I have a OPC Client connection to the following endpoint: opc.tcp://10.208.7.50:4880/FANUC/NanoUaServer
Which is a FANUC robot. The point should be static.
Somewhere in network run we (think we) have intermittent connection which drives this OPC connection into faulted state. We have two options we are doing to recover:
- if we power off/on that end device, something in the internal power up re-establishes with Ignition and device can communicate.
- On this page: OPC UA Client Connection Settings - Ignition User Manual 8.1 - Ignition Documentation (inductiveautomation.com)
- If I follow steps 4 (leaving same Endpoint in place) through 11, the connection re-establishes without powering down the end device.
While we troubleshoot the root cause of lost comms, is there a way in Ignition to automatically run a script if connection is faulted, doing basically the steps done in 4 through 11?
OR would adjusting one of these Client Connections on that page hold the connection through intermittency long enough to not time out? This setting seems what I could change so that it keeps attempting to connect, but I'm not sure:
There should be a gateway status tag you can use to monitor the connection, and then possibly toggling the connection disabled/enabled using system.opc.setServerEnabled - Ignition User Manual 8.1 - Ignition Documentation would have the same effect.
You shouldn't need to do this, though. For troubleshooting, it's best to leave a rolling Wireshark capture running, so that the time period when the connection goes from working to not working is captured.
Without logs and/or a capture I can't recommend any settings changes.
OK I'll work on getting some captures when the event happens again! Can also try the .setServerEnabled.
Going from disabled to enabled on the OPC connection itself does not fix. The only way I've found to reestablish is going through that discovery prompt, hitting next on each prompt until Finish appears, then it reconnects. (steps 4 through 11). OR powering off/on end device.
I can't think of any reason going through discovery is different than just edit/saving the connection, regardless of whether you disable it in between or not.
There's probably something broken with this server. What version of Ignition is this? I know we've added some workarounds for wonky server behavior when reconnecting in this past...
Version: 8.1.33
Once this event happens again I can try a few more things to see what all triggers the reconnect. Can try Wireshark too.
EDIT:
This appears to be the event in the Ignition Logs, but also this could be the operator response powering off the machine AFTER the event occured. Hard to say without seeing the event live:
