistics on the respective machines as well as send them to the
> > Console where it gets consolidated. However, no consolidated log file is
> > generated at the Console m/c. You'll need to save the test results in the
> > Console using an option present under the 'File' menu option.
> >
> > anirvan
> >
> > |-----Original Message-----
> > |From: josri [mailto:stopjo@yahoo.com]
> > |Sent: 02 December 2009 05:18
> > |To: grinder-use@lists.sourceforge.net
> > |Subject: [Grinder-use] Grinder Agent log
> > |
> > |
> > |Hi,
> > |
> > |I run console on a machine and multiple agents from different machines,
> > |the
> > |log is written in a separate log file on each machine. How is the
> > |result
> > |consolidated? Isn't it supposed to be one report when you run multiple
> > |agents? I have hudson running on the console machine and hudson is not
> > |able
> > |to see the log files. How do I make the logs to be in one single log
> > |file?
> > |
> > |Thank you
> > |Jo
> > |--
> > |View this message in context: http://old.nabble.com/Grinder-Agent-log-
> > |tp26600215p26600215.html
> > |Sent from the Grinder - User mailing list archive at Nabble.com.
> > |
> > |
> > |------------------------------------------------------------------------
> > |------
> > |Join us December 9, 2009 for the Red Hat Virtual Experience,
> > |a free event focused on virtualization and cloud computing.
> > |Attend in-depth sessions from your desk. Your couch. Anywhere.
> > |http://p.sf.net/sfu/redhat-sfdev2dev
> > |_______________________________________________
> > |grinder-use mailing list
> > |grinder-use@lists.sourceforge.net
> > |https://lists.sourceforge.net/lists/listinfo/grinder-use
> >
> >
> >
> ------------------------------------------------------------------------------
> > Join us December 9, 2009 for the Red Hat Virtual Experience,
> > a free event focused on virtualization and cloud computing.
> > Attend in-depth sessions from your desk. Your couch. Anywhere.
> > http://p.sf.net/sfu/redhat-sfdev2dev
> > _______________________________________________
> > grinder-use mailing list
> > grinder-use@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/grinder-use
> >
> >
>
> --
> View this message in context:
> http://old.nabble.com/Grinder-Agent-log-tp26600215p26603937.html
> Sent from the Grinder - User mailing list archive at Nabble.com.
>
>
>
> ------------------------------------------------------------------------------
> Join us December 9, 2009 for the Red Hat Virtual Experience,
> a free event focused on virtualization and cloud computing.
> Attend in-depth sessions from your desk. Your couch. Anywhere.
> http://p.sf.net/sfu/redhat-sfdev2dev
> _______________________________________________
> grinder-use mailing list
> grinder-use@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/grinder-use
>

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
grinder-use mailing list
grinder-use@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/grinder-use

Re: Running Grinder console without UI

2009-12-02T08:37:53Z

I'm afraid, I'm not able to find the discussion you are talking about. Searched with the term "console api", "Grinder console api".... May be not searching with the right term. Can you please help?

Thank you
Jo

--- On Tue, 12/1/09, Philip Aston <philip.aston@...> wrote:

From: Philip Aston <philip.aston@...>
Subject: Re: [Grinder-use] Running Grinder console without UI
To: "grinder-use" <grinder-use@...>
Date: Tuesday, December 1, 2009, 10:42 PM

josri wrote:
> Hi,
> Is there anyway to run Grinder console without the UI?
>
> Thank you
> Jo
>
-headless

You'll have to use the (undcoumented) console API to control it though.
Search this list for more details.

- Phil

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
grinder-use mailing list
grinder-use@...
https://lists.sourceforge.net/lists/listinfo/grinder-use

Re: Running Grinder console without UI

2009-12-01T22:42:02Z

josri wrote:
> Hi,
> Is there anyway to run Grinder console without the UI?
>
> Thank you
> Jo
>
-headless

You'll have to use the (undcoumented) console API to control it though.
Search this list for more details.

- Phil

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
grinder-use mailing list
grinder-use@...
https://lists.sourceforge.net/lists/listinfo/grinder-use

Re: Grinder Agent log

2009-12-01T22:29:11Z

Are you interested in the "data" log file or the "out" log file?
The Grinder results at Console correspond to the "data" log file format.

From your description I feel you want to get a handle on the individual agent "out" log file. But do you really want to capture information at individual thread level operating at every agent?

anirvan

On Wed, Dec 2, 2009 at 11:19 AM, josri <stopjo@...> wrote:

Thank you for the response.
I am starting the Grinder console from the hudson (Continuous Integration
tool) using Grinder Plugin. The plugin has a place to give the output
filename which is picked up to produce a chart at the end of the run. So,
if the log file is only generated in the agent, then what Grinder Plugin
would work only if the agent is running on the same machine? Also, when I
save the grinder results from the console, it only gives the summary :(

Thank you
Jo

Anirvan wrote:
>
> The result is consolidated in the Console. The way it should be set up is
> -
> run your agents on multiple systems. Configure each of the agents to talk
> to
> a single Console running on another system. When the agents run, they log
> their statistics on the respective machines as well as send them to the
> Console where it gets consolidated. However, no consolidated log file is
> generated at the Console m/c. You'll need to save the test results in the
> Console using an option present under the 'File' menu option.
>
> anirvan
>
> |-----Original Message-----
> |From: josri [mailto:stopjo@...]
> |Sent: 02 December 2009 05:18
> |To: grinder-use@...
> |Subject: [Grinder-use] Grinder Agent log
> |
> |
> |Hi,
> |
> |I run console on a machine and multiple agents from different machines,
> |the
> |log is written in a separate log file on each machine. How is the
> |result
> |consolidated? Isn't it supposed to be one report when you run multiple
> |agents? I have hudson running on the console machine and hudson is not
> |able
> |to see the log files. How do I make the logs to be in one single log
> |file?
> |
> |Thank you
> |Jo
> |--
> |View this message in context: http://old.nabble.com/Grinder-Agent-log-
> |tp26600215p26600215.html
> |Sent from the Grinder - User mailing list archive at Nabble.com.
> |
> |
> |------------------------------------------------------------------------
> |------
> |Join us December 9, 2009 for the Red Hat Virtual Experience,
> |a free event focused on virtualization and cloud computing.
> |Attend in-depth sessions from your desk. Your couch. Anywhere.
> |http://p.sf.net/sfu/redhat-sfdev2dev
> |_______________________________________________
> |grinder-use mailing list
> |grinder-use@...
> |https://lists.sourceforge.net/lists/listinfo/grinder-use
>
>

> ------------------------------------------------------------------------------

> Join us December 9, 2009 for the Red Hat Virtual Experience,
> a free event focused on virtualization and cloud computing.
> Attend in-depth sessions from your desk. Your couch. Anywhere.
> http://p.sf.net/sfu/redhat-sfdev2dev
> _______________________________________________
> grinder-use mailing list
> grinder-use@...
> https://lists.sourceforge.net/lists/listinfo/grinder-use
>
>

--
View this message in context: http://old.nabble.com/Grinder-Agent-log-tp26600215p26603937.html

Sent from the Grinder - User mailing list archive at Nabble.com.

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
grinder-use mailing list
grinder-use@...
https://lists.sourceforge.net/lists/listinfo/grinder-use

Re: Grinder Agent log

2009-12-01T21:49:54Z

Thank you for the response.
I am starting the Grinder console from the hudson (Continuous Integration tool) using Grinder Plugin. The plugin has a place to give the output filename which is picked up to produce a chart at the end of the run. So, if the log file is only generated in the agent, then what Grinder Plugin would work only if the agent is running on the same machine? Also, when I save the grinder results from the console, it only gives the summary :(

Thank you
Jo

Anirvan wrote:

The result is consolidated in the Console. The way it should be set up is -
run your agents on multiple systems. Configure each of the agents to talk to
a single Console running on another system. When the agents run, they log
their statistics on the respective machines as well as send them to the
Console where it gets consolidated. However, no consolidated log file is
generated at the Console m/c. You'll need to save the test results in the
Console using an option present under the 'File' menu option.

anirvan

|-----Original Message-----
|From: josri [mailto:stopjo@yahoo.com]
|Sent: 02 December 2009 05:18
|To: grinder-use@lists.sourceforge.net
|Subject: [Grinder-use] Grinder Agent log
|
|
|Hi,
|
|I run console on a machine and multiple agents from different machines,
|the
|log is written in a separate log file on each machine. How is the
|result
|consolidated? Isn't it supposed to be one report when you run multiple
|agents? I have hudson running on the console machine and hudson is not
|able
|to see the log files. How do I make the logs to be in one single log
|file?
|
|Thank you
|Jo
|--
|View this message in context: http://old.nabble.com/Grinder-Agent-log-
|tp26600215p26600215.html
|Sent from the Grinder - User mailing list archive at Nabble.com.
|
|
|------------------------------------------------------------------------
|------
|Join us December 9, 2009 for the Red Hat Virtual Experience,
|a free event focused on virtualization and cloud computing.
|Attend in-depth sessions from your desk. Your couch. Anywhere.
|http://p.sf.net/sfu/redhat-sfdev2dev
|_______________________________________________
|grinder-use mailing list
|grinder-use@lists.sourceforge.net
|https://lists.sourceforge.net/lists/listinfo/grinder-use

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
grinder-use mailing list
grinder-use@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/grinder-use

Ben Woodhouse is out of the office.

2009-12-01T20:00:29Z

I will be out of the office starting 30/11/2009 and will not return until
01/03/2010.

I am now based on customer site without access to my IBM email.

I can be contacted either at ben.woodhouse@... or on 07920 211824 /
ITN (37261616)

I will check my mail on a weekly basis, should the matter not be urgent.

Thanks

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
grinder-use mailing list
grinder-use@...
https://lists.sourceforge.net/lists/listinfo/grinder-use

Re: Grinder Agent log

2009-12-01T19:01:06Z

The result is consolidated in the Console. The way it should be set up is -
run your agents on multiple systems. Configure each of the agents to talk to
a single Console running on another system. When the agents run, they log
their statistics on the respective machines as well as send them to the
Console where it gets consolidated. However, no consolidated log file is
generated at the Console m/c. You'll need to save the test results in the
Console using an option present under the 'File' menu option.

anirvan

|-----Original Message-----
|From: josri [mailto:stopjo@...]
|Sent: 02 December 2009 05:18
|To: grinder-use@...
|Subject: [Grinder-use] Grinder Agent log
|
|
|Hi,
|
|I run console on a machine and multiple agents from different machines,
|the
|log is written in a separate log file on each machine. How is the
|result
|consolidated? Isn't it supposed to be one report when you run multiple
|agents? I have hudson running on the console machine and hudson is not
|able
|to see the log files. How do I make the logs to be in one single log
|file?
|
|Thank you
|Jo
|--
|View this message in context: http://old.nabble.com/Grinder-Agent-log-
|tp26600215p26600215.html
|Sent from the Grinder - User mailing list archive at Nabble.com.
|
|
|------------------------------------------------------------------------
|------
|Join us December 9, 2009 for the Red Hat Virtual Experience,
|a free event focused on virtualization and cloud computing.
|Attend in-depth sessions from your desk. Your couch. Anywhere.
|http://p.sf.net/sfu/redhat-sfdev2dev
|_______________________________________________
|grinder-use mailing list
|grinder-use@...
|https://lists.sourceforge.net/lists/listinfo/grinder-use

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
grinder-use mailing list
grinder-use@...
https://lists.sourceforge.net/lists/listinfo/grinder-use

Re: Running Grinder console without UI

2009-12-01T18:57:39Z

Not that I'm aware of. However, if you've got only one agent, then you do
not need to use the Console. In that case, set the useConsole parameter to
false in the grinder.properties file.

|-----Original Message-----
|From: josri [mailto:stopjo@...]
|Sent: 02 December 2009 04:34
|To: grinder-use@...
|Subject: [Grinder-use] Running Grinder console without UI
|
|
|Hi,
|Is there anyway to run Grinder console without the UI?
|
|Thank you
|Jo
|--
|View this message in context: http://old.nabble.com/Running-Grinder-
|console-without-UI-tp26600659p26600659.html
|Sent from the Grinder - User mailing list archive at Nabble.com.
|
|
|------------------------------------------------------------------------
|------
|Join us December 9, 2009 for the Red Hat Virtual Experience,
|a free event focused on virtualization and cloud computing.
|Attend in-depth sessions from your desk. Your couch. Anywhere.
|http://p.sf.net/sfu/redhat-sfdev2dev
|_______________________________________________
|grinder-use mailing list
|grinder-use@...
|https://lists.sourceforge.net/lists/listinfo/grinder-use

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
grinder-use mailing list
grinder-use@...
https://lists.sourceforge.net/lists/listinfo/grinder-use

Grinder Agent logs

2009-12-01T15:50:58Z

Hi,

I run console on a machine and multiple agents from different machines, the log is written in a separate log file on each machine. How is the result consolidated? Isn't it supposed to be one report when you run multiple agents? I have hudson running on the console machine and hudson is not able to see the log files. How do I make the logs to be in one single log file?

Thank you
Jo

Running Grinder console without UI

2009-12-01T15:03:33Z

Hi,
Is there anyway to run Grinder console without the UI?

Thank you
Jo

Grinder Agent log

2009-12-01T14:31:42Z

Performance Testing of any Flash based Web Site

2009-11-30T12:56:26Z

Hello All,

Is there a way to do Performance Testing of any Flash based Web Site. I
see that it is possible using Neoload and latest version of Load
Runner/SP. Can anyone please let me know if it is possible using The
Grinder.

Regards,
Rahman.

------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing.
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
grinder-use mailing list
grinder-use@...
https://lists.sourceforge.net/lists/listinfo/grinder-use

Re: FW: BUG :: Grinder 3.2b // SSLSocketFactory

2009-11-26T04:38:47Z

Re: [Grinder-use] FW: BUG :: Grinder 3.2b // SSLSocketFactory Ah, should work then... Thanks :-)

G.

Am 11/26/09 12:00 PM schrieb "Philip Aston" unter <philip.aston@...>:

Yes.

http://grinder.svn.sourceforge.net/viewvc/grinder/trunk/source/src/net/grinder/util/InsecureSSLContextFactory.java?view=markup

Guido Serra wrote:

Re: [Grinder-use] FW: BUG :: Grinder 3.2b // SSLSocketFactory Hi Phil,
It looks like it does not.

Did u overrided the CA Trust mechanism? That’s the only way u could have achieved that...

Regards,
G.

Am 11/26/09 11:31 AM schrieb "Philip Aston" unter <philip.aston@...>:



You do not need to do this. The Grinder will automatically trust your server's certificate.

- Phil

Guido Serra wrote:


Re: [Grinder-use] FW: BUG :: Grinder 3.2b // SSLSocketFactory Hi Phil,
If ur target is using a certificate which is “home signed”, u need to import his CA.
Java mechanism to do this is the trustStore. We need to be able to override it.

Regards,
G.

Am 11/26/09 11:05 AM schrieb "Philip Aston" unter <philip.aston@...>:




The current implementation will not reject any connections on the basis of trust. There is no requirement to provide trust certificates.

Can you explain why you need to do so?

- Phil

Guido Serra wrote:



Re: [Grinder-use] FW: BUG :: Grinder 3.2b // SSLSocketFactory Hi Phil,
With the current set of API it is not possible to override the trustStore.

We are creating internally a patch, would u like to integrate it? Should we do it?

Regards,
G.

Am 11/9/09 10:38 PM schrieb "Philip Aston" unter <philip.aston@...>:





Serra, Guido, VF-Group wrote:
> If the server requires to present an SSL certificate as credentials and ... The server certificate is signed with a CA which is not in the SUN default TrustStore, u need to override the SSLSocketFactory or provide a keyStore and a trustStore from the jvm command line
>

Or you could just trust everything, which is what The Grinder does.
Refer to the URL I quote below for details.

- Phil

> The method to override this, implemented in the HTTPConnection class, doesn't seem to work properly. It relies on the default values
>
> Regards,
> G.
>
>
> www.vodafone.com <http://www.vodafone.com> <http://www.vodafone.com> <http://www.vodafone.com>
>
>
> Vodafone Group Services GmbH
> Mannesmannufer 2 ∙ D-40213 Düsseldorf
> Amtsgericht Düsseldorf, HRB 53554
> Geschäftsführung: Dr. Joachim Peters, Rainer Wallek
>
>
>
> This message and any files or documents attached are confidential and may also be legally privileged or protected by other legal rules. It is intended only for the individual or entity named. If you are not the named addressee or you have received this email in error, please inform the sender immediately, delete it from your system and do not copy or disclose it or its contents or use it for any purpose. Thank you. Please also note that transmission cannot be guaranteed to be secure or error-free.
>
> ----- Original Message -----
> From: Philip Aston <philip.aston@...>
> To: grinder-use <grinder-use@...>
> Sent: Mon Nov 09 20:44:51 2009
> Subject: Re: [Grinder-use] FW: BUG :: Grinder 3.2b // SSLSocketFactory
>
> What is the reason you need a custom SSLSocketFactory?
>
> By default, the grinder does not validate any certificates, it trusts
> everything. See http://grinder.sourceforge.net/g3/ssl-support.html
>
> - Phil
>
> Guido Serra wrote:
>
>> Hi All,
>> Should I open an official bug, or ... Can u have a look at ur code,
>> please?
>>
>> Regards,
>> G.
>>
>> +++++++++++++++++++++++++++++++++++++++
>>
>> Hi All,
>> There’s a bug in grinder (the grinder.sf.net) source code.
>>
>> XXXXX implemented a very nice feature to manage certificates, but it
>> doesn’t work due to the way grinder manages internally the
>> SSLSocketFactory. The code from the framework tries to override that
>> class instance.
>>
>> Internally grinder doesn’t keep the sslFactory overridden by our
>> framework.
>>
>> See lines 596,597 of file src/HTTPClient/HTTPConnection.java :
>>
>>    final SSLSocket socket = (SSLSocket)defaultSSLFactory.createSocket();
>>     sslProtocols = socket.getSupportedProtocols();
>>
>> And line 3099:
>>
>>    sock = sslFactory.createSocket(sock, Host, Port, true);
>>
>> What scares me is this loop/comment in grinder.sf.net source code...
>>
>> int try_count = 3;
>>   /* what a hack! This is to handle the case where the server closes
>>    * the connection but we don't realize it until we try to send
>>    * something. The problem is that we only get IOException, but
>>    * we need a finer specification (i.e. whether it's an EPIPE or
>>    * something else); I don't trust relying on the message part
>>    * of IOException (which on SunOS/Solaris gives 'Broken pipe',
>>    * but what on Windoze/Mac?).
>>    */
>>
>>   while (try_count-- > 0)
>>   {
>>       try
>>       {
>>     // get a client socket
>>
>>
>> I commented out the line were we override the SSLContext from jython code.
>> And set the certificates I trust at JVM level... I had to specify
>> directly the server certificate as trustStore.
>>
>> -Djavax.net.ssl.trustStore=env/_certificates/*dgig.jks* \
>> -Djavax.net.ssl.trustStorePassword=vod@fone \
>> -Djavax.net.ssl.keyStorePassword=importkey \
>> -Djavax.net.ssl.keyStore=env/_certificates/10000000001189cc2bbbe_client2.jks
>> \
>>
>> With the DEVVodafone CA, it didn’t worked.
>>
>> NOTE! How I noticed the sslContext overwritten?
>>
>> >From the httpclient.log ... The first POST worked... While other 3
>> POST failed.
>>
>> The createSocket() function is instantiated 4 times in the
>> HTTPConnection.java ... 3 times in the loop above, and once by the
>> (SSLSocket)defaultSSLFactory .... (see enclosed file)
>>
>> I committed my mods... On trunk/ , do not trust them till XXXXX
>> reviews the code, please.
>>
>> Authentication realm: <http://85.205.240.168:8080> Subversion Repository
>> Password for 'guido.serra':
>> Adding (bin) env/_certificates/10000000001189cc2bbbe_client2.jks
>> Adding (bin) env/_certificates/DEV_TLServer_2048CAcert.truststore.jks
>> Adding (bin) env/_certificates/dgig.jks
>> Sending        src/com/vodafone/qa/common/httpclient.py
>> Sending        src/test.py
>>
>> Regards,
>> G.
>>

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july

_______________________________________________
grinder-use mailing list
grinder-use@...
https://lists.sourceforge.net/lists/listinfo/grinder-use

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
_______________________________________________
grinder-use mailing list
grinder-use@...
https://lists.sourceforge.net/lists/listinfo/grinder-use

Re: FW: BUG :: Grinder 3.2b // SSLSocketFactory

2009-11-26T03:00:18Z

Yes.

http://grinder.svn.sourceforge.net/viewvc/grinder/trunk/source/src/net/grinder/util/InsecureSSLContextFactory.java?view=markup

Guido Serra wrote:

Re: [Grinder-use] FW: BUG :: Grinder 3.2b // SSLSocketFactory Hi Phil,
It looks like it does not.

Did u overrided the CA Trust mechanism? That’s the only way u could have achieved that...

Regards,
G.

Am 11/26/09 11:31 AM schrieb "Philip Aston" unter <philip.aston@...>:

You do not need to do this. The Grinder will automatically trust your server's certificate.

- Phil

Guido Serra wrote:

Re: [Grinder-use] FW: BUG :: Grinder 3.2b // SSLSocketFactory Hi Phil,
If ur target is using a certificate which is “home signed”, u need to import his CA.
Java mechanism to do this is the trustStore. We need to be able to override it.

Regards,
G.

Am 11/26/09 11:05 AM schrieb "Philip Aston" unter <philip.aston@...>:



The current implementation will not reject any connections on the basis of trust. There is no requirement to provide trust certificates.

Can you explain why you need to do so?

- Phil

Guido Serra wrote:


Re: [Grinder-use] FW: BUG :: Grinder 3.2b // SSLSocketFactory Hi Phil,
With the current set of API it is not possible to override the trustStore.

We are creating internally a patch, would u like to integrate it? Should we do it?

Regards,
G.

Am 11/9/09 10:38 PM schrieb "Philip Aston" unter <philip.aston@...>:




Serra, Guido, VF-Group wrote:
> If the server requires to present an SSL certificate as credentials and ... The server certificate is signed with a CA which is not in the SUN default TrustStore, u need to override the SSLSocketFactory or provide a keyStore and a trustStore from the jvm command line
>

Or you could just trust everything, which is what The Grinder does.
Refer to the URL I quote below for details.

- Phil

> The method to override this, implemented in the HTTPConnection class, doesn't seem to work properly. It relies on the default values
>
> Regards,
> G.
>
>
> www.vodafone.com <http://www.vodafone.com> <http://www.vodafone.com>
>
>
> Vodafone Group Services GmbH
> Mannesmannufer 2 ∙ D-40213 Düsseldorf
> Amtsgericht Düsseldorf, HRB 53554
> Geschäftsführung: Dr. Joachim Peters, Rainer Wallek
>
>
>
> This message and any files or documents attached are confidential and may also be legally privileged or protected by other legal rules. It is intended only for the individual or entity named. If you are not the named addressee or you have received this email in error, please inform the sender immediately, delete it from your system and do not copy or disclose it or its contents or use it for any purpose. Thank you. Please also note that transmission cannot be guaranteed to be secure or error-free.
>
> ----- Original Message -----
> From: Philip Aston <philip.aston@...>
> To: grinder-use <grinder-use@...>
> Sent: Mon Nov 09 20:44:51 2009
> Subject: Re: [Grinder-use] FW: BUG :: Grinder 3.2b // SSLSocketFactory
>
> What is the reason you need a custom SSLSocketFactory?
>
> By default, the grinder does not validate any certificates, it trusts
> everything. See http://grinder.sourceforge.net/g3/ssl-support.html
>
> - Phil
>
> Guido Serra wrote:
>
>> Hi All,
>> Should I open an official bug, or ... Can u have a look at ur code,
>> please?
>>
>> Regards,
>> G.
>>
>> +++++++++++++++++++++++++++++++++++++++
>>
>> Hi All,
>> There’s a bug in grinder (the grinder.sf.net) source code.
>>
>> XXXXX implemented a very nice feature to manage certificates, but it
>> doesn’t work due to the way grinder manages internally the
>> SSLSocketFactory. The code from the framework tries to override that
>> class instance.
>>
>> Internally grinder doesn’t keep the sslFactory overridden by our
>> framework.
>>
>> See lines 596,597 of file src/HTTPClient/HTTPConnection.java :
>>
>>    final SSLSocket socket = (SSLSocket)defaultSSLFactory.createSocket();
>>     sslProtocols = socket.getSupportedProtocols();
>>
>> And line 3099:
>>
>>    sock = sslFactory.createSocket(sock, Host, Port, true);
>>
>> What scares me is this loop/comment in grinder.sf.net source code...
>>
>> int try_count = 3;
>>   /* what a hack! This is to handle the case where the server closes
>>    * the connection but we don't realize it until we try to send
>>    * something. The problem is that we only get IOException, but
>>    * we need a finer specification (i.e. whether it's an EPIPE or
>>    * something else); I don't trust relying on the message part
>>    * of IOException (which on SunOS/Solaris gives 'Broken pipe',
>>    * but what on Windoze/Mac?).
>>    */
>>
>>   while (try_count-- > 0)
>>   {
>>       try
>>       {
>>     // get a client socket
>>
>>
>> I commented out the line were we override the SSLContext from jython code.
>> And set the certificates I trust at JVM level... I had to specify
>> directly the server certificate as trustStore.
>>
>> -Djavax.net.ssl.trustStore=env/_certificates/*dgig.jks* \
>> -Djavax.net.ssl.trustStorePassword=vod@fone \
>> -Djavax.net.ssl.keyStorePassword=importkey \
>> -Djavax.net.ssl.keyStore=env/_certificates/10000000001189cc2bbbe_client2.jks
>> \
>>
>> With the DEVVodafone CA, it didn’t worked.
>>
>> NOTE! How I noticed the sslContext overwritten?
>>
>> >From the httpclient.log ... The first POST worked... While other 3
>> POST failed.
>>
>> The createSocket() function is instantiated 4 times in the
>> HTTPConnection.java ... 3 times in the loop above, and once by the
>> (SSLSocket)defaultSSLFactory .... (see enclosed file)
>>
>> I committed my mods... On trunk/ , do not trust them till XXXXX
>> reviews the code, please.
>>
>> Authentication realm: <http://85.205.240.168:8080> Subversion Repository
>> Password for 'guido.serra':
>> Adding (bin) env/_certificates/10000000001189cc2bbbe_client2.jks
>> Adding (bin) env/_certificates/DEV_TLServer_2048CAcert.truststore.jks
>> Adding (bin) env/_certificates/dgig.jks
>> Sending        src/com/vodafone/qa/common/httpclient.py
>> Sending        src/test.py
>>
>> Regards,
>> G.
>>

Re: FW: BUG :: Grinder 3.2b // SSLSocketFactory

2009-11-26T02:54:43Z

Re: [Grinder-use] FW: BUG :: Grinder 3.2b // SSLSocketFactory Hi Phil,
It looks like it does not.

Did u overrided the CA Trust mechanism? That’s the only way u could have achieved that...

Regards,
G.

Am 11/26/09 11:31 AM schrieb "Philip Aston" unter <philip.aston@...>:

You do not need to do this. The Grinder will automatically trust your server's certificate.

- Phil

Guido Serra wrote:

Re: [Grinder-use] FW: BUG :: Grinder 3.2b // SSLSocketFactory Hi Phil,
If ur target is using a certificate which is “home signed”, u need to import his CA.
Java mechanism to do this is the trustStore. We need to be able to override it.

Regards,
G.

Am 11/26/09 11:05 AM schrieb "Philip Aston" unter <philip.aston@...>:



The current implementation will not reject any connections on the basis of trust. There is no requirement to provide trust certificates.

Can you explain why you need to do so?

- Phil

Guido Serra wrote:


Re: [Grinder-use] FW: BUG :: Grinder 3.2b // SSLSocketFactory Hi Phil,
With the current set of API it is not possible to override the trustStore.

We are creating internally a patch, would u like to integrate it? Should we do it?

Regards,
G.

Am 11/9/09 10:38 PM schrieb "Philip Aston" unter <philip.aston@...>:




Serra, Guido, VF-Group wrote:
> If the server requires to present an SSL certificate as credentials and ... The server certificate is signed with a CA which is not in the SUN default TrustStore, u need to override the SSLSocketFactory or provide a keyStore and a trustStore from the jvm command line
>

Or you could just trust everything, which is what The Grinder does.
Refer to the URL I quote below for details.

- Phil

> The method to override this, implemented in the HTTPConnection class, doesn't seem to work properly. It relies on the default values
>
> Regards,
> G.
>
>
> www.vodafone.com <http://www.vodafone.com> <http://www.vodafone.com>
>
>
> Vodafone Group Services GmbH
> Mannesmannufer 2 ∙ D-40213 Düsseldorf
> Amtsgericht Düsseldorf, HRB 53554
> Geschäftsführung: Dr. Joachim Peters, Rainer Wallek
>
>
>
> This message and any files or documents attached are confidential and may also be legally privileged or protected by other legal rules. It is intended only for the individual or entity named. If you are not the named addressee or you have received this email in error, please inform the sender immediately, delete it from your system and do not copy or disclose it or its contents or use it for any purpose. Thank you. Please also note that transmission cannot be guaranteed to be secure or error-free.
>
> ----- Original Message -----
> From: Philip Aston <philip.aston@...>
> To: grinder-use <grinder-use@...>
> Sent: Mon Nov 09 20:44:51 2009
> Subject: Re: [Grinder-use] FW: BUG :: Grinder 3.2b // SSLSocketFactory
>
> What is the reason you need a custom SSLSocketFactory?
>
> By default, the grinder does not validate any certificates, it trusts
> everything. See http://grinder.sourceforge.net/g3/ssl-support.html
>
> - Phil
>
> Guido Serra wrote:
>
>> Hi All,
>> Should I open an official bug, or ... Can u have a look at ur code,
>> please?
>>
>> Regards,
>> G.
>>
>> +++++++++++++++++++++++++++++++++++++++
>>
>> Hi All,
>> There’s a bug in grinder (the grinder.sf.net) source code.
>>
>> XXXXX implemented a very nice feature to manage certificates, but it
>> doesn’t work due to the way grinder manages internally the
>> SSLSocketFactory. The code from the framework tries to override that
>> class instance.
>>
>> Internally grinder doesn’t keep the sslFactory overridden by our
>> framework.
>>
>> See lines 596,597 of file src/HTTPClient/HTTPConnection.java :
>>
>>    final SSLSocket socket = (SSLSocket)defaultSSLFactory.createSocket();
>>     sslProtocols = socket.getSupportedProtocols();
>>
>> And line 3099:
>>
>>    sock = sslFactory.createSocket(sock, Host, Port, true);
>>
>> What scares me is this loop/comment in grinder.sf.net source code...
>>
>> int try_count = 3;
>>   /* what a hack! This is to handle the case where the server closes
>>    * the connection but we don't realize it until we try to send
>>    * something. The problem is that we only get IOException, but
>>    * we need a finer specification (i.e. whether it's an EPIPE or
>>    * something else); I don't trust relying on the message part
>>    * of IOException (which on SunOS/Solaris gives 'Broken pipe',
>>    * but what on Windoze/Mac?).
>>    */
>>
>>   while (try_count-- > 0)
>>   {
>>       try
>>       {
>>     // get a client socket
>>
>>
>> I commented out the line were we override the SSLContext from jython code.
>> And set the certificates I trust at JVM level... I had to specify
>> directly the server certificate as trustStore.
>>
>> -Djavax.net.ssl.trustStore=env/_certificates/*dgig.jks* \
>> -Djavax.net.ssl.trustStorePassword=vod@fone \
>> -Djavax.net.ssl.keyStorePassword=importkey \
>> -Djavax.net.ssl.keyStore=env/_certificates/10000000001189cc2bbbe_client2.jks
>> \
>>
>> With the DEVVodafone CA, it didn’t worked.
>>
>> NOTE! How I noticed the sslContext overwritten?
>>
>> >From the httpclient.log ... The first POST worked... While other 3
>> POST failed.
>>
>> The createSocket() function is instantiated 4 times in the
>> HTTPConnection.java ... 3 times in the loop above, and once by the
>> (SSLSocket)defaultSSLFactory .... (see enclosed file)
>>
>> I committed my mods... On trunk/ , do not trust them till XXXXX
>> reviews the code, please.
>>
>> Authentication realm: <http://85.205.240.168:8080> Subversion Repository
>> Password for 'guido.serra':
>> Adding (bin) env/_certificates/10000000001189cc2bbbe_client2.jks
>> Adding (bin) env/_certificates/DEV_TLServer_2048CAcert.truststore.jks
>> Adding (bin) env/_certificates/dgig.jks
>> Sending        src/com/vodafone/qa/common/httpclient.py
>> Sending        src/test.py
>>
>> Regards,
>> G.
>>
>> ------------------------------------------------------------------------
>>
>> ------------------------------------------------------------------------------
>> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
>> trial. Simplify your report design, integration and deployment - and focus on
>> what you do best, core application coding. Discover what's new with
>> Crystal Reports now. http://p.sf.net/sfu/bobj-july
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> grinder-use mailing list
>> grinder-use@...
>> https://lists.sourceforge.net/lists/listinfo/grinder-use
>>
>>
>

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july

Re: FW: BUG :: Grinder 3.2b // SSLSocketFactory

2009-11-26T02:31:22Z

You do not need to do this. The Grinder will automatically trust your server's certificate.

- Phil

Guido Serra wrote:

Re: [Grinder-use] FW: BUG :: Grinder 3.2b // SSLSocketFactory Hi Phil,
If ur target is using a certificate which is “home signed”, u need to import his CA.
Java mechanism to do this is the trustStore. We need to be able to override it.

Regards,
G.

Am 11/26/09 11:05 AM schrieb "Philip Aston" unter <philip.aston@...>:

The current implementation will not reject any connections on the basis of trust. There is no requirement to provide trust certificates.

Can you explain why you need to do so?

- Phil

Guido Serra wrote:

Re: [Grinder-use] FW: BUG :: Grinder 3.2b // SSLSocketFactory Hi Phil,
With the current set of API it is not possible to override the trustStore.

We are creating internally a patch, would u like to integrate it? Should we do it?

Regards,
G.

Am 11/9/09 10:38 PM schrieb "Philip Aston" unter <philip.aston@...>:



Serra, Guido, VF-Group wrote:
> If the server requires to present an SSL certificate as credentials and ... The server certificate is signed with a CA which is not in the SUN default TrustStore, u need to override the SSLSocketFactory or provide a keyStore and a trustStore from the jvm command line
>

Or you could just trust everything, which is what The Grinder does.
Refer to the URL I quote below for details.

- Phil

> The method to override this, implemented in the HTTPConnection class, doesn't seem to work properly. It relies on the default values
>
> Regards,
> G.
>
>
> www.vodafone.com <http://www.vodafone.com>
>
>
> Vodafone Group Services GmbH
> Mannesmannufer 2 ∙ D-40213 Düsseldorf
> Amtsgericht Düsseldorf, HRB 53554
> Geschäftsführung: Dr. Joachim Peters, Rainer Wallek
>
>
>
> This message and any files or documents attached are confidential and may also be legally privileged or protected by other legal rules. It is intended only for the individual or entity named. If you are not the named addressee or you have received this email in error, please inform the sender immediately, delete it from your system and do not copy or disclose it or its contents or use it for any purpose. Thank you. Please also note that transmission cannot be guaranteed to be secure or error-free.
>
> ----- Original Message -----
> From: Philip Aston <philip.aston@...>
> To: grinder-use <grinder-use@...>
> Sent: Mon Nov 09 20:44:51 2009
> Subject: Re: [Grinder-use] FW: BUG :: Grinder 3.2b // SSLSocketFactory
>
> What is the reason you need a custom SSLSocketFactory?
>
> By default, the grinder does not validate any certificates, it trusts
> everything. See http://grinder.sourceforge.net/g3/ssl-support.html
>
> - Phil
>
> Guido Serra wrote:
>
>> Hi All,
>> Should I open an official bug, or ... Can u have a look at ur code,
>> please?
>>
>> Regards,
>> G.
>>
>> +++++++++++++++++++++++++++++++++++++++
>>
>> Hi All,
>> There’s a bug in grinder (the grinder.sf.net) source code.
>>
>> XXXXX implemented a very nice feature to manage certificates, but it
>> doesn’t work due to the way grinder manages internally the
>> SSLSocketFactory. The code from the framework tries to override that
>> class instance.
>>
>> Internally grinder doesn’t keep the sslFactory overridden by our
>> framework.
>>
>> See lines 596,597 of file src/HTTPClient/HTTPConnection.java :
>>
>>    final SSLSocket socket = (SSLSocket)defaultSSLFactory.createSocket();
>>     sslProtocols = socket.getSupportedProtocols();
>>
>> And line 3099:
>>
>>    sock = sslFactory.createSocket(sock, Host, Port, true);
>>
>> What scares me is this loop/comment in grinder.sf.net source code...
>>
>> int try_count = 3;
>>   /* what a hack! This is to handle the case where the server closes
>>    * the connection but we don't realize it until we try to send
>>    * something. The problem is that we only get IOException, but
>>    * we need a finer specification (i.e. whether it's an EPIPE or
>>    * something else); I don't trust relying on the message part
>>    * of IOException (which on SunOS/Solaris gives 'Broken pipe',
>>    * but what on Windoze/Mac?).
>>    */
>>
>>   while (try_count-- > 0)
>>   {
>>       try
>>       {
>>     // get a client socket
>>
>>
>> I commented out the line were we override the SSLContext from jython code.
>> And set the certificates I trust at JVM level... I had to specify
>> directly the server certificate as trustStore.
>>
>> -Djavax.net.ssl.trustStore=env/_certificates/*dgig.jks* \
>> -Djavax.net.ssl.trustStorePassword=vod@fone \
>> -Djavax.net.ssl.keyStorePassword=importkey \
>> -Djavax.net.ssl.keyStore=env/_certificates/10000000001189cc2bbbe_client2.jks
>> \
>>
>> With the DEVVodafone CA, it didn’t worked.
>>
>> NOTE! How I noticed the sslContext overwritten?
>>
>> >From the httpclient.log ... The first POST worked... While other 3
>> POST failed.
>>
>> The createSocket() function is instantiated 4 times in the
>> HTTPConnection.java ... 3 times in the loop above, and once by the
>> (SSLSocket)defaultSSLFactory .... (see enclosed file)
>>
>> I committed my mods... On trunk/ , do not trust them till XXXXX
>> reviews the code, please.
>>
>> Authentication realm: <http://85.205.240.168:8080> Subversion Repository
>> Password for 'guido.serra':
>> Adding (bin) env/_certificates/10000000001189cc2bbbe_client2.jks
>> Adding (bin) env/_certificates/DEV_TLServer_2048CAcert.truststore.jks
>> Adding (bin) env/_certificates/dgig.jks
>> Sending        src/com/vodafone/qa/common/httpclient.py
>> Sending        src/test.py
>>
>> Regards,
>> G.
>>
>> ------------------------------------------------------------------------
>>
>> ------------------------------------------------------------------------------
>> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
>> trial. Simplify your report design, integration and deployment - and focus on
>> what you do best, core application coding. Discover what's new with
>> Crystal Reports now. http://p.sf.net/sfu/bobj-july
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> grinder-use mailing list
>> grinder-use@...
>> https://lists.sourceforge.net/lists/listinfo/grinder-use
>>
>>
>

Re: FW: BUG :: Grinder 3.2b // SSLSocketFactory

2009-11-26T02:27:30Z

Re: [Grinder-use] FW: BUG :: Grinder 3.2b // SSLSocketFactory Hi Phil,
If ur target is using a certificate which is “home signed”, u need to import his CA.
Java mechanism to do this is the trustStore. We need to be able to override it.

Regards,
G.

Am 11/26/09 11:05 AM schrieb "Philip Aston" unter <philip.aston@...>:

The current implementation will not reject any connections on the basis of trust. There is no requirement to provide trust certificates.

Can you explain why you need to do so?

- Phil

Guido Serra wrote:

Re: [Grinder-use] FW: BUG :: Grinder 3.2b // SSLSocketFactory Hi Phil,
With the current set of API it is not possible to override the trustStore.

We are creating internally a patch, would u like to integrate it? Should we do it?

Regards,
G.

Am 11/9/09 10:38 PM schrieb "Philip Aston" unter <philip.aston@...>:



Serra, Guido, VF-Group wrote:
> If the server requires to present an SSL certificate as credentials and ... The server certificate is signed with a CA which is not in the SUN default TrustStore, u need to override the SSLSocketFactory or provide a keyStore and a trustStore from the jvm command line
>

Or you could just trust everything, which is what The Grinder does.
Refer to the URL I quote below for details.

- Phil

> The method to override this, implemented in the HTTPConnection class, doesn't seem to work properly. It relies on the default values
>
> Regards,
> G.
>
>
> www.vodafone.com <http://www.vodafone.com>
>
>
> Vodafone Group Services GmbH
> Mannesmannufer 2 ∙ D-40213 Düsseldorf
> Amtsgericht Düsseldorf, HRB 53554
> Geschäftsführung: Dr. Joachim Peters, Rainer Wallek
>
>
>
> This message and any files or documents attached are confidential and may also be legally privileged or protected by other legal rules. It is intended only for the individual or entity named. If you are not the named addressee or you have received this email in error, please inform the sender immediately, delete it from your system and do not copy or disclose it or its contents or use it for any purpose. Thank you. Please also note that transmission cannot be guaranteed to be secure or error-free.
>
> ----- Original Message -----
> From: Philip Aston <philip.aston@...>
> To: grinder-use <grinder-use@...>
> Sent: Mon Nov 09 20:44:51 2009
> Subject: Re: [Grinder-use] FW: BUG :: Grinder 3.2b // SSLSocketFactory
>
> What is the reason you need a custom SSLSocketFactory?
>
> By default, the grinder does not validate any certificates, it trusts
> everything. See http://grinder.sourceforge.net/g3/ssl-support.html
>
> - Phil
>
> Guido Serra wrote:
>
>> Hi All,
>> Should I open an official bug, or ... Can u have a look at ur code,
>> please?
>>
>> Regards,
>> G.
>>
>> +++++++++++++++++++++++++++++++++++++++
>>
>> Hi All,
>> There’s a bug in grinder (the grinder.sf.net) source code.
>>
>> XXXXX implemented a very nice feature to manage certificates, but it
>> doesn’t work due to the way grinder manages internally the
>> SSLSocketFactory. The code from the framework tries to override that
>> class instance.
>>
>> Internally grinder doesn’t keep the sslFactory overridden by our
>> framework.
>>
>> See lines 596,597 of file src/HTTPClient/HTTPConnection.java :
>>
>>    final SSLSocket socket = (SSLSocket)defaultSSLFactory.createSocket();
>>     sslProtocols = socket.getSupportedProtocols();
>>
>> And line 3099:
>>
>>    sock = sslFactory.createSocket(sock, Host, Port, true);
>>
>> What scares me is this loop/comment in grinder.sf.net source code...
>>
>> int try_count = 3;
>>   /* what a hack! This is to handle the case where the server closes
>>    * the connection but we don't realize it until we try to send
>>    * something. The problem is that we only get IOException, but
>>    * we need a finer specification (i.e. whether it's an EPIPE or
>>    * something else); I don't trust relying on the message part
>>    * of IOException (which on SunOS/Solaris gives 'Broken pipe',
>>    * but what on Windoze/Mac?).
>>    */
>>
>>   while (try_count-- > 0)
>>   {
>>       try
>>       {
>>     // get a client socket
>>
>>
>> I commented out the line were we override the SSLContext from jython code.
>> And set the certificates I trust at JVM level... I had to specify
>> directly the server certificate as trustStore.
>>
>> -Djavax.net.ssl.trustStore=env/_certificates/*dgig.jks* \
>> -Djavax.net.ssl.trustStorePassword=vod@fone \
>> -Djavax.net.ssl.keyStorePassword=importkey \
>> -Djavax.net.ssl.keyStore=env/_certificates/10000000001189cc2bbbe_client2.jks
>> \
>>
>> With the DEVVodafone CA, it didn’t worked.
>>
>> NOTE! How I noticed the sslContext overwritten?
>>
>> >From the httpclient.log ... The first POST worked... While other 3
>> POST failed.
>>
>> The createSocket() function is instantiated 4 times in the
>> HTTPConnection.java ... 3 times in the loop above, and once by the
>> (SSLSocket)defaultSSLFactory .... (see enclosed file)
>>
>> I committed my mods... On trunk/ , do not trust them till XXXXX
>> reviews the code, please.
>>
>> Authentication realm: <http://85.205.240.168:8080> Subversion Repository
>> Password for 'guido.serra':
>> Adding (bin) env/_certificates/10000000001189cc2bbbe_client2.jks
>> Adding (bin) env/_certificates/DEV_TLServer_2048CAcert.truststore.jks
>> Adding (bin) env/_certificates/dgig.jks
>> Sending        src/com/vodafone/qa/common/httpclient.py
>> Sending        src/test.py
>>
>> Regards,
>> G.
>>
>> ------------------------------------------------------------------------
>>
>> ------------------------------------------------------------------------------
>> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
>> trial. Simplify your report design, integration and deployment - and focus on
>> what you do best, core application coding. Discover what's new with
>> Crystal Reports now. http://p.sf.net/sfu/bobj-july
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> grinder-use mailing list
>> grinder-use@...
>> https://lists.sourceforge.net/lists/listinfo/grinder-use
>>
>>
>
>

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july

_______________________________________________
grinder-use mailing list
grinder-use@...
https://lists.sourceforge.net/lists/listinfo/grinder-use

Re: FW: BUG :: Grinder 3.2b // SSLSocketFactory

2009-11-26T02:05:47Z

The current implementation will not reject any connections on the basis of trust. There is no requirement to provide trust certificates.

Can you explain why you need to do so?

- Phil

Guido Serra wrote:

Re: [Grinder-use] FW: BUG :: Grinder 3.2b // SSLSocketFactory Hi Phil,
With the current set of API it is not possible to override the trustStore.

We are creating internally a patch, would u like to integrate it? Should we do it?

Regards,
G.

Am 11/9/09 10:38 PM schrieb "Philip Aston" unter <philip.aston@...>:

Serra, Guido, VF-Group wrote:
> If the server requires to present an SSL certificate as credentials and ... The server certificate is signed with a CA which is not in the SUN default TrustStore, u need to override the SSLSocketFactory or provide a keyStore and a trustStore from the jvm command line
>

Or you could just trust everything, which is what The Grinder does.
Refer to the URL I quote below for details.

- Phil

> The method to override this, implemented in the HTTPConnection class, doesn't seem to work properly. It relies on the default values
>
> Regards,
> G.
>
>
> www.vodafone.com
>
>
> Vodafone Group Services GmbH
> Mannesmannufer 2 ∙ D-40213 Düsseldorf
> Amtsgericht Düsseldorf, HRB 53554
> Geschäftsführung: Dr. Joachim Peters, Rainer Wallek
>
>
>
> This message and any files or documents attached are confidential and may also be legally privileged or protected by other legal rules. It is intended only for the individual or entity named. If you are not the named addressee or you have received this email in error, please inform the sender immediately, delete it from your system and do not copy or disclose it or its contents or use it for any purpose. Thank you. Please also note that transmission cannot be guaranteed to be secure or error-free.
>
> ----- Original Message -----
> From: Philip Aston <philip.aston@...>
> To: grinder-use <grinder-use@...>
> Sent: Mon Nov 09 20:44:51 2009
> Subject: Re: [Grinder-use] FW: BUG :: Grinder 3.2b // SSLSocketFactory
>
> What is the reason you need a custom SSLSocketFactory?
>
> By default, the grinder does not validate any certificates, it trusts
> everything. See http://grinder.sourceforge.net/g3/ssl-support.html
>
> - Phil
>
> Guido Serra wrote:
>
>> Hi All,
>> Should I open an official bug, or ... Can u have a look at ur code,
>> please?
>>
>> Regards,
>> G.
>>
>> +++++++++++++++++++++++++++++++++++++++
>>
>> Hi All,
>> There’s a bug in grinder (the grinder.sf.net) source code.
>>
>> XXXXX implemented a very nice feature to manage certificates, but it
>> doesn’t work due to the way grinder manages internally the
>> SSLSocketFactory. The code from the framework tries to override that
>> class instance.
>>
>> Internally grinder doesn’t keep the sslFactory overridden by our
>> framework.
>>
>> See lines 596,597 of file src/HTTPClient/HTTPConnection.java :
>>
>>    final SSLSocket socket = (SSLSocket)defaultSSLFactory.createSocket();
>>     sslProtocols = socket.getSupportedProtocols();
>>
>> And line 3099:
>>
>>    sock = sslFactory.createSocket(sock, Host, Port, true);
>>
>> What scares me is this loop/comment in grinder.sf.net source code...
>>
>> int try_count = 3;
>>   /* what a hack! This is to handle the case where the server closes
>>    * the connection but we don't realize it until we try to send
>>    * something. The problem is that we only get IOException, but
>>    * we need a finer specification (i.e. whether it's an EPIPE or
>>    * something else); I don't trust relying on the message part
>>    * of IOException (which on SunOS/Solaris gives 'Broken pipe',
>>    * but what on Windoze/Mac?).
>>    */
>>
>>   while (try_count-- > 0)
>>   {
>>       try
>>       {
>>     // get a client socket
>>
>>
>> I commented out the line were we override the SSLContext from jython code.
>> And set the certificates I trust at JVM level... I had to specify
>> directly the server certificate as trustStore.
>>
>> -Djavax.net.ssl.trustStore=env/_certificates/*dgig.jks* \
>> -Djavax.net.ssl.trustStorePassword=vod@fone \
>> -Djavax.net.ssl.keyStorePassword=importkey \
>> -Djavax.net.ssl.keyStore=env/_certificates/10000000001189cc2bbbe_client2.jks
>> \
>>
>> With the DEVVodafone CA, it didn’t worked.
>>
>> NOTE! How I noticed the sslContext overwritten?
>>
>> >From the httpclient.log ... The first POST worked... While other 3
>> POST failed.
>>
>> The createSocket() function is instantiated 4 times in the
>> HTTPConnection.java ... 3 times in the loop above, and once by the
>> (SSLSocket)defaultSSLFactory .... (see enclosed file)
>>
>> I committed my mods... On trunk/ , do not trust them till XXXXX
>> reviews the code, please.
>>
>> Authentication realm: <http://85.205.240.168:8080> Subversion Repository
>> Password for 'guido.serra':
>> Adding (bin) env/_certificates/10000000001189cc2bbbe_client2.jks
>> Adding (bin) env/_certificates/DEV_TLServer_2048CAcert.truststore.jks
>> Adding (bin) env/_certificates/dgig.jks
>> Sending        src/com/vodafone/qa/common/httpclient.py
>> Sending        src/test.py
>>
>> Regards,
>> G.
>>
>> ------------------------------------------------------------------------
>>
>> ------------------------------------------------------------------------------
>> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
>> trial. Simplify your report design, integration and deployment - and focus on
>> what you do best, core application coding. Discover what's new with
>> Crystal Reports now. http://p.sf.net/sfu/bobj-july
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> grinder-use mailing list
>> grinder-use@...
>> https://lists.sourceforge.net/lists/listinfo/grinder-use
>>
>>
>
>

Re: FW: BUG :: Grinder 3.2b // SSLSocketFactory

2009-11-25T08:58:08Z

Re: [Grinder-use] FW: BUG :: Grinder 3.2b // SSLSocketFactory Hi Phil,
With the current set of API it is not possible to override the trustStore.

We are creating internally a patch, would u like to integrate it? Should we do it?

Regards,
G.

Am 11/9/09 10:38 PM schrieb "Philip Aston" unter <philip.aston@...>:

Serra, Guido, VF-Group wrote:
> If the server requires to present an SSL certificate as credentials and ... The server certificate is signed with a CA which is not in the SUN default TrustStore, u need to override the SSLSocketFactory or provide a keyStore and a trustStore from the jvm command line
>

Or you could just trust everything, which is what The Grinder does.
Refer to the URL I quote below for details.

- Phil

> The method to override this, implemented in the HTTPConnection class, doesn't seem to work properly. It relies on the default values
>
> Regards,
> G.
>
>
> www.vodafone.com
>
>
> Vodafone Group Services GmbH
> Mannesmannufer 2 ∙ D-40213 Düsseldorf
> Amtsgericht Düsseldorf, HRB 53554
> Geschäftsführung: Dr. Joachim Peters, Rainer Wallek
>
>
>
> This message and any files or documents attached are confidential and may also be legally privileged or protected by other legal rules. It is intended only for the individual or entity named. If you are not the named addressee or you have received this email in error, please inform the sender immediately, delete it from your system and do not copy or disclose it or its contents or use it for any purpose. Thank you. Please also note that transmission cannot be guaranteed to be secure or error-free.
>
> ----- Original Message -----
> From: Philip Aston <philip.aston@...>
> To: grinder-use <grinder-use@...>
> Sent: Mon Nov 09 20:44:51 2009
> Subject: Re: [Grinder-use] FW: BUG :: Grinder 3.2b // SSLSocketFactory
>
> What is the reason you need a custom SSLSocketFactory?
>
> By default, the grinder does not validate any certificates, it trusts
> everything. See http://grinder.sourceforge.net/g3/ssl-support.html
>
> - Phil
>
> Guido Serra wrote:
>
>> Hi All,
>> Should I open an official bug, or ... Can u have a look at ur code,
>> please?
>>
>> Regards,
>> G.
>>
>> +++++++++++++++++++++++++++++++++++++++
>>
>> Hi All,
>> There’s a bug in grinder (the grinder.sf.net) source code.
>>
>> XXXXX implemented a very nice feature to manage certificates, but it
>> doesn’t work due to the way grinder manages internally the
>> SSLSocketFactory. The code from the framework tries to override that
>> class instance.
>>
>> Internally grinder doesn’t keep the sslFactory overridden by our
>> framework.
>>
>> See lines 596,597 of file src/HTTPClient/HTTPConnection.java :
>>
>>    final SSLSocket socket = (SSLSocket)defaultSSLFactory.createSocket();
>>     sslProtocols = socket.getSupportedProtocols();
>>
>> And line 3099:
>>
>>    sock = sslFactory.createSocket(sock, Host, Port, true);
>>
>> What scares me is this loop/comment in grinder.sf.net source code...
>>
>> int try_count = 3;
>>   /* what a hack! This is to handle the case where the server closes
>>    * the connection but we don't realize it until we try to send
>>    * something. The problem is that we only get IOException, but
>>    * we need a finer specification (i.e. whether it's an EPIPE or
>>    * something else); I don't trust relying on the message part
>>    * of IOException (which on SunOS/Solaris gives 'Broken pipe',
>>    * but what on Windoze/Mac?).
>>    */
>>
>>   while (try_count-- > 0)
>>   {
>>       try
>>       {
>>     // get a client socket
>>
>>
>> I commented out the line were we override the SSLContext from jython code.
>> And set the certificates I trust at JVM level... I had to specify
>> directly the server certificate as trustStore.
>>
>> -Djavax.net.ssl.trustStore=env/_certificates/*dgig.jks* \
>> -Djavax.net.ssl.trustStorePassword=vod@fone \
>> -Djavax.net.ssl.keyStorePassword=importkey \
>> -Djavax.net.ssl.keyStore=env/_certificates/10000000001189cc2bbbe_client2.jks
>> \
>>
>> With the DEVVodafone CA, it didn’t worked.
>>
>> NOTE! How I noticed the sslContext overwritten?
>>
>> >From the httpclient.log ... The first POST worked... While other 3
>> POST failed.
>>
>> The createSocket() function is instantiated 4 times in the
>> HTTPConnection.java ... 3 times in the loop above, and once by the
>> (SSLSocket)defaultSSLFactory .... (see enclosed file)
>>
>> I committed my mods... On trunk/ , do not trust them till XXXXX
>> reviews the code, please.
>>
>> Authentication realm: <http://85.205.240.168:8080> Subversion Repository
>> Password for 'guido.serra':
>> Adding (bin) env/_certificates/10000000001189cc2bbbe_client2.jks
>> Adding (bin) env/_certificates/DEV_TLServer_2048CAcert.truststore.jks
>> Adding (bin) env/_certificates/dgig.jks
>> Sending        src/com/vodafone/qa/common/httpclient.py
>> Sending        src/test.py
>>
>> Regards,
>> G.
>>
>> ------------------------------------------------------------------------
>>
>> ------------------------------------------------------------------------------
>> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
>> trial. Simplify your report design, integration and deployment - and focus on
>> what you do best, core application coding. Discover what's new with
>> Crystal Reports now. http://p.sf.net/sfu/bobj-july
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> grinder-use mailing list
>> grinder-use@...
>> https://lists.sourceforge.net/lists/listinfo/grinder-use
>>
>>
>
>
>
>
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
> trial. Simplify your report design, integration and deployment - and focus on
> what you do best, core application coding. Discover what's new with
> Crystal Reports now. http://p.sf.net/sfu/bobj-july
> _______________________________________________
> grinder-use mailing list
> grinder-use@...
> https://lists.sourceforge.net/lists/listinfo/grinder-use
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
> trial. Simplify your report design, integration and deployment - and focus on
> what you do best, core application coding. Discover what's new with
> Crystal Reports now. http://p.sf.net/sfu/bobj-july
> _______________________________________________
> grinder-use mailing list
> grinder-use@...
> https://lists.sourceforge.net/lists/listinfo/grinder-use
>

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
_______________________________________________
grinder-use mailing list
grinder-use@...
https://lists.sourceforge.net/lists/listinfo/grinder-use

Re: Filtering while recording

2009-11-19T22:24:17Z

anil_jacob@... wrote:
> Is there an option to use filters while recording a script using the
> TCPProxy with HTTP option.
>
> For e.g. if I don't want to capture HTTP requests to URLs other than
> my own internal hosts in the script.
>
> Anil

Nothing out of the box, but if you're comfortable with XSLT, you can
pull some tricks.

E.g. you can request the TCPProxy to produce an XML file using
etc/httpToXML.xsl, process that XML file with a custom stylesheet to
filter what you don't want, then apply the standard style sheet
(etc/httpToJythonScript.xsl). Or you can just create your own stylesheet
that does exactly what you want.

- Phil

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
_______________________________________________
grinder-use mailing list
grinder-use@...
https://lists.sourceforge.net/lists/listinfo/grinder-use

Filtering while recording

2009-11-19T17:24:00Z

Is there an option to use filters while recording a script using the TCPProxy with HTTP option.

For e.g. if I don't want to capture HTTP requests to URLs other than my own internal hosts in the script.

Anil

Re: Can we achieve both Auto Cookie and Manual Cookie Handling ...?

2009-11-18T13:28:22Z

kamalakar vadla wrote:
> Hi all,
>
> I am unable to get handle on Set-Cookie header if the auto cookie is
> enabled.
> But my requirement cookie need to be sent as apart of each request as
> param and at the same time cookie need be sent in the headers. This
> can be achieved by handling cookies manually . But i don't want to
> handle cookies manually..? is their any other possible solution to
> this problem..??

Add a cookie handler:
http://grinder.sourceforge.net/g3/script-gallery.html#cookies

- Phil

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
_______________________________________________
grinder-use mailing list
grinder-use@...
https://lists.sourceforge.net/lists/listinfo/grinder-use

Can we achieve both Auto Cookie and Manual Cookie Handling ...?

2009-11-18T03:48:55Z

Hi all,

I am unable to get handle on Set-Cookie header if the auto cookie is enabled.
But my requirement cookie need to be sent as apart of each request as param and at the same time cookie need be sent in the headers. This can be achieved by handling cookies manually . But i don't want to handle cookies manually..? is their any other possible solution to this problem..??

Thanks
Kamalakar

--
Kamalakar.Vadla
Pramati Technologies Pvt Ltd
Begumpet

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
trial. Simplify your report design, integration and deployment - and focus on
what you do best, core application coding. Discover what's new with
Crystal Reports now. http://p.sf.net/sfu/bobj-july
_______________________________________________
grinder-use mailing list
grinder-use@...
https://lists.sourceforge.net/lists/listinfo/grinder-use

Re: grinder-use Digest, Vol 42, Issue 9 (Running multiple scripts in sequence (or parallel))

2009-11-17T23:11:52Z

The % is the modulo function. This might help you understand the "poor
man's" approach:

>>> for x in range(1,20):
... print "%d %% 3 = %d" % (x, x%3)
...
1 % 3 = 1
2 % 3 = 2
3 % 3 = 0
4 % 3 = 1
5 % 3 = 2
6 % 3 = 0
7 % 3 = 1
8 % 3 = 2
9 % 3 = 0
10 % 3 = 1
11 % 3 = 2
12 % 3 = 0
13 % 3 = 1
14 % 3 = 2
15 % 3 = 0
16 % 3 = 1
17 % 3 = 2
18 % 3 = 0
19 % 3 = 1

Walt Tuvell wrote:

> Tom, this construct ("tid%N==M") determines which script each thread is
> going to run, depending on the thread's ID number. In other words, this
> speaks to the parallel (as opposed to sequence) running of tests.
>
> The "tid%N==M" technique is a "poor man's" way to distribute the work
> amongst the threads. Among other things, it limits each thread to
> running only one kind of test, and your distribution will vary depending
> on the relationship between the number of tests and the number of
> threads.
>
> But there's a much better way: The "rich man's" way to distribute the
> work is documented at the very bottom of the Grinder Complete User
> Guide, in the section on Weighted Distribution of Tests. That should
> respond to your requirements.
>
> - Walt
>
>
>
> On Tue, 2009-11-17 at 23:00 +0000, Tom Johnson wrote:
>
>> Hi again,
>>
>>
>>
>> OK, after working through the sequence script, the only portion of
>> the parallel script I am not grasping is the following portion:
>>
>>
>>
>>
>> tid = grinder.threadNumber
>>
>> if tid % 4 == 2:
>> self.testRunner = createTestRunner(scripts[1])
>> elif tid % 4 == 3:
>> self.testRunner = createTestRunner(scripts[2])
>> else:
>> self.testRunner = createTestRunner(scripts[0])
>>
>>
>>
>>
>> I understand that the tid = grinder.threadNumber assigns the thread
>> number to the value tid but I do not understand the if tid % 4 == 2:
>> along with the elif and else lines. I will be running 100 processes
>> with 20 threads in each with 14 scripts all needing to be run in
>> parallel.
>>
>>
>>
>> Tom
>>
>>
>>

Re: grinder-use Digest, Vol 42, Issue 9 (Running multiple scripts in sequence (or parallel))

2009-11-17T23:07:47Z

Ahh, now we can see what's going on!

sequence.py is working fine, but Aboutus.py is not. You will find you
can't even run it standalone. (I'd definitely recommend running each
script standalone before combining them with sequence.py).

The problem is that a class or method in the file is just too large. It
exceeds a Java restriction, inherited by Jython. Unfortunately, there is
little that can be done, other than to manually split the long part of
the script up into smaller pieces. The scripts recorded with the
TCPProxy usually do not exceed this limit, but it depends on how long
you record for.

- Phil

Tom Johnson wrote:

>
> Hi Phil,
>
>
>
> Before I go any further, THANK YOU for being so patient with me and
> helping me to understand how to do it correct. With that said........
>
> No need to be sorry, I figured it out when I tried it without the
> astricks. I think it was a good idea to try and bold it also.
>
>
>
> Yes, I do have the Aboutus in the c:\grinder directory. I am about
> frustrated enough to start populating in several places to see if I am
> having problems with location.
>
>
>
> Yes, an error file was also created and here is a copy of it:
>
>
>
> 11/17/09 3:21:14 PM (process tjohnson2-0): Error running worker
> process (Java exception initialising test script
> File "<string>", line 1, in ?
> File "C:\grinder\sequence.py", line 18, in ?)
> Java exception initialising test script
> File "<string>", line 1, in ?
> File "C:\grinder\sequence.py", line 18, in ?
> Caused by: java.lang.ClassFormatError: Invalid method Code length
> 79266 in class file Aboutus$py
> at java.lang.ClassLoader.defineClass1(Native Method)
> at java.lang.ClassLoader.defineClass(ClassLoader.java:620)
> at org.python.core.BytecodeLoader2.loadClassFromBytes(Unknown Source)
> at org.python.core.BytecodeLoader.makeClass(Unknown Source)
> at org.python.core.BytecodeLoader.makeCode(Unknown Source)
> at org.python.core.imp.createFromSource(Unknown Source)
> at org.python.core.imp.createFromSource(Unknown Source)
> at org.python.core.imp.loadFromSource(Unknown Source)
> at org.python.core.imp.find_module(Unknown Source)
> at org.python.core.imp.import_next(Unknown Source)
> at org.python.core.imp.import_name(Unknown Source)
> at org.python.core.imp.importName(Unknown Source)
> at org.python.core.ImportFunction.load(Unknown Source)
> at org.python.core.ImportFunction.__call__(Unknown Source)
> at org.python.core.PyObject.__call__(Unknown Source)
> at org.python.core.__builtin__.__import__(Unknown Source)
> at org.python.core.imp.importOne(Unknown Source)
> at org.python.pycode._pyx1.f$0(<string>:1)
> at org.python.pycode._pyx1.call_function(<string>)
> at org.python.core.PyTableCode.call(Unknown Source)
> at org.python.core.PyCode.call(Unknown Source)
> at org.python.core.Py.runCode(Unknown Source)
> at org.python.core.Py.exec(Unknown Source)
> at org.python.pycode._pyx0.f$0(C:\grinder\sequence.py:18)
> at org.python.pycode._pyx0.call_function(C:\grinder\sequence.py)
> at org.python.core.PyTableCode.call(Unknown Source)
> at org.python.core.PyCode.call(Unknown Source)
> at org.python.core.Py.runCode(Unknown Source)
> at org.python.core.__builtin__.execfile_flags(Unknown Source)
> at org.python.util.PythonInterpreter.execfile(Unknown Source)
> at
> net.grinder.engine.process.jython.JythonScriptEngine.initialise(JythonScriptEngine.java:96)
> at net.grinder.engine.process.GrinderProcess.run(GrinderProcess.java:234)
> at
> net.grinder.engine.process.WorkerProcessEntryPoint.run(WorkerProcessEntryPoint.java:87)
> at
> net.grinder.engine.process.WorkerProcessEntryPoint.main(WorkerProcessEntryPoint.java:59)
>
>
>
> I looked back at the other attempts because this was the first time I
> noticed an exception error listing and I see that it was in previous
> error logs also but not all.
>
>
>
> I did run the Aboutus.py script individually and it does work without
> errors but will run it again to make sure I did not miss anything.
>
>
>
> Tom
>
>
>
>
>
> ---- Phil wrote:
>
>
>
> Sorry, the asterisks were because I'd made it bold in my email :-)
>
> Taking this at face value, line 18 would be exec'ing "import Aboutus".
>
> Is Aboutus.py in the c:\grinder directory?
>
> Is there an error file create? If so, can you post it?
>
> - Phil
>

Re: grinder-use Digest, Vol 42, Issue 9 (Running multiple scripts in sequence (or parallel))

2009-11-17T17:06:18Z

Tom, this construct ("tid%N==M") determines which script each thread is
going to run, depending on the thread's ID number. In other words, this
speaks to the parallel (as opposed to sequence) running of tests.

The "tid%N==M" technique is a "poor man's" way to distribute the work
amongst the threads. Among other things, it limits each thread to
running only one kind of test, and your distribution will vary depending
on the relationship between the number of tests and the number of
threads.

But there's a much better way: The "rich man's" way to distribute the
work is documented at the very bottom of the Grinder Complete User
Guide, in the section on Weighted Distribution of Tests. That should
respond to your requirements.

- Walt

On Tue, 2009-11-17 at 23:00 +0000, Tom Johnson wrote:

> Hi again,
>
>
>
> OK, after working through the sequence script, the only portion of
> the parallel script I am not grasping is the following portion:
>
>
>
>
> tid = grinder.threadNumber
>
> if tid % 4 == 2:
> self.testRunner = createTestRunner(scripts[1])
> elif tid % 4 == 3:
> self.testRunner = createTestRunner(scripts[2])
> else:
> self.testRunner = createTestRunner(scripts[0])
>
>
>
>
> I understand that the tid = grinder.threadNumber assigns the thread
> number to the value tid but I do not understand the if tid % 4 == 2:
> along with the elif and else lines. I will be running 100 processes
> with 20 threads in each with 14 scripts all needing to be run in
> parallel.
>
>
>
> Tom
>
>
>
>
>
> ------------------------------------------------------------------------------
> Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day
> trial. Simplify your report design, integration and deployment - and focus on
> what you do best, core application coding. Discover what's new with
> Crystal Reports now. http://p.sf.net/sfu/bobj-july
> _______________________________________________ grinder-use mailing list grinder-use@... https://lists.sourceforge.net/lists/listinfo/grinder-use

Re: grinder-use Digest, Vol 42, Issue 9 (Running multiple scripts in sequence (or parallel))

2009-11-17T15:00:19Z

Hi again,

OK, after working through the sequence script, the only portion of the parallel script I am not grasping is the following portion:

        tid = grinder.threadNumber

        if tid % 4 == 2:
            self.testRunner = createTestRunner(scripts[1])
        elif tid % 4 == 3:
            self.testRunner = createTestRunner(scripts[2])
        else:
            self.testRunner = createTestRunner(scripts[0])

I understand that the tid = grinder.threadNumber assigns the thread number to the value tid but I do not understand the if tid % 4 == 2: along with the elif and else lines. I will be running 100 processes with 20 threads in each with 14 scripts all needing to be run in parallel.

Tom

Re: grinder-use Digest, Vol 42, Issue 9 (Running multiple scripts in sequence (or parallel))

2009-11-17T14:48:47Z

Hi Phil,

Well, very interesting. Out of 20 of my scripts, I now have 7 that are not running for some reason. I went through and tried each one individually and found the 7 that do not work anymore. I trimed my list down to just the ones that are still working and tried the sequence script again. Wouldn't you know it, the script is now working as advertised for only a couple of threads and processes. Will be trying for more shortly. Now to determine how the 7 got corupted to the point of not running.

Also, now to get the parallel script working.

Tom