Method and apparatus for defining failover events in a network device
View Patent ↗A critical interface may be defined for a network device such that if the critical interface goes DOWN, the network device will perform a forced failover. IP addresses are assigned to interface groups, and interface groups are assigned to a critical interface. The critical interface will go DOWN if any one of the interface groups goes DOWN. The interface groups will not go DOWN, however, unless all members' IP addresses assigned to the interface group go DOWN. By configuring the critical interface in this manner, the network manager has increased flexibility in defining which events should and should not trigger failover of the network device. Additionally, combinations of events may be grouped to enable the network manager to take into account fairly complex failure scenarios and specify, with precision, the action to be taken by the network device under myriad possible situations.
1. A method of forcing failover of a network device, the method comprising the steps of:
administratively defining at least two interface groups, each interface group being defined via a management system as a logical association of at least two members, wherein said interface groups will not have a state of “down” unless all members administratively defined as being included in the interface group have a state of “down”;
administrative assigning, via the management system, the at least two interface groups to a critical interface of the network device, wherein a state of the critical interface will be “up” if all interface groups assigned to the critical interface have a state of “up”, and the state of the critical interface will be “down” if the state of any one of the interface groups assigned to the critical interface have a state of “down”; and
forcing a failover of the network device upon a change in state of the critical interface to down, such that a change in state to “down” of any one of the interface groups assigned to the critical interface will force a failover of the network device and, hence, of all other interface groups assigned to the critical interface.
2. The method of claim 1 , wherein a first interface group assigned to the critical interface contains a tunnel interface, and a second interface group assigned to the critical interface contains an IP address associated with an entity on a network other than the tunnel interface, such that a change in state of the IP address to “down” will cause a forced failover of the network device interfaced to the tunnel despite the fact that the tunnel has not changed state to “down”.
3. The method of claim 2 , wherein the entity is a firewall interface or an LDAP/RADIUS interface.
4. The method of claim 2 , wherein the interface group may be associated with more than one network device.
5. The method of claim 2 , wherein the interface group may be assigned a name by a network manager.
6. The method of claim 1 , wherein defining at least one of the interface groups comprises selecting, by a network manager, at least two members from:
one or more physical interfaces,
one or more IP addresses, and
one or more tunnel interfaces,
wherein the interface group is defined such that a failure of all of the members selected to be part of the interface group is required to cause the state of the interface group to be “down”.
7. The method of claim 1 , wherein the network device is a physical network device.
8. The method of claim 1 , wherein the network device is a logical network device.
9. A network device, comprising a computer-readable medium containing instructions for controlling at least one processor to perform a method of forcing a failover of the network device under defined conditions, the method comprising the steps of:
administratively defining at least two of interface groups for the network device, each interface group being defined via a management system as a logical association of at least two members, wherein said interface groups will not have a state of “down” unless all members administratively defined as being included in the interface group have a state of “down”;
administratively assigning, via the management system, the at least two interface groups to a critical interface of the network device, wherein a state of the critical interface will be “up” if all interface groups assigned to the critical interface have a state of “up”, and the state of the critical interface will be “down” if the state of any one of the interface groups assigned to the critical interface have a state of “down”; and
forcing a failover of the network device upon a change in state of the critical interface to down, such that a change in state to “down” of any one of the interface groups assigned to the critical interface will force a failover of the network device and, hence, of all other interface groups assigned to the critical interface.
10. The network device of claim 9 , wherein a first interface group assigned to the critical interface contains a tunnel interface, and a second interface group assigned to the critical interface contains an IP address associated with an entity on a network other than the tunnel interface, such that a change in state of the IP address to “down” will cause a forced failover of the network device interfaced to the tunnel despite the fact that the tunnel has not changed state to “down”.
11. The network device of claim 10 , wherein the entity is a firewall interface or an LDAP/RADIUS interface.
12. The network device of claim 10 , wherein the interface group may be associated with more than one network device.
13. The network device of claim 10 , wherein the interface group may be assigned a name by a network manager.
14. The network device of claim 9 , wherein the defining at least one of the interface groups comprises selecting, by a network manager, at least two members from:
one or more physical interfaces,
one or more IP addresses, and
one or more tunnel interfaces; and
wherein the interface group is defined such that a failure of all of the members selected to be part of the interface group is required to cause the state of the interface group to be “down”.