Running a Master/Backup pair Ignition 8.1.39 using MS-SQL setup as 2 nodes AlwaysON redundancy. No domain controller nor using MS-SQL listener. Ignition is setup with a primary DB connection to SQL server on master gateway and a backup DB connection to SQL server on backup gateway. The failover datasource for the primary DB connection is the backup DB connection and vice-versa.
This setup is working fine, and the failover/failback works as expected. The MS-SQL is 2019 RTM version. This setup has been running for several years in an airgap setup. Since we are missing all the cumulative updates since RTM we decided to update to the latest CU, SQLServer2019-KB5054833-x64.
No other software or environment had changed besides MS-SQL CU patch. We stopped Ignition gateway using the batch file starting with the backup gateway, then the master gateway. We then proceed to update the CU to the secondary SQL replica, failover the SQL after the CU completes then update the previous primary replica which is now the secondary after failover. After CU completes, failback the replica, restoring the normal primary/secondary relationship. Starts the master gateway using the batch file, then starts the backup gateway.
The primary DB connection shows valid, but the secondary DB connection shows ERROR. However, if we failover the SQL replica, the secondary DB connection then shows "valid" with the primary DB connection continue to show valid. Failing back the replica did not change the "valid" status for either DB connection to ERROR. However, if the servers or gateways are restarted, the secondary DB connection again reports "ERROR" until the failover occurs. Functionally this seems to work, but the "ERROR" status raises concern since it doesn't reflect the true operational status.
I am attaching the ERROR message and stack trace from Ignition.
20260724-DBConnErrorStackTrace-SQLServer2019-KB5054833-x64.txt (4.1 KB)
20260724-DBConnError-SQLServer2019-KB5054833-x64.pdf (41.2 KB)