I've already accomplished what I need.  I was curious if anyone had an idea
for a more elegant solution.
The constant throughput sampler is similar, but not the right solution for
this case.  I did get some ideas from the throughput sampler source code.

On Tue, Jun 2, 2009 at 9:29 AM, sebb <[email protected]> wrote:

> In that case, the Constant Throughput Timer is probably what you want.
>
> See also:
>
>
> http://jakarta.apache.org/jmeter/usermanual/best-practices.html#beanshell_server
>
> for an example of how to use this to vary the throughput at run-time.
>
> You can add a
>
>
> http://jakarta.apache.org/jmeter/usermanual/component_reference.html#Duration_Assertion
>
> to mark overlong responses as failed.
>
> On 02/06/2009, Tony Lotts <[email protected]> wrote:
> > This is for validating how many users a single webservice app server can
> >  support within a response time requirement, and the rate at which those
> >  user's can send transactions.
> >
> >
> >
> >  On Mon, Jun 1, 2009 at 9:03 PM, sebb <[email protected]> wrote:
> >
> >  > On 01/06/2009, Tony Lotts <[email protected]> wrote:
> >  > >  I have created a timer which varies the delay based upon whether or
> not
> >  > the
> >  > >  response time of the previous transaction is > 2999 milliseconds.
> >  >
> >  > What are you trying to achieve here?
> >  >
> >  > >
> >  > >  The solution that I have uses a beanshell timer, beanshell
> >  > postprocessor,
> >  > >  and beanshell assertion.
> >  > >
> >  > >  The beanshell timer returns the delay property set by the beanshell
> >  > >  postprocessor.
> >  > >
> >  > >  The beanshell postprocessor evaluates the response time, and sets
> the
> >  > delay
> >  > >  property.
> >  > >
> >  > >  The beanshell assertion sets the delay to the assertion failure
> message.
> >  > >
> >  > >
> >  > >  Does anyone have any ideas for a more elegant solution?
> >  > >
> >  >
> >  > If you want to delay after the sampler, then you can do it in the
> >  > BeanShell post-processor.
> >  >
> >
> > > ---------------------------------------------------------------------
> >  > To unsubscribe, e-mail: [email protected]
> >  > For additional commands, e-mail: [email protected]
> >  >
> >  >
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: [email protected]
> For additional commands, e-mail: [email protected]
>
>

Reply via email to