JMeter: javax.net.ssl.SSLException when running 'many' threads

1
votes
timeStamp,elapsed,label,responseCode,responseMessage,threadName,dataType,success,failureMessage,bytes,sentBytes,grpThreads,allThreads,URL,Latency,IdleTime,Connect

1600278792845,6688,Category page,Non HTTP response code: javax.net.ssl.SSLException,Non HTTP response message: Connection reset,10.0.4.22-Thread Group 1-1619,text,false,Test failed: code expected to match /200/,3985,0,1830,1830,https://<my-test-url>,0,0,148

I'm running my JMeter tests from a master with three slave nodes. When using a 'low' number of threads (i.e. ~200) the tests run without any errors. But if I increase the number of threads to ~300, I start getting the above error message in my .jtl file. What does it mean?

If I run 1000 threads on a single slave node (still executed from the master node), I get no errors.

Some thoughts:

  • I would expect a java.lang.OutOfMemoryError exception if the slave node(s) itself couldn't run this many threads.
  • The tests are executed from an Azure environment, hence I don't believe its a network issue or such, since Azure should be able to handle this amount of traffic

System info:

  • Ubuntu 18.04 on all machines
  • Java 8
  • JMeter 5.0
1
load-testingjmeter-5.0

1 Answers

1
votes
  1. First of all make sure to check your application logs as it might indicate issues with the system under test. If you're sure that your application behaves fine you can consider following recommendations from JMeterSocketClosed wiki page
  2. If I run 1000 threads on a single slave node (still executed from the master node), I get no errors. - check the throughput (number of hits per second) as it might be the case that a single node lacks resources and cannot send requests fast enough and when you run 900 threads in distributed mode you conduct much higher throughput. Charts to check are Transactions per Second and Server Hits per Second
  3. I would expect a java.lang.OutOfMemoryError exception if the slave node(s) itself couldn't run this many threads. - not necessarily, JMeter can just execute requests more slowly due to excessive garbage collection
  4. Consider upgrading to JMeter 5.3 (or whatever is the latest stable version available at JMeter Downloads page) as you might be suffering from a JMeter issue which has already been fixed, as per JMeter Best Practices you should always be using the latest version of JMeter