Does the remote engine execute callbacks?

[pwaszkewitz]

As far as I know, a TestStand remote execution runs without a process model, so it will certainly not do ProcessModel callbacks. But will the execution of a step perform Engine callbacks in a remote execution?

I am especially interested in "Post-Step Failure engine callback" to be able to use the same error handling as in a local execution and "Post-Result List Entry engine callback".

Regards

Peter

[Dippi]

Hi Peter,

so on an other machine, you've got another instance of the TestStand engine, the two engines (on two machines) don't know anything about eachother's callbacks.
They're completely separate, execept for the parameters of the sequence that was called.
So if you run an execution entry point of a process model on the remote machine, the callbacks run in the sequence will be the callbacks of the process model on the remote machine

It should work if you duplicate the callbacks from the main machine in the sequencefile of the remote machine in order to get them to be executed. Example is PostResultListEntry....

Best regards

Dippi 

[pwaszkewitz]

Hi Dippi,

sorry that took so long. I tested the following:

• defined a sequence with all callbacks filled with a message box (Engine callbacks like SequenceFilePreStep, PostStep, as well as ProcessModel Callbacks, like PreUUT, Report options ...)
• configured both machines to use the standard NI Sequential Model
• put the sequence on both machines in the same (relative) path

I found that upon a SequenceCall with the option "On a remote computer" the EngineCallbacks (SequenceFilePreStep, PostStep etc.) are executed on the, the ProcessModel callbacks are not. Which is not really surprising as the call directly executes a sequence, i.e., does not use an Execution Entry Point.

How can I use an Execution Entry Point on a remote machine? By calling a sequence there which in turn does a local sequence call in a new execution?

Best regards

Peter

[CorbinH]

Peter,

You are right about using a remote sequence call that in-turn uses a new execution. Use a remote sequence call that calls a sequence file on the remote computer that only has another sequence call located in that sequence file. You can then pick which sequence file you would like ro run on the remote computer, the process model you would like to use, and which entry point to use.

[pwaszkewitz]

Corbin,

thank you for confirming this. Now I can think about what the best approach will be for us.

Peter