High availability file server for providing transparent access to all data before and after component failover

• US 6,854,072 B1
• Filed: 06/06/2001
• Issued: 02/08/2005
• Est. Priority Date: 10/17/2000
• Status: Expired due to Fees

• Current Assignee: Radisys Corporation (Reliance Industries Limited)
• Original Assignee: Continuous Computing Corp. (Reliance Industries Limited)
• Inventors: Sheehan, Kevin Sean, McEvoy, Patrick J., Cagle, Robert Penn
• Primary Examiner(s): Lim, Krisna
• Application Number:
  US09/874,249
• Time in Patent Office: 1,343 Days
• Field of Search: 714/1, 714/6, 714/11, 714/15, 709/203
• US Class Current: 714/15
• CPC Class Codes: H04L 49/351 for local area network [LAN], e.g. Ethernet switches
• CPC Class Codes: H04L 49/40 Constructional details, e.g. power supply, mechanical construction or backplane
• CPC Class Codes: H04L 49/552 by ensuring the integrity of packets received through redundant connections
• CPC Class Codes: H04L 49/65 Re-configuration of fast packet switches
• CPC Class Codes: H04L 61/103 across network layers, e.g. resolution of network layer into physical layer addresses or address resolution protocol [ARP]
• CPC Class Codes: H04L 61/5014 using dynamic host configuration protocol [DHCP] or bootstrap protocol [BOOTP]
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 69/12 Protocol engines
• CPC Class Codes: H04L 69/32 Architecture of open systems interconnection [OSI] 7-layer type protocol stacks, e.g. the interfaces between the data link level and the physical level
• CPC Class Codes: H04L 69/329 in the application layer [OSI layer 7]
• CPC Class Codes: H04L 69/40 for recovering from a failure of a protocol instance or entity, e.g. service redundancy protocols, protocol state redundancy or protocol service redirection management of faults, events, alarms or notifications in data switching networks H04L41/06

Abstract

A system and method for providing highly available and/or fault tolerant servers by implementing a lookup table or other data structure that allows a common file handle to be generated and interpreted by any component server of the HA server, thereby preventing stale file handles and removing local file system requirements as to media, implementation, and location.

Claims

1. An HA network file server comprising:

two or more component servers having associated file systems, the file systems having no requirements as to media, implementation, or location;

means for keeping data content of the file systems consistent among the component servers;

means for sharing access to the file systems over a network;

means for redirecting network traffic between the component servers; and

,

means for implementing file handle translation to allow one or more component servers to service transparently transactions of another component server.

2. An HA network file server comprising:

two or more component servers;

a local file system associated with each of the component servers, the local file system having no requirements as to media, implementation, or location; and

a translator for translating a file handle issued from one component server to the local file system of any component server to allow transactions associated with the issuing component server to be serviced transparently by any of the component servers.

3. The server of claim 2 wherein

each component server generates local file handles containing a component server ID and local file system information of the component server;

and,

the component servers coordinate to generate a common file handle to be translated by the translator to allow any component server to identify a corresponding file in its local file system, the common file handle issued for use in client/server transactions.

4. The server of claim 2 wherein

the translator maintains data structures that allow a component server to translate a file handle issued from another component server to its own file system information.

5. The server of claim 4 wherein the data structures associate a common file handle with local file handles generated by the local file systems of the component servers for a file, the common file handle used in client/server transactions.

6. The server of claim 5 wherein the common file handle is a generic file handle.

7. The server of claim 5 wherein the common file handle is the local file handle generated by the issuing component server.

8. The server of claim 2 wherein a file handle is translated by the translator in response to a failure of the issuing component server.

9. The server of claim 2 wherein a file handle is translated in response to increased workload to thereby effect a re-distribution of the workload among the component servers.

10. A method of providing high availability and fault tolerance to a server system having two or more component servers with associated file systems, the file systems having no requirements as to media, implementation, or location, the method comprising:

maintaining consistent data content among the file systems associated with the component servers;

sharing access to the file systems over a network;

redirecting network traffic between the component servers; and

,

implementing file handle translation to allow one or more component servers to service transparently the transactions of a first component server.

11. The method of claim 10 wherein implementing file handle translation comprises:

maintaining a data structure to allow any component server to associate a file handle issued by any other component server with the correct file in its own file system.

12. The method of claim 11 wherein implementing file handle translation further comprises:

using the data structure to translate a file handle issued by a first component server to a corresponding file in the file system of a second component server.

13. The method of claim 12 wherein using the data structure comprises:

associating a generic file handle with local file handles from the file systems of the component servers, the generic file handle issued for use in the associated transactions.

14. The method of claim 12 wherein using the data structure comprises:

associating the local file handles from the file systems of the component servers for a file, the local file handle of the first component server issued for use in associated client/server transactions.

15. The method of claim 12 wherein the first component server may be any component server.

16. The method of claim 10 wherein implementing file handle translation occurs in response to increased workload to allow the other component servers to service the associated transactions, thereby effecting transparent workload re-distribution among the component servers.

17. The method of claim 10 wherein implementing file handle translation occurs in response to the failure of the first component server, thereby effecting transparent file handle failover.

18. A method of providing high availability and fault tolerance to a network server system having two or more component servers with associated file systems, the file systems having no requirements as to media, implementation, or location, the method comprising:

issuing a file handle from a first component server,

translating the file handle to the file system(s) of one or more other component servers to allow the other component server(s) to service transparently the associated transactions.

19. The method of claim 18 further comprising:

maintaining a data structure for translating the file handle to the file systems of the component servers.

20. The method of claim 18 wherein issuing a file handle from a first component server comprises:

generating a local file handle for a file maintained in the file system of a first component server;

coordinating with remaining component servers to generate a common file handle that may be used by any of the component servers to identify the file in the associated file system of the component server;

generating the common file handle; and

,

returning the common file handle for use in associated transactions.

21. The method of claim 18 wherein translating the file handle occurs in response to the failure of the first component server.

22. The method of claim 20 wherein generating the common file handle comprises one of:

associating a generic file handle with local file handles from the file systems of the component servers, the generic file handle returned for use in the associated transactions;

or

associating the local file handles from the file systems of the component servers for a file, the local file handle of the first component server returned for use in the associated transactions.

23. The method of claim 22 wherein the first component server may be any of the component servers.

24. The method of claim 23 further comprising:

translating the file handle to the file system of one or more other component servers in response to increased workload to allow the other component servers to service the associated transactions, thereby effecting transparent workload re-distribution among the component servers.