Release Notes for MailSizr 2.5

Here are major new features/fixes in this release. They are NOT organized by any priority order.

  1. Support for separate script files.
    This option enable you to simulate 1 script per user. For example, in a 500 user test, you can choose to assign the same script to all users or have 1 script per user (for a total of 500 scripts during one single experiment).
  2. Flexibility to define collection intervals for statistics.
    You can define the way statistics are presented so you have MailSizr pre-analyze intervals of test as opposed to the entire test.
  3. Improvement to reporting capabilities.
    More details and better organization with a new look
  4. Embellishment and organization of documentation.
  5. Ability to log screen content to the report.
    This gives you the ability to log the HTML source pages which were returned from the experiment into your test report. May be a overkill for real tests, but can be very useful while debugging your application under load.
  6. Support to send mail to random users.
    This is the ability to pick a user in random using the MAILLIST.TXT file. One of the included scripts file provides an example.
  7. Timeout parameter (in Setup Applet)
    Previously, the timeout parameter applied to all commands, including the PAUSE commands. Now this setting applies only to active commands.
  8. RunFor directive.
    The RunFor directive in the older versions had a minor "Time Leak" which would show up only on simulations which lasted for very long periods of time. It has now been completely tightened up.
  9. Provided sample scripts with Documentation.
    Please send ideas for more of such scripts to us so we can include them in newer releases.
  10. User Desynchronization This has been modified to enable very quick rampup on the experiment
  11. WaitUntilRampup function
    This function is a new one. This can be defined by checking the appropriate box in the "Miscellaneous settings" tab of the Setup Applet. It allows the capability to stall simulation of commands in the script until all users have been created. This gives the user a flexibility of defining the workload generation pattern.
  12. Stability issues:
    Capability to handle socket and server exceptions when the network stack gets broken due to application/user load.