New updates

The world’s best load testing service gets even better!

Yesterday, on Aug 2, we pushed a small update that included the following features and changes:

• Ramp-up restrictions removed
  You can now ramp up or down the number of simulated users in a test as quickly as you like (well, almost – a ramp operation can be done in as little as 1 minute now).
• Ramp steps can be any length
  Previously, a single ramp-up/down step could have a duration of max 120 minutes. This limit has been removed. There is now just a single limit for the whole test schedule, which can be max 24 hours (1440 minutes).
• Changed default values for tests
  All defaults for test configurations, are now to ramp up to 50 VU during 10 minutes (previously, an automatically created test configuration would ramp up to 50 VU during 15 minutes, while a new test configuration created by a user would by default ramp up to 25 VU during 10 minutes).
• New load script API functionality: auto_cookie_handling
  There is now a new boolean option that can be set with http.set_option() – “auto_cookie_handling”. It is set to true by default, but if set to false it will turn off the automatic handling of cookies between requests, allowing the script programmer to design his/her own cookie management.
• Load generator bug fix
  Fixed an intermittent bug caused by issuing sleep statements with sleep time set to zero.

On June 20, we pushed another small update that included these changes:

• Several proxy recorder bug fixes:
  • Fixed problem with injected toolbar appearing in the wrong place
  • Fixed problem with extra HTML added
  • Fixed problem with proxy sometimes generating extra CRLF’s to requests
• New page on site: The state of web readiness 2012
  http://loadimpact.com/readiness
• Company address on invoices
  You now get your company’s address on your receipts/invoices, viewable online at loadimpact.com

On June 8, we released Load Impact 2.4 that contained the following fixes and improvements:

• New load script API functionality
  In this release we introduced a range of new API functions:
  • client.get_user_scenario_name() – returns name of currently executing user scenario
  • client.get_time_since_start() – returns elapsed time since the simulated user started executing
  • client.get_source_ip() – returns the source IP address seen in network traffic generated by this user
  • test.get_name() – returns the name of the currently running load test
  • test.get_time_since_start() – returns elapsed time since start of test execution
  • util.unique() – returns a string guaranteed to be unique across simulated clients in a test
• Extra IP addresses
  You can now configure your load test to use more source IP addresses when generating traffic. This comes at an extra cost as it requires more infrastructure (cloud) resources, but can be very useful for e.g. spreading traffic evenly if you have a load balancer.
• Small UI changes
  Several minor UI tweaks & fixes:
  • Changed “Test title” to “Test name”, for consistency
  • Fixed inconsistent naming of load zones. Load zones are now named as: CITY, COUNTRY CODE (PROVIDER)
    E.g. “Ashburn, US (Amazon)”
• Bugfixes
  • Fixed broken resend email activation link
  • Fixed bug allowing tests to be scheduled up to 1 hour in the past
  • Fixed pagination bug in URL table on test results page
  • Fixed deployment bug affecting graphical editor user scenarios containing Unicode characters
  • Fixed bug causing screen to gray out in certain cases when selecting script editor for new user scenario