Gatling version : 3.13.1Gatling flavor : Gatling build tool :
Hi,
As part of a single Simulation, I want to run:
Warm Up scenario, which starts from 0 RPS, gradually increases up to $target RPS, and holds the #target RPS for several seconds
After the Warm Up is finished, I’d like to run real load tests, that jump right away to $target RPS, and hold the traffic for some time
The setup is presented below:
import io.gatling.javaapi.core.CoreDsl.*
import io.gatling.javaapi.core.OpenInjectionStep.atOnceUsers
import io.gatling.javaapi.core.Simulation
import io.gatling.javaapi.http.HttpDsl.http
import java.time.Duration
class ComputerDatabaseSimulation : Simulation() {
private val warmUp = scenario("WarmUp")
.forever()
.on(exec(http("[WarmUp] GET /computer-database.gatling.io").get("https://computer-database.gatling.io/computers")))
private val testScn = scenario("TestScn")
.forever()
.on(exec(http("[TestScn] GET /computer-database.gatling.io").get("https://computer-database.gatling.io/computers")))
init {
setUp(
warmUp
.injectOpen(atOnceUsers(5))
.throttle(
reachRps(2).during(Duration.ofSeconds(5)),
holdFor(Duration.ofSeconds(10))
)
.andThen(
testScn
.injectOpen(atOnceUsers(5))
.throttle(
jumpToRps(2),
holdFor(Duration.ofSeconds(10))
)
)
)
}
}
Unfortunately, only “WarmUp” scenario is executed, the “TestScn” is totally skipped:
gradle gatlingRun --simulation=ComputerDatabaseSimulation
> Task :load-tests:gatlingRun
Simulation ComputerDatabaseSimulation started...
================================================================================
2024-11-17 15:33:13 GMT 5s elapsed
---- Requests ------------------------------------------------------------------
> Global (OK=4 KO=0 )
> [WarmUp] GET /computer-database.gatling.io (OK=4 KO=0 )
---- WarmUp --------------------------------------------------------------------
[--------------------------------------------------------------------------] 0%
waiting: 0 / active: 5 / done: 0
---- TestScn -------------------------------------------------------------------
[ ] 0%
waiting: 5 / active: 0 / done: 0
================================================================================
================================================================================
2024-11-17 15:33:18 GMT 10s elapsed
---- Requests ------------------------------------------------------------------
> Global (OK=14 KO=0 )
> [WarmUp] GET /computer-database.gatling.io (OK=14 KO=0 )
---- WarmUp --------------------------------------------------------------------
[--------------------------------------------------------------------------] 0%
waiting: 0 / active: 5 / done: 0
---- TestScn -------------------------------------------------------------------
[ ] 0%
waiting: 5 / active: 0 / done: 0
================================================================================
================================================================================
2024-11-17 15:33:18 GMT 10s elapsed
---- Requests ------------------------------------------------------------------
> Global (OK=14 KO=0 )
> [WarmUp] GET /computer-database.gatling.io (OK=14 KO=0 )
---- WarmUp --------------------------------------------------------------------
[--------------------------------------------------------------------------] 0%
waiting: 0 / active: 5 / done: 0
---- TestScn -------------------------------------------------------------------
[ ] 0%
waiting: 5 / active: 0 / done: 0
================================================================================
Parsing log file(s)...
Parsing log file(s) done in 0s.
Generating reports...
================================================================================
---- Global Information --------------------------------------------------------
> request count 14 (OK=14 KO=0 )
> min response time 118 (OK=118 KO=- )
> max response time 428 (OK=428 KO=- )
> mean response time 215 (OK=215 KO=- )
> std deviation 121 (OK=121 KO=- )
> response time 50th percentile 130 (OK=130 KO=- )
> response time 75th percentile 360 (OK=360 KO=- )
> response time 95th percentile 428 (OK=428 KO=- )
> response time 99th percentile 428 (OK=428 KO=- )
> mean requests/sec 1.4 (OK=1.4 KO=- )
---- Response Time Distribution ------------------------------------------------
> t < 800 ms 14 ( 100%)
> 800 ms <= t < 1200 ms 0 ( 0%)
> t >= 1200 ms 0 ( 0%)
> failed 0 ( 0%)
================================================================================
BUILD SUCCESSFUL in 15s
4 actionable tasks: 4 executed
What am I doing wrong?
Cheers!
The issue is with your forever loop in your warmUp scenario.
Maybe you’re expecting your throttle to stop all the virtual users of this scenario to stop after 15 seconds but that’s not how throttle works: the limit is lifted after 15s.
Hi,@slandelle !forever loop, but it doesn’t work the way I want: testScn is executed after warmUp, but we run only 5 requests per scenario.
Code sample:
import io.gatling.javaapi.core.CoreDsl.*
import io.gatling.javaapi.core.OpenInjectionStep.atOnceUsers
import io.gatling.javaapi.core.Simulation
import io.gatling.javaapi.http.HttpDsl.http
import java.time.Duration
class ComputerDatabaseSimulation : Simulation() {
private val warmUp = scenario("WarmUp")
.exec(http("[WarmUp] GET /computer-database.gatling.io").get("https://computer-database.gatling.io/computers"))
private val testScn = scenario("TestScn")
.exec(http("[TestScn] GET /computer-database.gatling.io").get("https://computer-database.gatling.io/computers"))
init {
setUp(
warmUp
.injectOpen(atOnceUsers(5))
.throttle(
reachRps(2).during(Duration.ofSeconds(5)),
holdFor(Duration.ofSeconds(10))
)
.andThen(
testScn
.injectOpen(atOnceUsers(5))
.throttle(
jumpToRps(2),
holdFor(Duration.ofSeconds(10))
)
)
)
}
}
Execution log:
gradle gatlingRun --simulation=ComputerDatabaseSimulation
> Task :load-tests:gatlingRun
Simulation ComputerDatabaseSimulation started...
================================================================================
2024-11-18 07:08:29 GMT 5s elapsed
---- Requests ------------------------------------------------------------------
> Global (OK=4 KO=0 )
> [WarmUp] GET /computer-database.gatling.io (OK=4 KO=0 )
---- WarmUp --------------------------------------------------------------------
[###########################################################---------------] 80%
waiting: 0 / active: 1 / done: 4
---- TestScn -------------------------------------------------------------------
[ ] 0%
waiting: 5 / active: 0 / done: 0
================================================================================
================================================================================
2024-11-18 07:08:32 GMT 8s elapsed
---- Requests ------------------------------------------------------------------
> Global (OK=10 KO=0 )
> [WarmUp] GET /computer-database.gatling.io (OK=5 KO=0 )
> [TestScn] GET /computer-database.gatling.io (OK=5 KO=0 )
---- WarmUp --------------------------------------------------------------------
[##########################################################################]100%
waiting: 0 / active: 0 / done: 5
---- TestScn -------------------------------------------------------------------
[##########################################################################]100%
waiting: 0 / active: 0 / done: 5
================================================================================
Parsing log file(s)...
Parsing log file(s) done in 0s.
Generating reports...
================================================================================
---- Global Information --------------------------------------------------------
> request count 10 (OK=10 KO=0 )
> min response time 372 (OK=372 KO=- )
> max response time 478 (OK=478 KO=- )
> mean response time 390 (OK=390 KO=- )
> std deviation 30 (OK=30 KO=- )
> response time 50th percentile 381 (OK=381 KO=- )
> response time 75th percentile 391 (OK=391 KO=- )
> response time 95th percentile 478 (OK=478 KO=- )
> response time 99th percentile 478 (OK=478 KO=- )
> mean requests/sec 1.11 (OK=1.11 KO=- )
---- Response Time Distribution ------------------------------------------------
> t < 800 ms 10 ( 100%)
> 800 ms <= t < 1200 ms 0 ( 0%)
> t >= 1200 ms 0 ( 0%)
> failed 0 ( 0%)
================================================================================
BUILD SUCCESSFUL in 11s
4 actionable tasks: 3 executed, 1 up-to-date
The fix seems easy right now: don’t use atOnceUsers(5), use sth like constantUsersPerSec(2.0).during(Duration.ofSeconds(15)).
Code sample:
import io.gatling.javaapi.core.CoreDsl.*
import io.gatling.javaapi.core.Simulation
import io.gatling.javaapi.http.HttpDsl.http
import java.time.Duration
class ComputerDatabaseSimulation : Simulation() {
private val warmUp = scenario("WarmUp")
.exec(http("[WarmUp] GET /computer-database.gatling.io").get("https://computer-database.gatling.io/computers"))
private val testScn = scenario("TestScn")
.exec(http("[TestScn] GET /computer-database.gatling.io").get("https://computer-database.gatling.io/computers"))
init {
setUp(
warmUp
.injectOpen(constantUsersPerSec(2.0).during(Duration.ofSeconds(15)))
.throttle(
reachRps(2).during(Duration.ofSeconds(5)),
holdFor(Duration.ofSeconds(10))
)
.andThen(
testScn
.injectOpen(constantUsersPerSec(2.0).during(Duration.ofSeconds(10)))
.throttle(
jumpToRps(2),
holdFor(Duration.ofSeconds(10))
)
)
)
}
}
Unfortunately, testScn is not executed:
gradle gatlingRun --simulation=ComputerDatabaseSimulation
> Task :load-tests:gatlingRun
Simulation ComputerDatabaseSimulation started...
================================================================================
2024-11-18 07:17:21 GMT 5s elapsed
---- Requests ------------------------------------------------------------------
> Global (OK=4 KO=0 )
> [WarmUp] GET /computer-database.gatling.io (OK=4 KO=0 )
---- WarmUp --------------------------------------------------------------------
[#########--------------- ] 13%
waiting: 20 / active: 6 / done: 4
---- TestScn -------------------------------------------------------------------
[ ] 0%
waiting: 20 / active: 0 / done: 0
================================================================================
================================================================================
2024-11-18 07:17:26 GMT 10s elapsed
---- Requests ------------------------------------------------------------------
> Global (OK=14 KO=0 )
> [WarmUp] GET /computer-database.gatling.io (OK=14 KO=0 )
---- WarmUp --------------------------------------------------------------------
[##################################--------------- ] 46%
waiting: 10 / active: 6 / done: 14
---- TestScn -------------------------------------------------------------------
[ ] 0%
waiting: 20 / active: 0 / done: 0
================================================================================
================================================================================
2024-11-18 07:17:26 GMT 10s elapsed
---- Requests ------------------------------------------------------------------
> Global (OK=14 KO=0 )
> [WarmUp] GET /computer-database.gatling.io (OK=14 KO=0 )
---- WarmUp --------------------------------------------------------------------
[##################################--------------- ] 46%
waiting: 10 / active: 6 / done: 14
---- TestScn -------------------------------------------------------------------
[ ] 0%
waiting: 20 / active: 0 / done: 0
================================================================================
Parsing log file(s)...
Parsing log file(s) done in 0s.
Generating reports...
================================================================================
---- Global Information --------------------------------------------------------
> request count 14 (OK=14 KO=0 )
> min response time 353 (OK=353 KO=- )
> max response time 458 (OK=458 KO=- )
> mean response time 377 (OK=377 KO=- )
> std deviation 23 (OK=23 KO=- )
> response time 50th percentile 374 (OK=374 KO=- )
> response time 75th percentile 376 (OK=376 KO=- )
> response time 95th percentile 458 (OK=458 KO=- )
> response time 99th percentile 458 (OK=458 KO=- )
> mean requests/sec 1.4 (OK=1.4 KO=- )
---- Response Time Distribution ------------------------------------------------
> t < 800 ms 14 ( 100%)
> 800 ms <= t < 1200 ms 0 ( 0%)
> t >= 1200 ms 0 ( 0%)
> failed 0 ( 0%)
================================================================================
BUILD SUCCESSFUL in 13s
4 actionable tasks: 3 executed, 1 up-to-date
Can you please help me fix the above example? Much appreciated, thanks
That’s a bad interaction between:
sequential scenarios
and the behavior of throttle
Gatling will automatically interrupt your test at the end of the throttle, just like it does with maxDuration
It looks pretty complicated to change this, don’t expect a solution any time soon.
I recommend that you go with another solution than throttle and design you scenario and your injection profile in a proper way to generate your expected traffic.
init {
setUp(
warmUp
.injectOpen(
rampUsersPerSec(0.0).to(2.0).during(5),
constantUsersPerSec(2.0).during(10)
)
.andThen(
testScn
.injectOpen(
constantUsersPerSec(2.0).during(10)
)
)
)
}
system
December 19, 2024, 4:24pm
5
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.