How does OPC Expert detect a failed connection to the primary OPC server and initiate a switchover to the standby?
OPC Expert uses continuous, built-in connection monitoring to detect failures on the primary OPC server path. When a failure is detected, it initiates a switchover to the standby server automatically, without requiring manual intervention, a separately configured watchdog tag, or a heartbeat signal from the underlying control system.
What OPC Expert Monitors
OPC Expert actively monitors the state of its connection to both the primary and standby OPC servers at all times. It tracks network-level connectivity, OPC server availability, and data quality. A failure can be triggered by any of several conditions: loss of network access to the server machine, the OPC server process terminating unexpectedly, the server becoming unreachable due to hardware failure, or the data itself becoming stale or invalid even when the communication path appears intact.
This last point is significant. OPC Expert does not only watch for connection drops. It can also detect that data has stopped changing when it should be changing, or that a value has crossed a threshold that indicates a process anomaly. This means you can configure failover to activate based on data behavior, not just on raw connectivity status.
Built-in and Custom Trigger Conditions
OPC Expert includes built-in automatic triggers that handle common failure scenarios without any user configuration. These cover connection loss, server unavailability, and bad data quality as reported by the OPC server itself. In addition, users can define custom trigger conditions tailored to their specific application. Custom triggers can be based on stale data detection, value thresholds, or calculated expressions involving multiple OPC items.
This combination of automatic and user-defined triggers means that OPC Expert can respond to both infrastructure failures and data-quality anomalies. An OPC server that is technically running and connected but is delivering frozen or out-of-range process values can still trigger a failover to the standby path.
How the Switchover Behaves in Practice
In Hot redundancy mode, because OPC Expert is already receiving data from the standby server simultaneously, the switchover is immediate and results in no data loss. The failover is transparent to any downstream application connected to OPC Expert’s embedded OPC server output.
In Warm mode, OPC Expert must signal the standby server to begin active data collection from the control system. This takes a short amount of time as the standby server initiates its subscription cycle. The transition window is measurable in seconds. In Cold mode, the standby server must be started before OPC Expert can establish a connection, resulting in a longer transition that depends on how quickly the standby environment comes online.
After failover, OPC Expert continuously attempts to restore the primary connection in the background. Once the primary server is available again, OPC Expert can optionally fail back to it automatically, depending on how the redundancy configuration is set up.
Related Tools
OPC Redundancy is the feature within OPC Expert responsible for all aspects of failover detection, switchover execution, and connection restoration. It handles both the monitoring logic and the communication management required to keep downstream applications receiving data through transitions.
Trigger OPC Events is a separate OPC Expert feature that allows users to configure action-based responses to OPC data conditions, including sending email alerts and executing scripts. While distinct from the redundancy failover logic, it can complement a redundancy setup by notifying operations staff when a failover event has occurred.
Conclusion
OPC Expert does not require a manually configured watchdog or heartbeat tag to detect primary server failures. It monitors connections and data quality continuously and initiates switchover automatically based on both built-in and user-defined trigger conditions. The result is a failover mechanism that covers network failures, server crashes, hardware problems, and data anomalies without requiring changes to the upstream OPC servers or the downstream application.