Nederlands - nl-NLEnglish (United Kingdom)
PDF Print E-mail

Zyntax Tuxedo Adaptor (ZTA) for Rational Robot and TestManager


An adaptor for BEA Tuxedo (c) environments which is integrated with Rational's Robot and TestManager.

The Tuxedo Adaptor allows users to perform load, stress, performance, and benchmarking against applications which utilize BEA's Tuxedo. This includes packaged applications such as PeopleSoft or Clarify, and generated application such as those from CoolGen.

For a complete product overview, see the Powerpoint Presentation .

Recording Tuxedo Sessions

The Rational TestSuite product line provides some (limited) support for Tuxedo applications up until Tuxedo version 6, but there is no support for higher versions 7.x and 8.x. The below example shows a fragment of a typical script for an applications using Tuxedo© VIEWS.

 

 

Although some data can still be recognized, using datapools in this script can be pretty cumbersome, especially if you want to use data strings with different lengths than the recorded strings.

For Tuxedo 7+, the situation is even worse, since due to the lack of support for these protocols, the script generator defaults to socket level scripts, and then your script may look like this:

 

 

In this case, none of the data is recognizable, and neither is the underlying structure in terms of Tuxedo programming (the tux_*() calls in the first example). Obviously, using datapools in this case will be very difficult and time-consuming.

When recording with the Zyntax Tuxedo Adapter, the corresponding script fragment looks like this:

 

 

As you can see, both the data and the underlying program structure are visible, and using datapools during replay will be easy. Each data command shows the data type, the field name (automatically taken from the Tuxedo .DEF or .VV files), and the data for that field.

The vertical bars in the data strings (e.g. for field "v2s3") are used when a certain field is an array, causing multiple values to be written with a single line.

The pad*() macros are used to zero-pad or space-pad string fields that are larger than the actual strings that are sent.

Replaying Tuxedo Sessions

The ZTA comes with a replay library which translates the script commands back to actual Tuxedo calls, and packs the data for you in a platform-independent way. Using datapools is straightforward; one simple replaces the strings in the script with strings from the datapool.

An interesting feature about the ZTA is the ability to retrieve data from incoming data buffers. This may be needed in the case that the server returns session identifiers (or "cookies") which must be passed on subsequent calls. The ZTA provides a simple command to do this task:

 

 

In this example the 4 byte signed integer "identifier" is retrieved from the buffer called "_odata", at offset 24 (this offset information is supplied by ZTA). The retrieved data is formatted into a string which can be used in further calls.

Supported platforms

Currently, the following platforms are supported: Windows 2000/NT, Linux, HPUX. Support for SunOS is under development, and other platforms may be supported on request.

 

If you want to know more about the ZTA, simply send a short email to This e-mail address is being protected from spambots. You need JavaScript enabled to view it and we'll send you the information proactively. Note that we respect your email address, and never give it out. We strongly dislike spam.

 
Please report web-related issues to webmaster@zyntax.com
Copyright 2010 Zyntax Consulting BV, All Rights Reserved