System, method and program to identify failed components in storage area network

• US 20060080430A1
• Filed: 10/07/2004
• Published: 04/13/2006
• Est. Priority Date: 10/07/2004
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Barsuk, Vyacheslav
• Granted Patent: US 7,457,871 B2
• Time in Patent Office: Days
• US Class Current: 709/224
• CPC Class Codes: H04L 43/00 Arrangements for monitoring or testing data switching networks
• CPC Class Codes: H04L 43/0811 by checking connectivity
• CPC Class Codes: H04L 43/0817 by checking functioning
• CPC Class Codes: H04L 67/1097 for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS]
• CPC Class Codes: H04L 67/60 Scheduling or organising the servicing of application requests, e.g. requests for application data transmissions using the analysis and optimisation of the required network resources admission control or resource allocation H04L47/70

Abstract

System, method and computer program product for identifying a failed component in a system comprising application servers, storage servers and a switch fabric. The switch fabric has first ports coupled to the application servers, second ports coupled to third ports of the storage servers and internal switches to interconnect the first ports to second ports. Each of the application servers compiles records of its own attempts to communicate with the storage servers via the switch fabric. Each of the records indicates one of the third ports and one of the storage servers for each of the communication attempts. From the records a determination is made if any of the communications was successful to one of the storage servers. If not, a determination is made that the one storage server may have failed. If any of the communications was successful to the one storage server, a determination is made that the one storage server is active and a determination is made from the records if any of the communications was successful to each of the third ports of the one storage server. If not, a determination is made that each of the third ports of the one storage server may have failed. If any of the communications was successful to each of the third ports, a determination is made that each of the third ports of the one storage server is active. According to a feature of the present invention, a determination is also made based on the records if a switch or first port has failed.

Claims

1. A method for identifying a failed component in a system comprising application servers, storage servers and a switch fabric, said switch fabric having first ports coupled to said application servers to receive requests to access storage managed by said storage servers, second ports coupled to third ports of said storage servers to forward the access requests to said storage servers and receive responses from said storage servers, and internal switches to interconnect said first ports to second ports to pass said requests and responses through said switch fabric, said method comprising:

each of said application servers compiling records of its own attempts to communicate with said storage servers via said switch fabric, each of said records indicating one of said third ports and one of said storage servers for each of said communication attempts; and

determining from said records if any of said communications was successful to one of said storage servers, and if not, determining that said one storage server may have failed, or if so, determining that said one storage server is active and determining from said records if any of said communications was successful to each of said third ports of said one storage server, and if not, determining that said each third port of said one storage server may have failed, or if so, determining that said each third port of said one storage server is active.

2. A method as set forth in claim 1 wherein each of said records also indicates one of said first ports and one of said switches for said each communication attempt, and further comprising:

determining from said records if any of said communications was successful to each of said switches leading to said each third port of said one storage server, and if not, determining that said each switch may have failed, or if so, determining that said each switch is active and determining from said records if any of said communications was successful to each of said first ports connected to said each switch leading to said one storage server, and if not, determining that said each first port connected to said each switch leading to said one storage server or a connection between one of said application servers and said each first port connected to said each switch leading to said one storage server may have failed, or if so, determining that said each first port connected to said each switch leading to said one storage server is active.

3. A method as set forth in claim 1 wherein said switch fabric is hardwired to said application servers and said storage servers.

4. A method as set forth in claim 3 wherein said applications servers communicate with storage server via said fabric switch using fiber channel protocol.

5. A method as set forth in claim 2 wherein said switch fabric is hardwired to said application servers and said storage servers.

6. A method as set forth in claim 5 wherein said applications servers communicate with storage server via said fabric switch using fiber channel protocol.

7. Apparatus for identifying a failed component in a system comprising application servers, storage servers and a switch fabric, said switch fabric having first ports coupled to said application servers to receive requests to access storage managed by said storage servers, second ports coupled to third ports of said storage servers to forward the access requests to said storage servers and receive responses from said storage servers, and internal switches to interconnect said first ports to second ports to pass said requests and responses through said switch fabric, said apparatus comprising:

means for receiving from each of said application servers records of its own attempts to communicate with said storage servers via said switch fabric, each of said records indicating one of said third ports and one of said storage servers for each of said communication attempts; and

means for determining from said records if any of said communications was successful to one of said storage servers, and if not, determining that said one storage server may have failed, or if so, determining that said one storage server is active and determining from said records if any of said communications was successful to each of said third ports of said one storage server, and if not, determining that said each third port of said one storage server may have failed, or if so, determining that said each third port of said one storage server is active.

8. Apparatus as set forth in claim 7 wherein each of said records also indicates one of said first ports and one of said switches for said each communication attempt, and further comprising:

means for determining from said records if any of said communications was successful to each of said switches leading to said each third port of said one storage server, and if not, determining that said each switch may have failed, or if so, determining that said each switch is active and determining from said records if any of said communications was successful to each of said first ports connected to said each switch leading to said one storage server, and if not, determining that said each first port connected to said each switch leading to said one storage server or a connection between one of said application servers and said each first port connected to said each switch leading to said one storage server may have failed, or if so, determining that said each first port connected to said each switch leading to said one storage server is active.

9. Apparatus as set forth in claim 7 wherein said switch fabric is hardwired to said application servers and said storage servers.

10. Apparatus as set forth in claim 9 wherein said applications servers communicate with storage server via said fabric switch using fiber channel protocol.

11. Apparatus as set forth in claim 8 wherein said switch fabric is hardwired to said application servers and said storage servers.

12. Apparatus as set forth in claim 11 wherein said applications servers communicate with storage server via said fabric switch using fiber channel protocol.

13. A computer program product for identifying a failed component in a system comprising application servers, storage servers and a switch fabric, said switch fabric having first ports coupled to said application servers to receive requests to access storage managed by said storage servers, second ports coupled to third ports of said storage servers to forward the access requests to said storage servers and receive responses from said storage servers, and internal switches to interconnect said first ports to second ports to pass said requests and responses through said switch fabric, said computer program product comprising:

a computer readable medium;

first program instructions to receive from each of said application servers records of its own attempts to communicate with said storage servers via said switch fabric, each of said records indicating one of said third ports and one of said storage servers for each of said communication attempts; and

second program instructions to determine from said records if any of said communications was successful to one of said storage servers, and if not, determine that said one storage server may have failed, or if so, determine that said one storage server is active and determine from said records if any of said communications was successful to each of said third ports of said one storage server, and if not, determine that said each third port of said one storage server may have failed, or if so, determine that said each third port of said one storage server is active; and

wherein

said first and second program instructions are recorded on said medium.

14. A computer program product as set forth in claim 13 wherein each of said records also indicates one of said first ports and one of said switches for said each communication attempt, and further comprising:

third program instructions to determine from said records if any of said communications was successful to each of said switches leading to said each third port of said one storage server, and if not, determine that said each switch may have failed, or if so, determine that said each switch is active and determine from said records if any of said communications was successful to each of said first ports connected to said each switch leading to said one storage server, and if not, determine that said each first port connected to said each switch leading to said one storage server or a connection between one of said application servers and said each first port connected to said each switch leading to said one storage server may have failed, or if so, determine that said each first port connected to said each switch leading to said one storage server is active; and

wherein

said third program instructions are recorded on said medium.

15. A computer program product as set forth in claim 13 wherein said switch fabric is hardwired to said application servers and said storage servers.

16. A computer program product as set forth in claim 15 wherein said applications servers communicate with storage server via said fabric switch using fiber channel protocol.

17. A computer program product as set forth in claim 14 wherein said switch fabric is hardwired to said application servers and said storage servers.

18. A computer program product as set forth in claim 17 wherein said applications servers communicate with storage server via said fabric switch using fiber channel protocol.