OPC tags stay unhealthy even after device is reconnected


We have Ignition Edge installed on a moxa switch which is connected to a Modbus TCP device. I created a device under config>Opcua>Devices. Everything is working fine. Every now and then if I power cycle the whole panel (Edge device + Modbus device), my tags change to “Bad_NotConnected”. If I then go to config>Opcua>Devices I see the device is connected. If I then go to config>opc>OPC Quick client, I can drill down to the device and r/w any value without any issues. The problem is that all my tags stay “Bad_NotConnected” until I restart the edge device or re-enable the device under config>Opcua>Devices (sometimes) or restart the opc ua module.

It is as if the GW eventually makes connection to the Modbus device but the tags don’t. Any Idea what could cause/solve this issue?

I did set Max Retry Count = 10 (config>Opcua>Devices>MyMBDevice>edit) but that didn’t seem to help.

If I need to include any additional information, please let me know. The issue is that this system is going to be running on a remote site from the end of this week and if for some reason a power loss/cycle occurs on site, I will have to remotely login and reset whatever is needed to get all the opc tags to become healthy again.

What version of ignition?

Ignition Edge Version 8.1.3
And module drivers are shown below:

Hmm. 8.1.5 includes a fix for something that sounds a bit like this…

I went and took a look at the changelog the only thing I could see maybe relating to the issue was the following:


2044: TCP Driver can saturate shared execution engine if many devices are defined and unavailable
To prevent saturation of the shared execution engine thread pool, the TCP Driver now uses the opc-ua-executor thread pool to connect.

But since 8.1.4 is the current stable build and since this system is now up and running, would it be best if I wait for 8.1.5 stable build?

Hmm, you’re right, I don’t see the entry I’m talking about. I’m trying to find out why it’s not there.

Are you thinking of the Allen Bradley driver issue instead?

No, we missed a couple change log entries.

The one I’m thinking of will read something like this:

Fixed a bug that caused tags to get stuck with Bad_NotConnected or Uncertain_InitialValue quality on startup or reconnect.

This one?

Nightly Changelog: 8.1.5-b20210413 - addendum

2081: OPC tags get stuck with Bad_NotConnected or Uncertain_InitialValue on startup or reconnect
Fixed a bug that caused tags to get stuck with Bad_NotConnected or Uncertain_InitialValue quality on startup or reconnect.

Yeah that one. It wasn’t in there this morning :slight_smile:

Yes that definitely got added or I missed it multiple times. Thank you for clearing this up for me guys.