How should response times be interepreted when testing Web Services?


I am performing load testing on a webscoket connection. My simulation opens the ws/socketIO connection and then sends several messages to the socketIO server.

Examining the generated statistics and graphs I do not understand what is the meaning of the response time in the case of sending messages via ws, since a ws message does not have an associated response as in the case of http requests. What do the numbers represent?

Thanks for your help.


This is the time between your outbound message and the checks completion.

Hi again,

Thanks for the answer.

What is not clear to me, is that even in a simulation with no checks, only ws send messages, I still see in the report number of responses and response time. Is it the case that Gatling still performs (internally) check on incoming messages?

What is even more puzzling, is that I tried connecting to a ws server which does not emit any messages at all. So there should be no incoming responses in the simulation. Still in the report, responses and times are reported.

What does Gatling measure in this case?

Sorry, there’s also the response times for the CONNECT requests for opening the WebSockets.

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.