A service virtual machine provides service to any number of virtual machines on a hypervisor over a first communication channel. When an anomaly is detected within the provided service, any virtual machine using the first communication channel switches to a second communication channel and receives service from a second virtual machine. The second virtual machine may execute upon the same computer or on a different computer. Hooking points within the hypervisor provide a means for the service virtual machines to monitor traffic and provide service to the protected virtual machines. When a service virtual machine is suspended, it is repopulated, upgraded or rebooted, and then restored to service. Once restored, any protected virtual machine may be switched back to the restored service virtual machine. Virtual machines may be switched to a different communication channel by modifying a configuration file. Both communication channels may be in use at the same time.