Does the Burst Load Strategy work in 2.5.1?
I am trying to test a simple mock service with a Burst test, and I question whether it actually bursts or not.
My settings are:
Threads: 5
Strategy: Burst
Bursty Delay: 15
Burst Duration: 3
Limit: 90 seconds
It appears as though it starts with bursts (not delays), which is fine, and I see output from the Mock Service's console. However, as I watch the console, it continues to constantly be populated with request data during the 15 second delay. It is as though the Burst test just constantly hits the Mock Service, and the concept of Burst is ignored. According to the docs, all 5 Threads should sleep for the delay time, and it does not appear that they are sleeping.
It also states that the delay is in seconds, but to just be sure, I set the delay time to 10000 (10 seconds if the measurement is milliseconds), which produced identical results.
Now, it is possible that the logs are simply slower to output data to the console, so I increased my Limit to 120 seconds, Burst Delay to 30, and decreased the Burst duration to 2 seconds, and received the same result: requests constantly being written to the Mock Service console.
The numbers in the table (all columns) are also constantly updated, which is another indication to me that a true "burst" is not occurring.
Please advise.
Thanks,
Sean
My settings are:
Threads: 5
Strategy: Burst
Bursty Delay: 15
Burst Duration: 3
Limit: 90 seconds
It appears as though it starts with bursts (not delays), which is fine, and I see output from the Mock Service's console. However, as I watch the console, it continues to constantly be populated with request data during the 15 second delay. It is as though the Burst test just constantly hits the Mock Service, and the concept of Burst is ignored. According to the docs, all 5 Threads should sleep for the delay time, and it does not appear that they are sleeping.
It also states that the delay is in seconds, but to just be sure, I set the delay time to 10000 (10 seconds if the measurement is milliseconds), which produced identical results.
Now, it is possible that the logs are simply slower to output data to the console, so I increased my Limit to 120 seconds, Burst Delay to 30, and decreased the Burst duration to 2 seconds, and received the same result: requests constantly being written to the Mock Service console.
The numbers in the table (all columns) are also constantly updated, which is another indication to me that a true "burst" is not occurring.
Please advise.
Thanks,
Sean
