In the digital world, network service disruptions are inevitable, even affecting cloud platforms and content delivery networks with 100% uptime guarantees. The key question is how organizations respond to such outages. Will they seamlessly switch to redundant services, ensuring uninterrupted user experience? And if so, will this failover process be automated or manual?

Midsize and large organizations often have redundant systems in place for survival during outages, but what they might lack is an automated mechanism to redirect traffic to these backups when a primary service fails.

IBM NS1 Connect Filter Chain™ technology leverages DNS capabilities to automatically divert traffic between service providers in the event of network disruptions. By setting up basic rules, NS1 Connect monitors network status and switches endpoints as needed, with predefined rules and priorities ensuring a seamless process.

On the NS1 platform, filter chain configurations are applied to individual DNS zone records, determining how NS1 handles queries for each record. These filter chains employ unique logics to process queries and offer flexibility in achieving specific outcomes based on operational or business requirements.

To cater to different preferences in directing failover traffic, a guide has been compiled on constructing active-active, active-passive, and manual failover systems using filter chains.

Active-active Failover

In an active-active scenario, NS1 or third-party sources monitor the status of endpoints in an application delivery infrastructure. When an outage is detected, NS1 automatically directs traffic to selected secondary systems that are already operational as part of the load balancing setup. This dynamic process ensures continuous operation by redistributing load to functioning systems.

The filter chain starts with an “Up” filter, indicating the operational status of service provider endpoints.

Subsequently, a “Shuffle” or “Weighted Shuffle” filter is applied to distribute traffic to functioning providers if the “Up” filter identifies a non-operational endpoint. The distribution can be random (Shuffle) or based on preset weights (Weighted Shuffle).

Lastly, the “Select First N” filter determines the number of answers to provide to inbound queries, which typically aligns with RFC 1912, specifying a single answer per CNAME query.

Active-passive Failover

In an active-passive setup, monitoring data from NS1 or external sources guides the routing of traffic to secondary systems when primary systems fail. Unlike active-active, the secondary systems are only activated when needed, serving as backup options.

Similar to active-active, an “Up” filter is initiated to determine the online status of services based on monitoring data.

A “Priority” filter prioritizes active systems over passive or backup systems, sorting higher priority answers to the top of the list. If no higher priority options are available, NS1 progresses down the list until an active resource is located.

Similar to the active-active failover scenario, the “Select First N” filter dictates the number of answers provided, typically one.

Manual Failover

For cases requiring manual intervention in failover decisions, the filter chain system is utilized, enabling manual activation when targeting traffic redirection. In this approach, the “Up” filter allows manual status assignments for services.

Following the same logic as active-passive failover, the “Priority” filter organizes active systems ahead of passive alternatives. Available resources are ranked based on priority, with the system progressing through the list until a suitable resource is identified.

Just like in previous cases, the “Select First N” filter determines the number of responses provided, commonly set to one.

Multi-cloud or Multi-CDN Availability

In more complex scenarios where service availability nuances need consideration, the filter chain system provides a nuanced approach to traffic steering. Regional outages or performance issues can lead to low-quality service experiences, warranting a more detailed availability assessment.

The “Pulsar Availability Threshold” filter introduces a percentage-based metric to manage service usage based on availability metrics.

Subsequently, the “Weighted Shuffle” filter allocates traffic to available providers meeting defined availability criteria, weighted according to specific priorities.

The “Pulsar Performance Sort” filter directs traffic to the fastest available services based on the preceding filters, excluding low-performing services as per set thresholds.

Conclusively, the “Select First N” filter again specifies the number of responses provided, typically one.

For comprehensive insights on leveraging filter chains to enhance performance, ensure resilience, and reduce costs, delve deeper into the NS1 Connect platform.

### Summary:
This guide delves into the NS1 Connect Filter Chains, a technology by IBM that enhances network resilience by automatically rerouting traffic in the event of service disruptions. By exploring active-active, active-passive, and manual failover scenarios, organizations can strengthen their network redundancy and optimize user experience during outages.

### FAQs:

**Q: How does NS1 Connect Filter Chains help in network resilience?**
A: NS1 Connect Filter Chains use DNS capabilities to automatically redirect traffic between service providers during network disruptions, ensuring seamless failover processes.

**Q: What are some common types of failover systems outlined in the article?**
A: The article discusses active-active, active-passive, and manual failover systems using NS1 Connect Filter Chains.

**Q: How does the “Select First N” filter impact inbound queries?**
A: The “Select First N” filter specifies the number of answers provided to inbound queries, typically aligning with industry standards mandating a single answer per query.