System health monitoring and recovery

Claim 2

A network device, comprising: a least one communication port to allow communication with at least one other network device; a storage to store a rules database; and at least one health monitor module to: maintain a system health metric and to receive an event notification of a task performance; alter the health metric based upon the outcome of the task performance, wherein there are multiple tasks that affect the health metric; and access the rules database for a matching rule for the event notification and perform any tasks determined by the rules database, including causing a switch over to a standby unit if the health metric for a first unit falls below a predetermined threshold, wherein the switch over is prevented if a health metric for the standby unit is lower than the health metric for the first unit.

Claim 3

The network device of claim 1 , wherein the unit further comprises a supervisory engine.

Claim 4

The network device of claim 1 , wherein the rules database includes a set of rules for each unit.

Claim 5

The network device of claim 1 , wherein the network device further comprises several subsystems and each unit includes a health monitor.

Claim 6

The network device of claim 4 , wherein the health monitor on one unit is designated as a central health monitor and receives at least one health metric from each of the other units.

Claim 9

The method of claim 6 , wherein establishing a health metric further comprises establishing a health metric for each of several subsystem units.

Claim 10

The method of claim 7 , wherein a central health monitor receives the health metric for each subsystem and produces a system health metric.

Claim 11

The method of claim 6 , wherein applying input rules further comprises: adding values to the health metric for successful task performance; and subtracting values from the health metric for unsuccessful task performance.

Claim 12

The method of claim 9 , wherein the values are weighted depending upon the task.

Claim 13

The method of claim 6 , wherein the health metric further comprises a normalized value between 0 and 1.

Claim 14

The method of claim 6 , wherein the action rules are invoked when the health metric for a first unit passes a predetermined threshold.

Claim 15

A network device, comprising: a means for allowing communication with at least one other network device; a means for storing a rules database; and at least one means for: maintaining a system health metric for receiving an event notification about a task performance; altering the health metric based upon the outcome of the task performance, wherein there are multiple tasks that affect the health metric; accessing the rules database for a matching rule for the event notification; and performing any tasks determined by the rules database including altering the health metric in response to the event, wherein the health metric is altered by any one of multiple events and including causing a switch over to a standby unit if the health metric for a first unit falls below a predetermined threshold such that the switch over is prevented when a health metric for the standby unit is lower than the health metric for the first unit.

Claim 16

A network device, comprising: a means for allowing communication with at least one other network device; a means for storing a rules database; and at least one means for: maintaining a system health metric for receiving an event notification about a task performance; altering the health metric based upon the outcome of the task performance, wherein there are multiple tasks that affect the health metric; accessing the rules database for a matching rule for the event notification; and performing any tasks determined by the rules database including altering the health metric in response to the event, wherein the health metric is altered by any one of multiple events and including causing a switch over to a standby unit if the health metric for a first unit falls below a predetermined threshold such that the switch over is prevented when a health metric for the standby unit is lower than the health metric for the first unit.

Claim 17

The network device of claim 13 , wherein the network device further comprises at least one supervisory engine.

Claim 18

The network device of claim 13 , wherein the rules database includes a set of rules for each unit.

Claim 19

The network device of claim 13 , wherein a health monitor on one unit is designated as a central health monitor and receives at least one health metric from any other units.

Claim 20

An article containing machine-readable code that, when executed, causes the machine to: establish a health metric; apply input rules to the health metric, wherein the rules alter the health metric based upon task performance of each of several tasks; and invoke action rules, wherein the action rules are invoked depending upon the health metric, wherein the action rule causes a switch over to a standby unit if the health metric falls below a predetermined threshold wherein the switch over is prevented if a health metric for the standby unit is lower than that of the first unit.

Claim 21

An article containing machine-readable code that, when executed, causes the machine to: establish a health metric; apply input rules to the health metric, wherein the rules alter the health metric based upon task performance of each of several tasks; and invoke action rules, wherein the action rules are invoked depending upon the health metric, wherein the action rule causes a switch over to a standby unit if the health metric falls below a predetermined threshold wherein the switch over is prevented if a health metric for the standby unit is lower than that of the first unit.

Claim 22

The article of claim 17 , wherein the code, when executed, causes the machine to establish a health metric for each of several units.

Claim 23

The article of claim 17 , wherein the code, when executed, causes the machine to designate a central health monitor and causes the central health monitor to receive the health metric for each unit and produces a system health metric.

Claim 24

The article of claim 17 , wherein the code that causes the machine to apply input rules further causes the machine to: add values to the health metric for successful task performance; and subtract values from the health metric for unsuccessful task performance.

Claim 25

The article of claim 17 , wherein the code that causes the machine to add values further causes the machine to weight the values depending upon the task.