What is Health Checking?
Health checking is an important part of load balancing
Health checking is a key function of an application delivery controller (ADC) or load balancer, ensuring that traffic is only routed to servers that are able to respond effectively. To perform a health check, the application delivery controller (ADC) polls the members of the server cluster to test whether an application is available. If a server responds successfully to a given number of probes, it is considered healthy, and remains eligible to receive new requests. If the server fails to respond successfully, it is considered unhealthy, removed from the cluster, and flagged for attention by IT.
The load balancing function can then redistribute traffic across the remaining members of the cluster to ensure high availability and optimal service. Once IT has addressed the underlying issue in the server and brought it back to health, it can be restored to the cluster and begin delivering apps again, with the load balancer redistributing traffic accordingly.
Load balancing health checks are enabled by a full-featured application delivery controller (ADC), and are also provided by cloud providers such as Google, Amazon, Microsoft, and Oracle. Indicators used to measure the ongoing health of servers can include such metrics as time series of total bytes in and out from each server; time series of traffic rates (in Mbps) in and out from each server; percent of error traffic over range; number of good SSL connections; average application server latency by service; client-side latency; and the response time for specific SQL database queries.
Key Takeaways
- Health checking is a load balancer function that continuously polls servers to verify they can respond to requests before routing traffic to them
- Servers that fail a set number of health probes are removed from the cluster and flagged for IT attention, preventing users from hitting degraded or down servers
- Once a failed server is fixed and restored, the load balancer automatically reintroduces it to the cluster and redistributes traffic accordingly
- Health metrics can include bytes in/out, traffic rates, SSL connections, application latency, and database query response times
- A10 Thunder ADC integrates health checking as part of its server load balancing and global server load balancing capabilities
How A10 Networks Supports Heath checking
The A10 Networks Thunder® ADC application delivery and load balancing solution ensures high availability and rapid failover with little or no downtime by continuously monitoring server health. The solution’s health checking functionality is part of a rich set of server load balancing and global server load balancing capabilities.
FAQs
Load balancers can perform several types of checks including TCP connection tests, HTTP/HTTPS request probes, DNS query checks, and application-specific checks that validate actual service responses rather than just network connectivity.
Depending on the load balancer configuration, existing sessions may be allowed to complete (graceful removal) or immediately redirected to healthy servers, with the approach varying based on session persistence settings and the severity of the failure.
Active health checking proactively sends test probes to servers at regular intervals, while passive health checking monitors real traffic responses to detect failures. Many ADCs use both in combination for more accurate detection.
Health checking is a foundational component of high availability. By automatically detecting and removing unhealthy servers, it ensures the load balancer only routes traffic to servers that can deliver reliable responses, minimizing downtime for end users.
Yes. Advanced ADCs allow custom health check scripts that validate application-specific responses, for example, checking that a specific string appears in an HTTP response or that a database query returns the expected result, going beyond simple ping or TCP checks.