WebService in VeriStand

[thumble]

I'm building a real-time PXI frontend based on VeriStand. All specific functions are embedded into custom devices. The frontend must be stand-alone capable (that means it must start automatically after power on, without any computer connected). For some reasons I plane to use web services to have some remote control (I can't use the VeriStand gateway) and I wonder how to include this configuration into my VeriStand project, so it can be managed and (re)deployed along with the frontend code, as a whole software part.

Thx in advance for help.

[Mathieu_P.]

Hello,

I'm note sure of what you are looking for. Do you want something like Remote Front Panel in LabVIEW ?

I ask that because Web Services in LabVIEW don't have a front panel.

Regards,

[thumble]

I our case the architecture is one remote control node that must interract with many server nodes and not only through MMI commands (the remote control code must be able to exchange data with many nodes). Furthermore the remote control node can be turned off without incidence on server nodes. Currently we have changed our mind and plane to use shared variables to do that. But the initial question remains (but is no more an urgence) and can be reformulated as: how to deploy a web service on a VeriStand realtime node along with the VeriStand project.

[Mathieu_P.]

Hello thumble,

Why can't you use the Veristand Gateway ? The goal of the gateway is to provide access from multiple host to the RT target without disrupting the determinism.

It seems appropriate.

[thumble]

We would have preferred, but we can’t because:

Solution 1 (The preferred): one single global VeriStand project with 11 targets => when any of the targets is down, the Gateway shuts down and we lose control other the 10 remaining ones. Furthermore if any of the targets is unavailable (i.e. for maintenance reason) we can’t deploy any of the other.

Solution 2 (the selected one): one separate VeriStand project for each target => since the gateway can be connected to only one target at given time, we would have to connect the target N, read the channels, disconnect, and go on with next one (11 times). That seems not efficient.

The best would be to have a fine control of the targets in a multi-target configuration: ability to deploy each target individually and Gateway able to tolerate the loss of a target and even to reconnect it when it is rebooted (with project auto-start configured).