Run test: tests.test_cosim.test_BicycleDynamicAndDriver
Observe an output similar to the following:
FAILED [100%]['java', '-jar', 'D:\\srcctrl\\github\\pyfmu\\src\\pyfmu\\resources\\validation\\coe.jar', '-v', '--configuration', 'C:\\Users\\clagm\\AppData\\Local\\Temp\\tmpvo5w0t60\\coe.json', '--oneshot', '--starttime', '0.0', '--endtime', '10.0']
Version: 1.0.7-SNAPSHOT
WARN [main] (CoeInitialize.scala:395) - Make sure the execution tool: '' is available during this simulation.
An error has ocurred during instantiation of the FMU: Unable to instantiate slave
ERROR [main] (CoeInitialize.scala:420) - Could not instantiate: {bd}.bdi
WARN [main] (CoeInitialize.scala:395) - Make sure the execution tool: '' is available during this simulation.
ERROR [main] (CoeInitialize.scala:127) - Invalid start condition the instances could not be created in: List({bd}.bdi)
ERROR [main] (RequestProcessors.java:320) - Internal error in initialization
org.intocps.orchestration.coe.AbortSimulationException: Invalid start condition the instances could not be created in: List({bd}.bdi)
at org.intocps.orchestration.coe.scala.CoeInitialize$.initialize(CoeInitialize.scala:128)
at org.intocps.orchestration.coe.scala.Coe.initialize(Coe.scala:140)
at org.intocps.orchestration.coe.httpserver.RequestProcessors.processInitialize(RequestProcessors.java:311)
at org.intocps.orchestration.coe.httpserver.RequestHandler.initialize(RequestHandler.java:362)
at org.intocps.orchestration.coe.httpserver.RequestHandler.handleRequest(RequestHandler.java:212)
at org.intocps.orchestration.coe.util.SimulationExecutionUtil.callHttpAny(SimulationExecutionUtil.java:166)
at org.intocps.orchestration.coe.util.SimulationExecutionUtil.run(SimulationExecutionUtil.java:200)
at org.intocps.orchestration.coe.CoeMain.runOneShotSimulation(CoeMain.java:213)
at org.intocps.orchestration.coe.CoeMain.main(CoeMain.java:159)
Simulation failed Invalid start condition the instances could not be created in: List({bd}.bdi)
This is difficult to debug error because the message we get throws the developed into a different direction wrt to the cause.
I also would expect that removing those argument would make them available inside kwargs, but apparently this is changing the behavior of the wrapper in a drastic way...
Yes, the log messages printed by maestro are somewhat lacking. I am quite confident that a python stack trace is printed to stdout.
It is possible that maestro capture this and neglect to forward it to the output.
Steps to reproduce:
visible
andlogging_on
keyword arguments, so that the constructor looks like this:tests.test_cosim.test_BicycleDynamicAndDriver
This is difficult to debug error because the message we get throws the developed into a different direction wrt to the cause. I also would expect that removing those argument would make them available inside kwargs, but apparently this is changing the behavior of the wrapper in a drastic way...