Reducing service disruption using multiple virtual IP addresses for a service load balancer

A service node comprising:

• one or more processors operably coupled to a memory;
• a forwarding unit configured for execution by the one or more processors;
• a service load balancer implemented in the forwarding unit;
• a control unit, configured for execution by the one or more processors, that maps a first virtual address for a particular service to a plurality of private addresses that identify a plurality of servers, wherein each server provides the particular service, and wherein a Uniform Resource Locator (URL) for the service resolves to the first virtual address;
• wherein the control unit configures the service load balancer of the service node to load balance network traffic for the first virtual address across the servers and forwarding, by the service node, network traffic for the first virtual address between the servers and a plurality of subscriber devices;
• wherein the control unit, in response to determining that at least one server of plurality of servers will no longer service new connections for the service from the subscriber devices, associates a second virtual address with a subset of the private addresses that excludes the private address for the at least one server, and causes the URL to resolve to the second virtual address;
• and
• wherein the forwarding unit, in response to configuring the service load balancer of the service node to load balance network traffic for the second virtual address across the subset of the servers, forwards network traffic for the first virtual address between the servers and the subset of the subscriber devices.

A non-transitory computer-readable medium comprising instructions for causing at least one programmable processor of a service node to:

• maps a first virtual address for a particular service to a plurality of private addresses that identify a plurality of servers, wherein each server provides the particular service, and wherein a Uniform Resource Locator (URL) for the service resolves to the first virtual address;
• configures a service load balancer of the service node to load balance network traffic for the first virtual address across the servers and forwarding, by the service node, network traffic for the first virtual address between the servers and a plurality of subscriber devices;
• in response to determining that at least one server of plurality of servers will no longer service new connections for the service from the subscriber devices, associates a second virtual address with a subset of the private addresses that excludes the private address for the at least one server, and causing the URL to resolve to the second virtual address;
• and
• in response to configuring the service load balancer of the service node to load balance network traffic for the second virtual address across the subset of the servers, forwards network traffic for the first virtual address between the servers and the subset of the subscriber devices.