Cisco switch monitoring

[PFrog]

Diagnostic feedback shouldn't inform how you respond to a connection loss in the first place.

If we have systems shut down they could affect equipment that rely on them, I almost have to treat a comm loss as a central system went offline. I'm trying to determine the difference between a network issue, or a system was powered down/failed. If we have network outages/dropouts, I would like to keep the equipment going, and ignore the bad data from the RIO units.

 

[JeremyM]

a) Cable between PLC and IDF broken.
b) IDF down for service.
c) IDF down for outage.

Distinguish these with GSV, ping, or any other network diagnostic feedback. Spoiler: You can't because the diagnostic itself relies upon network integrity.

You need hardwired, maintained relay feedback from the switch to PLC backplane IO, for example. Commercial appliances typically don't implement relays, so managed industrial ones are needed.

The "next best" option is to infer service downtime by schedule, but you'd be chasing the problem any time the window is missed or changed. Or daylight savings.

 

[PFrog]

a) Cable between PLC and IDF broken.
b) IDF down for service.
c) IDF down for outage.

Distinguish these with GSV, ping, or any other network diagnostic feedback. Spoiler: You can't because the diagnostic itself relies upon network integrity.

You need hardwired, maintained relay feedback from the switch to PLC backplane IO, for example. Commercial appliances typically don't implement relays, so managed industrial ones are needed.

The "next best" option is to infer service downtime by schedule, but you'd be chasing the problem any time the window is missed or changed. Or daylight savings.

Network integrity/cabling issues with the switches is not the main issue. The main issue is the switch dropping out due to scheduled/non-scheduled maintenance (at least this is what I have seen so far). I'm aware of the issues with this. Again, I'm looking to see if someone is aware of a device I can plug in/ or some type of coding that I can apply to verify switch comms at its port/address. If it doesn't exist that is fine, I'll probably use a RIO unit within its panel and verify that for switch status. It would always be powered up (switch power).

 

[Nova5]

Thats why it would be and is generally best to have your own switches for the controls system. IT cannot muck about with them and cause system disruptions that if things react unexpectedly something is broken or someone ends up injured.