"Entry point 'Single Pass' does not exist." with TestStand Base Deployment Engine

[iseGmbH]

Hi,

I'm currently trying to set up a machine solely for running test sequences and therefore only has the "Base Deployment Engine" installed.

When running a dummy sequence I'm getting the error "Entry point does not exist" for both 'Single Pass' and 'MainSequence', although I can run at least the 'MainSequence' manually from the test runner:

$ ./TestExec.exe -outputToStdIO -operatorInterface -runEntryPoint 'MainSequence' dummy.seq
Entry point 'MainSequence' does not exist.               # <- Error from the shell command
Execution 'MainSequence - dummy.seq [2]' ended           # <- After MANUALLY running the 'MainSequence'
Status: Error
[...]

Regarding such error messages I found this solved forum post: https://forums.ni.com/t5/NI-TestStand/entry-point-Test-Batch-does-not-exist/m-p/989821#M26021

The solution however suggests that something is wrong with the <TestStand>\Components\Models\TestStandModels\ParallelModel.seq file. With only the Base Deployment Engine installed the 'Models' directory does not exist.

I don't know where to go from here. Am I supposed to manually define process models? Do I have to start the execution another way?

I'd appreciate any pointers, even if it's just some documentation I might have overlooked.

Cheers!

[iseGmbH]

I get a similar error when I try to address the Step to start at directly:

$ ./TestExec.exe dummy.seq -goto "Seq[\"MainSequence\"].Main[\"Property Loader\"]"

But when I click "Run MainSequence" manually, it runs just fine (nevermind the error, that's a concern for later):

Could it be that it's just not loaded properly so the runner doesn't find anything at all? This makes no sense to me, and the documentation isn't much help either 😕

[iseGmbH]

I revisited some of the video-courses and documentation I found yesterday and I think I am missing something that I need to deploy, but didn't figure out what it is yet.

Since this machine is meant to be a Jenkins test node which might run tests for different devices, I want to keep the local installation as clean as possible without deploying anything permanently. That's why I currently manually deploy into the test's working directory for each individual run, namely:

• the sequence files
• our custom step types (type palettes, icons and code modules incl. dependencies)
• the pre-compiled test runner (TestExec)

The TestStand runtime engine is NOT deployed individually, that's the only thing installed permanently on the test node.

I'm currently looking into anything related to process models since I'm thinking what I'm missing is related to that, but I'm not sure. All the documentation I found is either geared towards full development environments which have these things already installed, or deployment on completely "blank" machines.

If you have anything in mind that you think might be able to help me, please feel free to share 😊 I have no prior experience with TestStand and figuring that out completely by myself could probably take me a while 😬

[iseGmbH]

NI support was very helpful in figuring out the missing part. I don't know whether I missed it when installing TestStand on the agent or how it came to be, but the missing component was the TestStand Runtime.

This means using a deployment including the runtime, or a station-wide installation of it via NI Package Manager fixed this issue for me.

A package for this can be created for example using the TestStand Deployment Tool by creating a package-based deployment that does not include any files, but has the box "Include TestStand Runtime" checked (it might make sense to include some FrontendCallbacks or so if you're using the global environment).

Hopefully this will be helpful to someone else in the future 🙂