| || |
HTTPS protocal - generating load
I am using HTTPS protocal for my Application.
Vusers starts failing after 35 , and Most of the TCP connections opened by the Vusers are in TIME_WAIT state.
I get the below ERRORS when the Vusers started failing
1 ) ERROR 10060: (A connection attempt failed because the connected party did not properly respond after a period of time
2)ERROR 10048: (Only one usage of each socket address (protocol/network address/port) is normally permitted. )
3)ERROR 10055: (An operation on a socket could not be performed because the system lacked sufficient buffer space or because a queue was full. ).
I searched for these errors and i fine tuned the network parameters in windows machine
Changed the following values in the windows registry
in the windows registry
in Testexecuter_ini file.
But No improvrment.
Configurations of the machine from where the load is generated
CPU : pentium 4
RAM :1 GB
I tried accessing the application after generating 40 user load from controller and manually tried in the same machine.The application could not be opened.
But accessed the application from another machine.Able to acceesing it fine.
I used one load generator and i was able to scale upto 100 vusers.But the vusers were not stable in the during the stand by time as they do more iterations and require more TCp connections
Is there any way to increase the TCP connections in windows Operating system?
or any other means to fix this problem. [img]/images/graemlins/confused.gif[/img]
Re: HTTPS protocal - generating load
OpenSTA is the memory intensive tool. So you need to benchmark how much the virtual users it will handle in the single system. After benchmarking you need to add number of load generator for your load/performance testing.
It is better to tune the “TestExecuter_web.ini” parameters in OpenSTA based on your test execution.
I guess this information will help you for your testing.
Thanks & Regards,