Memory Issue When running with 1000 users with taurus + blazemeter API reporting

106 views
Skip to first unread message

Stalin Decruz

unread,
Mar 29, 2017, 11:17:54 AM3/29/17
to codename-taurus
Hi Team,

Am looking for little clarification on ways to avoid the memory usage While using taurus.

1.changing user.properties
2.How to limit the artifacts that gets collect while using taurus.

please throw some light.

Thanks,
Stalin

Andrey Pokhilko

unread,
Mar 29, 2017, 11:27:12 AM3/29/17
to codenam...@googlegroups.com

Hi,

Can you elaborate what exactly you want? Which file concerns you? Where do you see the memory usage, which process? Also, what is your OS?


Andrey Pohilko
Chief Scientist
P: +7 (909) 631-21-69
BlazeMeter Inc.
--
You received this message because you are subscribed to the Google Groups "codename-taurus" group.
To unsubscribe from this group and stop receiving emails from it, send an email to codename-taur...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/codename-taurus/824620b6-1e39-4110-b7dd-8bac4f57d88c%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Stalin Decruz

unread,
Mar 30, 2017, 5:31:52 AM3/30/17
to codename-taurus, andrey....@contractor.ca.com
Thanks for reply Andrey!
Here are the details,

OS: Linux centos
Issue : upto 600 users i did not see any issue once it crossed the level the memory started spiking up reaching >90 % throwing Memory exceptions.

Looking for all best ways to optimize memory and use memory effectively


1. In the user properties of thaurus -Jmeter if I change the saveservice( for eg jmeter.save.saveservice.assertions=false). Will taurus effect that change?
2.Also i want to know if there is any place where i can do the  configurations that will limit the reportiing.For example i dont need error logs, i dont need engine health.How i can set this so i can save some memory of my master machine.

Andrey Pokhilko

unread,
Mar 30, 2017, 5:50:57 AM3/30/17
to Stalin Decruz, codename-taurus

Engine health does not cost you any significant memory.

Can you investigate a bit deeper and tell me which process eats too much memory at the moment of spike - is it jmeter+java or taurus+python?

You can experiment with these flags, setting them to false: http://gettaurus.org/docs/JMeter/#JMeter-Test-Log


Andrey Pohilko
Chief Scientist
P: +7 (909) 631-21-69
BlazeMeter Inc.
Reply all
Reply to author
Forward
0 new messages