Best Practices for Global Data watchdog?

[TCC-Systems]

Native "Health Timeout" and "Last Value" settings in Concept/Unity/Control Expert IO Scanner allows inter-PLC communication with fallback to "0" in event of comms loss (built-in-watchdog)

I've now transitioned to Global Data (Multicast Publisher/Subscriber inter-PLC comms).  Do I need to build my own watchdog bit into the Global Data DDTs or is there a similar "fallback to 0" feature?

[Trinxs1]

I would strongly recommend installing a watchdog as the first or last register in the data transfer from each PLC.  I would also ensure that each participating PLC publishes at least a watchdog.

1. Publishing and subscribing are functions of the global data card but it does not confirm that the CPU is running.  It can be in halt and the data will still be published.
2. If you wish to confirm that the subscribers are active and running the only way for a publisher to confirm this is by subscribing to data the device publishes.