Method of improving the availability of a computer clustering system through the use of a network medium link state function
A method for improving the availability and reliability of a computer clustering system having first and second servers that are connected to each other by a communication link and that each have an associated mass storage device. In one implementation, the method begins when the first server detects a loss of communication over the communication link from the second server to the first server that prevents the servers from communicating with each other. The first server then analyzes the communication link to determine if the communication link is functioning properly. If the communication link is functioning properly, then the first server continues operation, assuming a right to survive without communicating with the second server or with the mass storage device of the first server or with the mass storage device of the second server, and the first server services network requests that would otherwise be serviced by the second server.
1. A method for improving the availability and reliability of a computer clustering system having first and second servers that are connected to each other by a communication link and that each have an associated mass storage device, the method comprising:
the first server detecting, over the communication link, a loss of communication between the first and second servers;
the first server analyzing the communication link to determine if the communication link is functioning properly; and
if the communication link is determined to be functioning properly, the first server:
continuing operation;
assuming a right to survive without communicating with the second server or with the mass storage device of the first server or with the mass storage device of the second server;
servicing a network request that would otherwise be serviced by the second server; and
the first server discontinuing operation if the communication link is determined to be not functioning properly.
2. The method as recited in claim 1 , wherein the right to survive is associated with the second server prior to any assumption of the right to survive by the first server.
3. The method as recited in claim 1 , wherein detecting a loss of communication over the communication link comprises determining that a heartbeat signal associated with the second server cannot be detected on the communication link.
4. The method as recited in claim 1 , wherein detecting a loss of communication over the communication link comprises:
the first server attempting to communicate with the second server across the communication link; and
the first server analyzing the results of the attempted communication across the communication link to determine the servers cannot communicate with each other.
5. The method as recited in claim 1 , further comprising servicing, with the second server, any network requests made to the first server after the first server discontinues operation.
6. The method as recited in claim 1 , wherein analyzing the communication link to determine if the communication link is functioning properly comprises checking at least one transmission characteristic associated with the communication link to determine if the at least one transmission characteristic is within specified parameters.
7. The method as recited in claim 1 , wherein the servicing of a network request that would otherwise be serviced by the second server is facilitated by reconfiguration of the computer clustering system, the reconfiguration being performed by the first server.
8. The method as recited in claim 1 , further comprising determining, prior to the detection of the loss of communication, that the first and second server cannot mirror changes to the mass storage devices.
9. The method as recited in claim 1 , further comprising mirroring, with the first server, the mass storage device of the first server with a mass storage device of the second server, the mirroring being performed over the communication link.
10. The method as recited in claim 1 , further comprising mirroring data from the mass storage device of the second server to the mass storage device of the first server.
11. The method as recited in claim 1 , further comprising:
the first server determining that communication from the second server is restored; and
the second server discontinuing operation.
12. The method as recited in claim 1 , further comprising monitoring, with the first server, the communication link.
13. A computer program product for implementing a method for improving the availability and reliability of a computer clustering system having first and second servers that are connected to each other by a communication link and that each have an associated mass storage device, the computer program product comprising: a computer-readable medium having computer-executable instructions stored upon and executed by a computer processor for implementing the method, the method comprising:
the first server detecting, over the communication link, a loss of communication between the first and second servers;
the first server analyzing the communication link to determine if the communication link is functioning properly; and
if the communication link is determined to be functioning properly, the first server:
continuing operation;
assuming a right to survive without communicating with the second server or with the mass storage device of the first server or with the mass storage device of the second server;
servicing a network request that would otherwise be serviced by the second server; and
the first server discontinuing operation if analysis of the communication link indicates that the communication link is not functioning properly.
14. The computer program product as recited in claim 13 , wherein the right to survive is associated with the second server prior to any assumption of the right to survive by the first server.
15. The computer program product as recited in claim 13 , wherein detecting a loss of communication over the communication link comprises determining that a heartbeat signal associated with the second server cannot be detected on the communication link.
16. The computer program product as recited in claim 13 , wherein detecting a loss of communication over the communication link comprises:
the first server attempting to communicate with the second server across the communication link; and
the first server analyzing the results of the attempted communication across the communication link to determine the servers cannot communicate with each other.
17. The computer program product as recited in claim 13 , wherein the method further comprises servicing, with the second server, any network requests made to the first server after the first server discontinues operation.
18. The computer program product as recited in claim 13 , wherein analyzing the communication link to determine if the communication link is functioning properly comprises checking at least one transmission characteristic associated with the communication link to determine if the at least one transmission characteristic is within specified parameters.
19. The computer program product as recited in claim 13 , wherein the servicing of a network request that would otherwise be serviced by the second server is facilitated by reconfiguration of the computer clustering system, the reconfiguration being performed by the first server.
20. The computer program product as recited in claim 13 , wherein the method further comprises determining, prior to the detection of the loss of communication, that the first and second server cannot mirror changes to the mass storage devices.
21. The computer program product as recited in claim 13 , wherein the method further comprises mirroring, with the first server, the mass storage device of the first server with a mass storage device of the second server, the mirroring being performed over the communication link.
22. The computer program product as recited in claim 13 , wherein the method further comprises mirroring data from the mass storage device of the second server to the mass storage device of the first server.
23. The computer program product as recited in claim 13 , wherein the method further comprises:
the first server determining that communication from the second server is restored; and
the second server discontinuing operation.
24. The computer program product as recited in claim 13 , wherein the method further comprises monitoring, with the first server, the communication link.