jmeter-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Deepak Shetty <shet...@gmail.com>
Subject Re: Dynamically changing sampler name during test
Date Fri, 20 Apr 2012 17:55:06 GMT
Hi
 i think this works
a. extract out the comment using regex post processor or equivalent
b. Add a BeanShell Post Processor  as child of the sampler write code like
prev.setSampleLabel(vars.get("variablenamefromstepa")); => replace the name
with the value you extracted

An alternative is
If you are processing the results later then you extract the comment to a
variable . you use sample_variables to write this variable to the result
file and you process the result file to deal with this extra value .

regards
deepak

On Fri, Apr 20, 2012 at 10:44 AM, Marcelo Jara <marcelojara@hotmail.com>wrote:

>
> Hello! I have a unique (I think) situation here. I'll try to explain it as
> clearly as possible. This may not be doable in Jmeter, but here is the
> requirement:
>
> I have a website and am using the web logs to run performance tests. The
> weblogs will contain all the different URLs that customers would normally
> hit. These can be searches, product pages, home page hits, etc. The URL of
> the request will not tell me what page I'm going to since a user can be
> redirected to other pages depending on the state of the user and system
> (data is always changing). One thing all these pages have in common is a
> comment in the source which describes what page I am at.
>
> Example: http://www.mysite.com/c/this/is/an/example/url  can be sent to a
> search results page. So the comment on that page will be "page:
> Search_Results".
> Example: http://www.mysite.com/c/this/is/another/example/url can be
> redirected to the home page because an error may have occurred in some
> backend system. The comment on source will be "page: Home_Page".
>
>
> The problem I'm running into right now is that since I only have one
> sampler which is randomly choosing a URL to hit, the reports will only have
> the sampler name. What I'm looking for is an aggregate report that would
> list the data broken down my the different page types.
>
>
> I'm currently using LoadRunner for this which we're trying to get rid of.
> For those with LoadRunner, I'm achieving this by putting a timer around the
> request, then parsing out the page name, and the using a built-in function
> called lr_set_transaction(name,duration) where I can pass the page name and
> response time.
>
> I hope this makes sense.
>
>
> Thanks,
>
> MJ
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message