What’s new in JMeter 5.5?

Reading Time: 3 minutes

Hi Folks,
In this blog post, we are going to see what’s new in JMeter 5.5, Apache community has released 5.5 with a couple of new features, several improvements, and fixes.

What’s new in JMeter 5.5?

  • A brand-new thread group, Open Model Thread Group
  • JMeter now supports Java 17
  • Log4j 2.17.2 is now available in JMeter
  • JMeter 5.5 is the ability to replace the Oro-based Regex implementation by the built-in Java based one. To choose the Java-based one, set the JMeter property jmeter.regex.engine to the value java
  • Kotlin language is used in some core classes and tests (e.g. Open Model Thread Group)

JMeter 5.5 has several bug fixes and enhancements in almost all the areas

Bug fixes done in the release:

  • Bug 65027 – Now we detect mime-type for files automatically when adding files to HTTP Sampler
  • Bug 65020 – HTTP Sampler/Files upload tab – add missing buttons

  • Pull request #650 – HTTP Sampler timestamp fix when an exception caught
  • Bug 65353 – Now, JMeter makes the estimator used for calculating percentiles on the dashboard configurable
  • Bug 65328Pull request #666 – From Now, HTTP 308 Permanent Redirect is not supported
  • Bug 65515 – Allow pooling of Prepared Statements in JDBC
  • Bug 63061 – Sort View Results in Table in a human expected order
  • Bug 64962 – Save CSV sub-results recursively from View Results Tree
  • Bug 65784 – No Graphs displayed in Aggregate Report/Response Time Graph
  • Bug 65884 – GUI doesn’t display response for multipart request manually encoded
  • Bug 65269 – JSON Extractor and JSON JMESPath Extractor ignore sub-samples
  • Bug 65352 – Warning logged when Boundary Extractor doesn’t find any match
  • Bug 65681 – Use default values for null values when extracting with JSONPostProcessor
  • Bug 65782 – Use correct message format for MessageFormat in HTMLAssertion
  • Bug 65794 – JSON Assertion always successful with indefinite paths
  • Pull request #665 – Now, Increased visible lines of code in IfController and WhileController

JMeter fixes many more other bugs. You can check from here

Non-Functional changes done in the release:

  • Bug 65128 – Pull request #643 – Add missing documentation about the Same user on each iteration for Thread Groups
  • Bug 65232 – Hide splash screen when an error displayed because the test plan could not parsed
  • Updated Groovy to 3.0.11 (from 3.0.7)
  • Updated Apache ActiveMQ to 15.6.4 (from 15.6.0)
  • Updated Apache HttpClient to 4.5.13 (from 4.5.12)
  • Updated json-path to 2.7.0 (from 2.4.0)
  • Updated JUnit to 4.13.2 and 5.8.2 (from 4.13.1 and 5.7.0)
  • Updated Apache log4j2 to 2.17.2 (from 2.13.3)
  • Updated Gradle to 7.3 (from 7.2)
  • Updated jmespath to 0.5.1
  • Updated ktlint to 0.40.0
  • Updated hsqldb to 2.5.2 (from 2.5.0)
  • Updated PH CSS and PH commons to 6.5.4 and 10.1.6 (from 6.2.3 and 9.5.1)
  • Updated RSyntaxTextArea to 3.2.0 (from 3.1.1)
  • Updated SLF4J to 1.7.36 (from 1.7.30)
  • Updated SvgSalamander to 1.1.2.4 (from 1.1.2.1)
  • Updated WireMock-JRE8 to 2.30.0 (from 2.24.1)
  • Updated com.github.vlsi.vlsi-release-plugins 1.76 (from 1.74)
  • Updated jackson to 2.13.3 (from 2.10.5)
  • Updated jmespath to 0.5.1
  • Updated Saxon-HE to 11.2 (from 9.9.1-8)
  • Updated Apache xmlgraphics commons to 2.7 (from 2.6)

For more details you can check from here

Open Model Thread Group

  • Open Model Thread Group defines a pool of users that will execute a particular test case against the server
  • The users are generated according to the schedule
  • The load profile consists of a sequence of constant, increasing, or decreasing load
  • The thread groups terminate threads as soon as the schedule ends

Parameters:

  • Name: Descriptive name for this thread group that is shown in the tree
  • Schedule: The expression that configures the schedule. For example: rate(5/sec) random_arrivals(1 min) pause(5 sec)
  • Random Seed: different thread groups should better have different seed values. Constant seed ensures the thread group generates the same delays each test start. The value of “0” means the schedule is truly random

I hope you enjoyed it and it helped you!! stay connected for more interesting blogs.

References :

https://jmeter.apache.org/changes.html

Written by 

I am curious about to learn new technologies. I have good understanding of performance testing concepts.