Method for recovery of paths between storage area network nodes with probationary period and desperation repair

• US 6,952,734 B1
• Filed: 08/21/2000
• Issued: 10/04/2005
• Est. Priority Date: 08/21/2000
• Status: Expired due to Term

• Current Assignee: QNAP Systems Incorporated
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Burris, Dustin D., Gunlock, Richard D.
• Primary Examiner(s): Kang, Paul H.
• Application Number:
  US09/642,801
• Time in Patent Office: 1,870 Days
• Field of Search: 709/227, 709/224, 709/223, 709/226, 709/238, 709/239, 714/25, 714/43
• US Class Current: 709/227
• CPC Class Codes: G06F 11/0727 in a storage system, e.g. in a DASD or network based storage system drivers for digital recording or reproducing units G06F3/06; circuits for error detection or correction within digital recording or reproducing units G11B20/18; 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], H04L67/1097
• CPC Class Codes: G06F 11/076 by exceeding a count or rate limit, e.g. word- or bit count limit
• CPC Class Codes: G06F 11/201 between storage system components
• 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]

Abstract

A node for a storage area network has a processor, at least one port for connection to a storage area network, and a memory system. The memory system contains machine readable instructions for managing the at least one port, including instructions for determining a status associated with the at least one port capable of holding at least failed, probationary, normal and active status. For any failed port, the instructions include instructions for detecting a repair and when repair is detected for advancing the status to active or probationary status—probationary status being set when no exchange is pending to a target node reachable only through the repaired port. For probationary ports, the instructions detect when the port operates without error for a predetermined period and advances the status from probationary to active or normal. Network exchanges are preferentially routed over an active or normal port, rather than transmitting them over probationary ports.

Claims

1. A node for a storage area network comprising:

at least one processor;

at least one port for connection to a storage area network coupled to the at least one processor; and

a memory system coupled to the at least one processor;

wherein the memory system contains machine readable instructions comprising instructions for;

determining a status associated with the at least one port, the status capable of having at least an active, a normal, a probationary, and a failed value;

when the status has a failed value, of detecting a repair associated with the at least one port and when repair is detected advancing the status to a value selected from the group consisting of active and probationary status;

detecting when the port operates without error for a predetermined period and thereupon advancing the status from the probationary value to a value selected from the group consisting of active and normal; and

preferentially routing exchanges over a port of the at least one port having a status value selected from the group of active and normal when a port having such status exists and a target node of the exchange is reachable over that port.

2. The node of claim 1, wherein the port is a fibre channel N_—

port, L_—

port, or NL_—

port.

3. The node of claim 1, wherein the memory system contains machine readable instructions further comprising instructions for:

determining that a port having an associated status of probationary has encountered excessive errors, and when said excessive errors are detected for changing that status to failed.

4. The node of claim 3, wherein the machine readable instructions for determining that a port having an associated status of probationary has encountered excessive errors includes instructions of ignoring errors detected within a predetermined period of time of the port seeing a network login attempt made by a second node of the network.

5. The node of claim 4, wherein the memory system contains machine readable instructions further comprising instructions for taking laser diodes of a port of the at least one port offline when status associated with that port is changed to failed, and for placing those laser diodes online when said status is changed to a status selected from the group consisting of active, normal, and probationary.

6. The node of claim 1, wherein the memory system contains machine readable instructions further comprising instructions for conducting login attempts over each port of the at least one port, for determining target nodes reachable through each port, and for recording in the memory system identities of target nodes reachable through each port.

7. The node of claim 6, wherein the memory system contains machine readable instructions further comprising instructions for promoting status associated with a port from probationary to a value selected from the group consisting of normal and active when an exchange to a particular target node of the target nodes is pending and that target node is not reachable through any port already having an associated status selected from the group consisting of normal and active.

8. The node of claim 6, wherein the memory system contains machine readable instructions further comprising instructions for testing a port for repair if that port has failed status, an exchange to a particular target node of the target nodes is pending, and the particular target node is not reachable through any other port already having an associated status selected from the group consisting of probationary, normal and active.

9. A method for assigning exchanges to ports of a node of a storage area network, the node having at least one port for connecting the node to the storage area network, comprising the steps of:

determining a status associated with each port of the at least one port, the status capable of having at least an active, a normal, a probationary, and a failed value;

when the status has a failed value, of detecting a repair associated with the associated port and when repair is detected advancing the status to a value selected from the group consisting of active and probationary status;

detecting when each port having probationary status operates without error for a predetermined period and thereupon advancing the status from probationary to a status selected from the group consisting of active and normal; and

preferentially assigning exchanges to ports having a status value selected from the group of active and normal when at least one port having such status exists and a target node of the exchange is reachable over that port.

10. The method of claim 9 wherein the port is a fibre channel N_—

port, L_—

port, or NL_—

port.

11. The method of claim 9, further comprising the step of:

determining that a port having an associated status of probationary has encountered excessive errors, and when said excessive errors are detected for changing that status to failed.

12. The method of claim 11, wherein the step of determining that a port having an associated status of probationary has encountered excessive errors ignores at least some errors detected within a predetermined period of time of the port seeing a network login attempt made by a second node of the network.

13. The method of claim 11, further comprising the steps of:

taking laser diodes of a port offline when status associated with that port is changed to failed, and for placing those laser diodes online when said status is changed to a status selected from the group consisting of active, normal, and probationary.

14. The method of claim 13, further comprising the steps of:

conducting login attempts over each port having online laser diodes,

determining target nodes reachable through each port, and

recording in the memory system identities of target nodes reachable through each port.

15. The method of claim 14, further comprising the steps of:

promoting status associated with a port from probationary to a value selected from the group consisting of normal and active when an exchange to a particular target node of the target nodes is pending and that target node is not reachable through any port already having an associated status selected from the group consisting of normal and active.

16. The node of claim 15, further comprising the step of testing a port for repair if that port has failed status, an exchange to a particular target node of the target nodes is pending, the particular target node is possibly reachable through the port having failed status and the particular target node is not reachable through any other port having an associated status selected from the group consisting of probationary, normal and active.

17. A method comprising:

monitoring to determine a status of individual ports of a node of a storage area network;

the status comprising one of;

active, normal, probationary, or failed;

in an event of an individual port having a failed status, detecting a repair of the individual port, and upon detecting the repair, advancing the status of the individual port to one of;

active or probationary;

detecting when an individual port having a probationary status operates without error for a predetermined period and thereupon advancing the individual port from the probationary status to a status comprising;

active or normal; and

preferentially assigning exchanges to individual ports having a status of active or normal when at least one port having such status exists and a target node of the exchange is reachable over that port.