Ranorex Forum

I have scheduled single test cases execution by Jenkins. Specific Ranorex agents have been set in command line parameters. Some of the tests needs to be run on a specific agent, but most of them could be run on any agent. Some questions:

1. Is it possible to define agent group's tag in command line, instead of a particular agent name (/agent: parameter)?
2. How many tests an agent can buffer? Number of Jenkins executors has to be much bigger than amount of agents to avoid idle time in some of the agents. Can there be tens of test cases in one agent's queue?
3. Why the .exe file remains reserved in Ranorex Studio during the execution in agent? This prevents building anything of the same solution during the agent execution.

Hi,

I believe, that the primary purpose of Ranorex remote execution, has been designed for people without CI in place. Basically, its purpose is to run tests on remote machine directly from Studio. In other words, using Ranorex agents, if you are already using Jenkins or similar CI solution, is a bit overkill

I'm using Jenkins and all my tests are built via Jenkins and started either from Ranorex (Jenkins) execution plugin or command line. There is no need to use "Agents" middle-layer, which only complicates whole business. True, if there is command line support for Ranorex agents, why not to use it? But as you already experienced it, the remote execution is still somehow tight with Studio. So using command line (agent-less) execution is still better option. At least I think so

Hi Timo,

Jenkins has it's own possiblities to work with nodes/agents:

stage('RunRX') {
	agent { label 'RunRxWithOracleDB' }  // label assigned to certain nodes which are capable of running Rx 
	steps {
		echo "start run of ranorex test"
		script {
			bat "YourRxProject.exe /testsuite:${params.Testsuite} /runconfig:${params.Runconfig}"	
		}
	}		
}