Gateway Connection Lost
Master is not available
java.net.ConnectException: Connection refused
Backup connection status is unknown.
Searching for active Gateway
We also noticed that the Vision Client running on the primary PC displays http://0.0.0.0:8088 as the Master address in Diagnostics, while the secondary PC correctly displays 192.168.54.5:8088.
Troubleshooting Performed
We have already tried the following:
Verified Gateway redundancy configuration and synchronization.
Confirmed that the Backup Gateway correctly becomes Active when the Master stops.
Performed redundancy resynchronization successfully.
Deleted the Ignition/Vision Client cache on Windows 11.
Downloaded and reinstalled the Vision Client Launcher.
Reconfigured and restarted the Vision Client, pointing it to the correct Master IP address.
Despite these attempts, the issue persists, and we have not been able to identify the root cause.
Questions
Is there any additional configuration required in Ignition 8.3.9 for automatic Vision Client failover?
Why does Vision report "Backup connection status is unknown" even though Gateway redundancy is working correctly?
Could this be related to the Gateway HTTP address configuration or Vision Client Launcher settings?
Is this a known issue or regression in Ignition 8.3.9?
The Gateway redundancy mechanism seems to be working correctly, but the Vision Clients do not appear to recognize the Backup Gateway.
Any suggestions or troubleshooting recommendations would be greatly appreciated.
What do you gateways report as their public HTTP(S) address? Do the gateways talk to each other over the same IP addresses that clients use? (The latter is generally required to use "automatic" in your gateway public address setting.)
For best results, use DNS names everywhere in your settings, set each gateway to use its DNS name as its public address (not automatic), and ensure the DNS names resolve correctly everywhere you run clients.
I’ve done some additional testing using Auto Detect HTTP Address.
The Master Gateway now correctly reports its IP address, but the Backup Gateway still shows “Connection Status: Unknown” in the Vision Client diagnostics.
I can also reproduce the issue by manually triggering a Request Failover from the backup server.
When I do this:
The Backup Gateway correctly becomes active.
The redundancy failover appears to work correctly at the Gateway level.
However, the Vision Client does not reconnect to the Backup Gateway and remains in an error state, apparently because it doesn’t know the backup server’s address.
In the Vision Client diagnostics, I currently see:
Master: Connected to http://192.168.54.5:8088
Backup: Connection status is unknown.
Both Gateway IP addresses are reachable from the client network.
I understand that using DNS names instead of IP addresses is recommended, but since both gateways and clients are on the same local network, I would first like to understand why the backup address is not being detected correctly.
Is there a specific setting in Ignition 8.3 where I need to configure the Backup Gateway’s public HTTP address so that Vision Clients can discover it correctly?
Do you mean those settings?
(I’m not sure whether you’re referring to the Public HTTP Address under Network > Webserver > Public HTTP Address, or to the settings shown in the attached screenshots.)
First pic is Master node, second pic is Slave node.
if I request a failover, everything works perfectly from the Gateway side. The only issue is with the Vision client, which, for some reason, reports the backup connection status as “Unknown.”
This project was upgraded from version 8.1, so I’m wondering if I might be missing some new configuration that needs to be set up
Checking the vision configuration file on the master node the configuration looks good to me:
Sorry, I'm used to our old configuration model and web interface.
Web server settings are one of the things that are not synchronized between master and backup (because by definition they don't make sense to synchronize). So navigate to the backup gateway and in the same location you configured an explicit public address for the master, populate an explicit public address for the backup.
Ultimately this configuration ends up in the gateway.xml file on disk.
First of all, thank you all for your suggestions and for the time you dedicated to helping me.
After double-checking all the configuration parameters and finding no obvious errors, I decided to completely uninstall Ignition from both the Master and Backup PCs, remove the remaining configuration and cache folders, and perform a clean installation of both Gateways.
Once both Gateways were reinstalled, I configured the redundancy using exactly the same IP addresses and settings as before.
This time, everything worked correctly right from the start, both at the Gateway redundancy level and within the Vision Clients.
Since I was not personally involved in the original installation and configuration of these PCs, I cannot confirm whether any changes were made to the original configuration over time, or whether any version upgrades, including major version upgrades, had been performed since the initial installation.
In any case, after performing a clean installation, the system immediately behaved exactly as expected.
Just a quick update that might be useful to someone else in the future.
It seems that the entire issue was caused by a mismatch in the installed module versions after applying my two licenses to the primary and backup Ignition gateways.
Once Ignition Customer Support corrected and aligned the module versions, everything started working as expected again.
One additional note: as soon as the licenses were applied, the Vision runtime immediately resumed communication and correctly detected the backup gateway status, without needing to delete any cache folders or restart the Vision Client.
