RE: [bcc][faked-from] RE: comparing results

["Krahe, Chris" <Ch...@aquilent.com>]

Keith-

I'd use it if the listeners could read from the database.  Specifically, I'd want the listener to accept a select statement or stored procedure call that I type in :)

And I'm sure you already know this, but if done at run time it would have to be _real_ efficient, especially in the multi-client scenario.

Personally, I don't think I'd mind having a tool that could bulk-import results into the database after a test run.  The tool would be smart enough to warn me if I was trying to import the same file twice.  It would also let me delete database contents for a selected file.

-Chris

Chris Krahe
Systems Architect
Aquilent, Inc (ack-wil-lent)
http://www.aquilent.com/


-----Original Message-----
From: Lancaster, Keith [mailto:klancaster@noblestar.com]
Sent: Friday, January 09, 2004 10:15 AM
To: JMeter Users List
Subject: [bcc][faked-from] RE: comparing results
Importance: Low


I have been thinking about working on a SQL storage mechanism for JMeter.
The other tools I use all store to rel dbs, so I know *I* could really use
it. Opinions?

Keith Lancaster 

-----Original Message-----
From: ChristopherPesarchick@westfieldgrp.com
To: JMeter Users List
Sent: 1/9/04 7:33 AM
Subject: RE: comparing results





Maybe a good enhancement is to have JMeter save the results to a
database.
That way you always have your results in one location and it would be
easy
to do comparisons.

A person could have a set of SQL statements that gets executed to create
some report.
That way your are not always recreating a spreadsheet.

Something to think about.

-Chris



 

                      "Sonam Chauhan"

                      <sonam.chauhan@ce        To:       "'JMeter Users
List'" <jm...@jakarta.apache.org>                       
                      .com.au>                 cc:

                                               Subject:  RE: comparing
results                                                        
                      01/08/2004 11:11

                      PM

                      Please respond to

                      "JMeter Users

                      List"

 

 





> For example: An Aggregate Report might show percentage increase (or
> decrease) of average, min, and max stats; A Graph Results might
overlay
> one set of graphs on another.

Hi Chris - you may be better off exporting data to Excel, and doing this
analysis there.

At the bottom of this email, I've included a Perl snippet that parses an
*XML* format .jtl log file (LFILE) and writes out a '|' delimited stats
file (SFILE) for export to Excel.

With regards,
Sonam Chauhan
--
Corporate Express Australia Ltd.
Phone: +61-2-9335-0725, Fax: 9335-0753, Email: sonamc@ce.com.au


> -----Original Message-----
> From: Krahe, Chris [mailto:Chris.Krahe@aquilent.com]
> Sent: Friday, 9 January 2004 1:26 PM
> To: JMeter User (E-mail)
> Subject: comparing results
>
> Has anyone used a listener (or similar tool) that can read and compare
2
> or more results files (.csv or .jtl) from the same test plan?
>
> For example: An Aggregate Report might show percentage increase (or
> decrease) of average, min, and max stats; A Graph Results might
overlay
> one set of graphs on another.
>
> Thanks in advance...
> -Chris
>
>
> Chris Krahe
> Systems Architect
> Aquilent, Inc (ack-wil-lent)
> http://www.aquilent.com/
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: jmeter-user-unsubscribe@jakarta.apache.org
> For additional commands, e-mail: jmeter-user-help@jakarta.apache.org


Perl snippet:
-------------------------------------------------
 while (<LFILE>) {

  # Grab the relevant details from the log entry.
  # A normal log entry line logs one HTTP 'sampler'
operation:
  #   <sampleResult timeStamp="..." ...
threadName="..." label="..." time="..." />
  # In case the opertion had HTTP redirects, the redirects
show up as nested <sampleResult> elements,
  # but still on the same line:
  #   <sampleResult timeStamp= ... > <sampleResult
timeStamp=.... > ... </sampleResult>
  # We are only interested in the data in the first
<sampleResult> element, so
  # we use non-greedy pattern match operator ( '.*?' or
'.+?') to ignore any latter <sampleResult> elements,
  /timeStamp="(\d+)".+?threadName="(.*?)".+?label="(.+?)"
time="(\d+?)".+?success="(.+?)"/;

  my $timestamp = $1; # unix timestamp
  my $threadname = $2; # thread label
  my $label = $3; # operation label
  my $time = $4; # operation time in milliseconds
  my $success = $5; # boolean success indicator for this
operation
  # We then output stats with the '|' symbol as a
delimiter, except if:
  #   (a) we could not parse the information
successfully
  #   (b) This was a sleep operation using the 'Sleep
Test' sampler
  #        We ignore this because it is really a
delay, not an HTTP operation.
  print SFILE "$timestamp | $threadname| $label | $success
| $time \n"
   unless ($label =~ /Sleep Test/  ||
!defined($label));
 }
-------------------------------------------------



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




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

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

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