JMeter load distribution test - Waiting for possible Shutdown/StopTestNow/HeapDump/ThreadDump message on port 4445

[Arshad Salmani]

I have added 16 Concurrency thread group in my test plan and when running with 2K users with distributed load across 19 slave server I am able to see the summary report on the command prompt when run the command tail -f -n 2 MVP_2K_Run.log , But when running the same test plan with 15K thread (users) across the 19 slave server and run the same tail command then stuck at "Waiting for possible Shutdown/StopTestNow/HeapDump/ThreadDump message on port 4445" and there is no summary showing on the command prompt but the Jtl file is getting generated.

I have run my test plan with 15K users and distributed the load on 19 servers using below command

nohup /opt/apache-jmeter-5.4.1/bin/jmeter -Djava.rmi.server.hostname=10.104.180.** -n -t MVP5.2_BaseLinePT_15K.jmx -l Novus_MVP5.2_BaseLinePT_15K_23Oct.jtl -R <<1 to 19 slave servers ip address> :4000 > MVP5.2_BaseLinePT_15K_23Oct.log 2>&1 &

JTL and log files are got created on server under the specified path but summary of execution like thread started , active and finished, are not showing on the command prompt when running the below command.

tail -f -n 2 MVP5.2_BaseLinePT_15K_23Oct.log

It shows the below details only.

Remote engines have been started:[1 to 19 slaver server IP details] Waiting for possible Shutdown/StopTestNow/HeapDump/ThreadDump message on port 4445

Please help me solve these problems.

Also, the generated jtl file is getting very heavy for 60K VU (approximately 4.5 GB in size) please help me how to reduce the size of the jtl file.

[Dmitri T]

I don't think you can evenly distribute 2000 or 15000 threads between 19 slaves.

If you have 2000 users in Thread Groups and 19 slaves the total load is 38000.

If you have 15000 users in Thread Groups and 19 slaves the total load is 285000.

If you need to execute your test with 285000 concurrent users and it hangs:

1. Check jmeter-server.log files on slaves for any suspicious entries
2. Make sure to follow recommendations from 9 Easy Solutions for a JMeter Load Test “Out of Memory” Failure to properly tune your JMeter instances for high loads
3. Take thread and heap dumps and see where exactly your test is stuck. It might be the case your system under test is overloaded and doesn't respond and JMeter by default waits for the response forever, it might worth defining reasonable timeouts in your Samplers

With regards to the .jtl file size you might want to amend your results file configuration to store only those metrics which are absolutely required.

[Arshad Salmani]

Hi,

Actually, I divided the total number of threads by 19 and then added it to the concurrency thread group.

For example, if I have to test one scenario with 2000 VU, then 2000/19 = 105. So we are passing 105 Vu in the 1st currency thread group for 1 hour.

1500 Vu in the second scenario, then 1500/19 = 79, so we added 79 VU in the second concurrency thread group for 1 hour.

Like this, I have added it to the 16-concurrency thread group for 1 hour.

Arshad Salmani

Senior QA Engineer

+971 56 828 0883

asalmani@shory.com

Sky Tower, Office 1006

Al Reem Island, Abu Dhabi, UAE

Shory.com

Connect with us on LinkedIn

Follow us on Twitter | Instagram | Facebook

Shory Insurance Brokers is a licensed broker by the Central Bank of UAE with license number (287)

--
You received this message because you are subscribed to the Google Groups "jmeter-plugins" group.
To unsubscribe from this group and stop receiving emails from it, send an email to jmeter-plugin...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/jmeter-plugins/a7c7f298-4868-4d32-9c1e-381654961177n%40googlegroups.com.

The content of this email and any attachments are deemed confidential and intended for the recipient specified in message only. It is strictly forbidden to share any part of this message with any third party, without a written consent of the sender. If you received this message by mistake, please reply to this message and follow with its deletion. 

[rahiman shaik]

Bro look after the memory print condition and if required increase the heap size, then do it again

To view this discussion on the web visit https://groups.google.com/d/msgid/jmeter-plugins/CAMrdpQcVO%2B37uBS279JAtPaTMeDbVCA4WiJkodrYWcTuzMUuvA%40mail.gmail.com.