Results 1 to 3 of 3
  1. #1

    Testing communication chain incl. MQs

    Hello *,

    we do have an architecture involving a range of systems and whole farm of MQs:

    [Start]--> MQ1 --> System 1 (including some MQs as well)-->MQ2 --> System 2 --> post to several parallel MQs (MQ3) --> [End]

    Now I wonder:
    is it possible to test posting messages into the first MQ1 via Loadrunner and to retrieve the "matching" messages from MQ2 and MQs3 (there is a unique identifier somewhere in the XML). The objective is to measure the time it takes for the information to ripple through the MQs and systems.

    Of course, all MQs are not unique but working in a load-balanced cluster in parallel. So, I would not know in which exact Queue my result ends up.

    Has anyone ever built somethin like "VU listening on several MQs" and got some measurements between "insert at [Start]" and "retrieve at [End]".

    Btw, it is Loadrunner 11 Patch 3 with MQ Series version 7.

    Any idea or expierience with setting up an effective and effiecient script for the above?

    Best regards

  2. #2

    Re: Testing communication chain incl. MQs

    Open lrun.h in your <loadrunner home\include directory. Search for the term 'distributed.' You are likely to find a couple of functions that could be of some assistance.

    I have observed inconsistent operations on thee functions. You are much better off if you can pull information from an audit log on each of the MQ instances and sync on your message ID or use a database which is storing transacational information as these asynchronous transactions are processed.

    If you go the log processing route consider the use of Microsoft logparser to process the combined logs from all of your hosts for the min/max instance grouped by your message ID.
    James Pulley

    Replace ineffective offshore contracts, LoadRunnerByTheHour. Starting @ $19.95/hr USD.

    Put us to the test, skilled expertise is less expensive than you might imagine.

    Twitter: @LoadRunnerBTH @PerfBytes

  3. #3

    Re: Testing communication chain incl. MQs

    Best practice would be to find the source points and emulate the traffic using the source application. I am sure your MQ1 will get the message from *some other* application (could be another server, application etc). Work with your solution/functional architects and find out the *top* business process that will *generate* load on MQ1. Same goes to generating traffic on MQ3. When you do this, you will test MQ2 by default.

    Work with your middleware support team and extract the message response times using a SQL query or login to the middlware manager and *export* the table contents. Most of the middleware interface applications come with an app and db servers to store the history. Chances are high that you will able to get it directly from the support teams rather than complicating it in loadrunner.

    from my experience [img]/images/graemlins/smile.gif[/img]
    The views expressed in this forum are mine. My organization does not subscribe to the substance, veracity or truthfulness of the said opinion



Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
BetaSoft Inc.
All times are GMT -8. The time now is 05:43 PM.

Copyright BetaSoft Inc.