I am looking for a way to mark JMS “sessions” of gatling as errors (KO) if they don’t get a reply during the given timeout.
At the moment my simulations would keep running forever waiting for those delayed/lost reply messages unless a simulation.maxDuration
is specified, in which case the missing messages still won’t be counted as KO
and I have to evaluate the ratio of the “pending” requests at the moment of completion, which is not very handy for parameterized automatic runs.
I thought that the JMS protocol’s “receiveTimeout” would help me, but this seems to be a technical parameter that doesn’t influence the reporting and is detached from the actual request.
The “gatling.http.ahc.requestTimeout” seems to be close to what I am looking for, but looks to be http specific.
Questions:
- Is it possible to implement this timing out logic using the gatling DSL, and if so - where should I look at?
- Would such feature be a useful contribution to JMS / generic DSL?