Redundancy is a broad concept, as explored in our Demystifying Redundancy Blog. In this post, we narrow the focus on how to manage redundant OPC servers—a critical topic for ensuring reliable process communications to your HMI, SCADA, MES, Historian, etc. Redundancy is crucial because it ensures continuous operation and data integrity in the event of system failures, minimizing downtime and potential losses.
Simply deploying two OPC servers connected to your PLCs doesn’t make them redundant. We're sometimes asked about having the two OPC servers communicate and decide when to failover. That sounds good but has a flaw. If the client loses network connection to one OPC server but the two OPC servers are still communicating, and the HMI/SCADA doesn’t have any internal failover mechanism, then no failover will occur.
While some HMI/SCADA systems offer native redundancy support and can make that decision, it often involves scripting or custom code, making the process more complex. That’s where supplemental software like Cogent DataHub steps in, simplifying redundancy by managing connections to both OPC servers. Your OPC client interacts with DataHub as if it were a single server, abstracting away the complexity. With the proper architecture for the location of the Cogent DataHub, ideally on the same computer as the HMI/SCADA server, the scenario mentioned earlier would not be a problem.
In this blog, we'll discuss 4 key steps for managing OPC server redundancy and achieving high availability and reliability in your server communications using Cogent DataHub.
1. Define your OPC Server Redundancy Requirements
To effectively manage redundancy, start by determining your specific needs. Here are some key decisions to make:
- Primary vs. Secondary Servers:
- Identify which server will act as the primary and which as the secondary failover.
- Identify which server will act as the primary and which as the secondary failover.
- Failover Triggers and Criteria:
- Decide what conditions should trigger a failover. Common triggers include:
- Poor tag quality (e.g., Bad or Not Good).
- Connection loss (Not Connected state).
- Stalled updates to critical values for a set time.
- For setups with redundant network paths, ensure your architecture accounts for potential device communication loss along one path.
- Decide what conditions should trigger a failover. Common triggers include:
- Return to Normal Operation:
- Define how and when the system should revert to the primary server once it’s operational.
Balancing sensitivity is critical—triggers that are too strict may cause unnecessary failovers, while lenient ones risk excessive downtime.
2. Choose the Best Redundancy Method for your System
Cogent DataHub supports different redundancy options for OPC UA and DA servers, allowing flexibility for your specific business needs. Here's how they work:
- Hot Redundancy: DataHub maintains active connections to both primary and secondary servers. When the primary fails, DataHub instantly switches to the secondary, providing seamless data to OPC clients. This approach minimizes failover times to mere milliseconds but requires both downstream OPC servers to poll devices simultaneously, which could increase network and hardware load.
For scenarios where reducing network load is a priority, DataHub supports warm and cold redundancy.
- Warm Redundancy: Connections are pre-established but activated only during failover.
- Cold Redundancy: Connections are established and activated only after failover, resulting in longer switch times.
Choosing the best method depends on your tolerance for downtime versus added network traffic.
3. Implement your OPC Server Redundancy Configuration
4. Extend Redundancy Beyond your OPC Servers
DataHub’s redundancy features extend past OPC servers. It can manage any two identical data sets across supported sources like OPC A&E, Modbus, ODBC, and even custom applications via the DataHub API. For example, you can mix protocols (e.g., OPC DA as primary and Modbus as secondary) as long as tag names match.
DataHub also provides real-time status and control over redundancy through system data points. These tags let you:
- Monitor the active connection.
- Force a failover by writing to the “preferred source” tag.
- Trigger email/SMS alerts during failovers to stay informed.
- Log failover events for historical tracking and analysis.
Ready to Elevate your OPC Server Redundancy?
Cogent DataHub makes managing redundancy straightforward while offering a host of other features to enhance your integration projects. From data bridging to real-time notifications and gateway solutions for MQTT and External Historians, DataHub is a versatile tool for modern industrial systems.
Be sure to contact our team to discuss your specific needs and subscribe to our blog today!
Don’t have Cogent DataHub yet? Download your free trial today and see how it can simplify your redundancy management.
