vlsi opened a new issue, #5955:
URL: https://github.com/apache/jmeter/issues/5955
### Use case
Fine-tuning HTTP samplers requires runing a server that could respond fast
with minimal resource utilization.
What do you think if we integrate a small mock server into JMeter sources?
I guess it would simplify to test performance-related changes.
### Possible solution
I tried the following, and I got ~88K/sec for HTTPClient4 implementation and
~60K/sec for Java implementation (~8 threads, Java 17, M1 Max)
```kotlin
import io.undertow.Undertow
import io.undertow.util.Headers
fun main() {
val server = Undertow.builder()
.addHttpListener(3000, "0.0.0.0")
.setHandler { exchange ->
// Sets the return Content-Type to text/html
exchange.responseHeaders
.put(Headers.CONTENT_TYPE, "application/json")
// Returns a hard-coded HTML document
exchange.responseSender
.send("""{"result":"ok"}""")
}.build()
// Boot the web server
server.start()
}
```
I guess we could predefined several routes (e.g. chunked encoding, large
response, small response, etc).
The drawback would be to version churn (Renovate bot would ask us upgrading
the versions as they release)
WDYT?
### Possible workarounds
_No response_
### JMeter Version
5.5
### Java Version
_No response_
### OS Version
_No response_
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
