METHODS FOR IMPROVING MANAGEMENT OF INPUT OR OUTPUT OPERATIONS IN A NETWORK STORAGE ENVIRONMENT WITH A FAILURE AND DEVICES THEREOF

• US 20160239394A1
• Filed: 02/13/2015
• Published: 08/18/2016
• Est. Priority Date: 02/13/2015
• Status: Abandoned Application

• Current Assignee: NetApp, Inc.
• Original Assignee: NetApp, Inc.
• Inventors: Ramprasad Darisa, Venkata, Ravindranath Allu, Nandakumar, Nagarajan, Rajesh
• Application Number:
  US14/622,014
• Publication Number:
  US 20160239394A1
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 11/1423 by reconfiguration of paths
• CPC Class Codes: G06F 11/1425 by reconfiguration of node membership
• CPC Class Codes: G06F 11/2069 Management of state, configuration or failover
• CPC Class Codes: G06F 11/2094 Redundant storage or storage space G06F11/2056 takes precedence
• CPC Class Codes: G06F 2201/84 Using snapshots, i.e. a logical point-in-time copy of the data
• CPC Class Codes: G06F 2201/85 Active fault masking without idle spares active fault masking without idle spare hardware where processing functionality is redundant G06F11/2035

Abstract

This technology identifies one or more nodes with a failure, designates the identified one or more nodes as ineligible to service any I/O operation, and disables I/O ports of the identified one or more nodes. Another one or more nodes are selected to service any I/O operation of the identified one or more nodes based on a stored failover policy. Any of the I/O operations are directed to the selected another one or more nodes for servicing and then routing of any of the serviced I/O operations via a switch to the identified one or more nodes to execute any of the routed I/O operations with a storage device. An identification is made when the identified one or more nodes is repaired. The designation as ineligible is removed and one or more I/O ports of the identified one or more nodes are enabled when the repair is identified.

Claims

1. A method for improving management of input or output (I/O) operations in a network storage environment with a failure, the method comprising:

identifying, by at least one of a plurality of node controller computing devices, another one of the plurality of node controller computing devices with a failure;

designating, by the at least one of the plurality of node controller computing devices, as ineligible to service any I/O operation and disabling one or more I/O ports of the identified one of the plurality of node controller computing devices with the failure;

selecting, by the at least one of the plurality of node controller computing devices, another one of the plurality of node controller computing devices without a failure to service any I/O operation of the identified one of the plurality of node controller computing devices with the failure based on a stored failover policy;

directing, by the at least one of the plurality of node controller computing devices, any of the I/O operations to the selected another one of the plurality of node controller computing devices for servicing and then routing of any of the serviced I/O operations via a switch to the identified one of the plurality of node controller computing devices with the failure to execute any of the routed I/O operations with a storage device;

identifying, by the at least one of the plurality of node controller computing devices, when the identified one of the plurality of node controller computing devices with the failure is repaired; and

removing, by the at least one of the plurality of node controller computing devices, the designation as ineligible and enabling one or more I/O ports of the identified one of the plurality of node controller computing devices identified with the repair.

2. The method as set forth in claim 1 wherein the identified one of the plurality of node controller computing devices with the failure further comprises two of the plurality of node controller computing devices in a pair with the failure;

and

wherein the selecting another one of the plurality of node controller computing devices without a failure further comprises;

selecting, by the at least one of the plurality of node controller computing devices, another pair of the plurality of node controller computing devices without a failure to service any I/O operation of the identified pair of the plurality of node controller computing devices with the failure based on the stored failover policy.

3. The method as set forth in claim 2 further comprising:

identifying, by the at least one of the plurality of node controller computing devices, when a repair of one of the two of the plurality of node controller computing devices in the pair with the failure is initiated;

wherein the directing any of the I/O operations to the selected another one of the plurality of node controller computing devices without a failure for servicing and then routing of any of the serviced I/O operations further comprises;

halting, by the at least one of the plurality of node controller computing devices, the servicing of any of the routed I/O operations with the one of the two of the plurality of node controller computing devices in a pair with the failure with the identified initation of the repair; and

allowing, by the at least one of the plurality of node controller computing devices, the other one of the two of the plurality of node controller computing devices in a pair with the failure which does not have the identified initation of the repair to take over the servicing of any of the routed I/O operations.

4. The method as set forth in claim 1 wherein the identified one of the plurality of node controller computing devices with the failure further comprises an independent node controller computing device in the plurality of node controller computing devices with the failure;

and

wherein the selecting another one of the plurality of node controller computing devices without a failure further comprises;

selecting, by the at least one of the plurality of node controller computing devices, another independent one of the plurality of node controller computing devices without a failure to service any I/O operation of the identified independent one of the plurality of node controller computing devices with the failure based on the stored failover policy.

5. The method as set forth in claim 4 further comprising:

identifying, by the at least one of the plurality of node controller computing devices, when a repair of the identified independent one of the plurality of node controller computing devices with the failure is initiated;

wherein the directing any of the I/O operations to the selected another one of the plurality of node controller computing devices for servicing and then routing of any of the serviced I/O operations further comprises;

halting, by the at least one of the plurality of node controller computing devices, the servicing of any of the routed I/O operations with the identified independent one of the plurality of node controller computing devices with the failure and with the identified initation of the repair; and

allowing, by the at least one of the plurality of node controller computing devices, buffering of any of the routed I/O operations in the another independent one of the plurality of node controller computing devices for a stored buffer time.

6. The method as set forth in claim 1 wherein the failure comprises a failure of a NVRAM battery failure in one or more of the plurality of node controller computing devices.

7. A non-transitory computer readable medium having stored thereon instructions for improving management of input or output (I/O) operations in a network storage environment with a failure comprising executable code which when executed by a processor, causes the processor to perform steps comprising:

identifying one of the one or more of the plurality of node controller computing devices with a failure;

designating as ineligible to service any I/O operation and disabling one or more I/O ports of the identified one of the plurality of node controller computing devices with the failure;

selecting another one of the plurality of node controller computing devices without a failure to service any I/O operation of the identified one of the plurality of node controller computing devices with the failure based on a stored failover policy;

directing any of the I/O operations to the selected another one of the plurality of node controller computing devices for servicing and then routing of any of the serviced I/O operations via a switch to the identified one of the plurality of node controller computing devices with the failure to execute any of the routed I/O operations with a storage device;

identifying when the identified one of the plurality of node controller computing devices with the failure is repaired; and

removing the designation as ineligible and enabling one or more I/O ports of the identified one of the plurality of node controller computing devices identified with the repair.

8. The medium as set forth in claim 7 wherein the identified one of the plurality of node controller computing devices with the failure further comprises two of the plurality of node controller computing devices in a pair with the failure;

and

wherein the selecting another one of the plurality of node controller computing devices without a failure further comprises;

selecting another pair of the plurality of node controller computing devices without a failure to service any I/O operation of the identified pair of the plurality of node controller computing devices with the failure based on the stored failover policy.

9. The medium as set forth in claim 8 further comprising:

identifying when a repair of one of the two of the plurality of node controller computing devices in the pair with the failure is initiated;

wherein the directing any of the I/O operations to the selected another one of the plurality of node controller computing devices without a failure for servicing and then routing of any of the serviced I/O operations further comprises;

halting the servicing of any of the routed I/O operations with the one of the two of the plurality of node controller computing devices in a pair with the failure with the identified initation of the repair; and

allowing the other one of the two of the plurality of node controller computing devices in a pair with the failure which does not have the identified initation of the repair to take over the servicing of any of the routed I/O operations.

10. The medium as set forth in claim 7 wherein the identified one of the plurality of node controller computing devices with the failure further comprises an independent node controller computing device in the plurality of node controller computing devices with the failure;

and

wherein the selecting another one of the plurality of node controller computing devices without a failure further comprises;

selecting another independent one of the plurality of node controller computing devices without a failure to service any I/O operation of the identified independent one of the plurality of node controller computing devices with the failure based on the stored failover policy.

11. The medium as set forth in claim 10 further comprising:

identifying when a repair of the identified independent one of the plurality of node controller computing devices with the failure is initiated;

wherein the directing any of the I/O operations to the selected another one of the plurality of node controller computing devices for servicing and then routing of any of the serviced I/O operations further comprises;

halting the servicing of any of the routed I/O operations with the identified independent one of the plurality of node controller computing devices with the failure and with the identified initation of the repair; and

allowing buffering of any of the routed I/O operations in the another independent one of the plurality of node controller computing devices for a stored buffer time.

12. The medium as set forth in claim 7 wherein the failure comprises a failure of a NVRAM battery failure in one or more of the plurality of node controller computing devices.

13. A network storage management system comprising:

a plurality of node controller computing devices, wherein one or more of the plurality of node controller computing devices comprise a memory coupled to a processor which is configured to be capable of executing programmed instructions comprising and stored in the memory to;

identify one of the one or more of the plurality of node controller computing devices with a failure;

designate as ineligible to service any I/O operation and disabling one or more I/O ports of the identified one of the plurality of node controller computing devices with the failure;

select another one of the plurality of node controller computing devices without a failure to service any I/O operation of the identified one of the plurality of node controller computing devices with the failure based on a stored failover policy;

direct any of the I/O operations to the selected another one of the plurality of node controller computing devices for servicing and then routing of any of the serviced I/O operations via a switch to the identified one of the plurality of node controller computing devices with the failure to execute any of the routed I/O operations with a storage device;

identify when the identified one of the plurality of node controller computing devices with the failure is repaired; and

remove the designation as ineligible and enabling one or more I/O ports of the identified one of the plurality of node controller computing devices identified with the repair.

14. The system as set forth in claim 13 wherein the identified one of the plurality of node controller computing devices with the failure further comprises two of the plurality of node controller computing devices in a pair with the failure;

and

wherein the processor coupled to the memory is further configured to be capable of executing at least one additional programmed instruction for the select another one of the plurality of node controller computing devices without a failure further comprises and is stored in the memory to;

select another pair of the plurality of node controller computing devices without a failure to service any I/O operation of the identified pair of the plurality of node controller computing devices with the failure based on the stored failover policy.

15. The system as set forth in claim 14 wherein the processor coupled to the memory is further configured to be capable of executing at least one additional programmed instruction further comprising and stored in the memory to:

identify when a repair of one of the two of the plurality of node controller computing devices in the pair with the failure is initiated;

wherein the processor coupled to the memory is further configured to be capable of executing at least one additional programmed instruction for the direct any of the I/O operations to the selected another one of the plurality of node controller computing devices without a failure for servicing and then routing of any of the serviced I/O operations further comprising and stored in the memory to;

halt the servicing of any of the routed I/O operations with the one of the two of the plurality of node controller computing devices in a pair with the failure with the identified initation of the repair; and

allow the other one of the two of the plurality of node controller computing devices in a pair with the failure which does not have the identified initation of the repair to take over the servicing of any of the routed I/O operations.

16. The system as set forth in claim 13 wherein the identified one of the plurality of node controller computing devices with the failure further comprises an independent node controller computing device in the plurality of node controller computing devices with the failure;

and

wherein the processor coupled to the memory is further configured to be capable of executing at least one additional programmed instruction for the select another one of the plurality of node controller computing devices without a failure further comprising and stored in the memory to;

select another independent one of the plurality of node controller computing devices without a failure to service any I/O operation of the identified independent one of the plurality of node controller computing devices with the failure based on the stored failover policy.

17. The system as set forth in claim 16 wherein the processor coupled to the memory is further configured to be capable of executing at least one additional programmed instruction further comprising and stored in the memory to:

identify when a repair of the identified independent one of the plurality of node controller computing devices with the failure is initiated;

wherein the processor coupled to the memory is further configured to be capable of executing at least one additional programmed instruction for the direct any of the I/O operations to the selected another one of the plurality of node controller computing devices without a failure for servicing and then routing of any of the serviced I/O operations further comprising and stored in the memory to;

halt the servicing of any of the routed I/O operations with the identified independent one of the plurality of node controller computing devices with the failure and with the identified initation of the repair; and

allow buffering of any of the routed I/O operations in the another independent one of the plurality of node controller computing devices for a stored buffer time.

18. The system as set forth in claim 13 wherein the failure comprises a failure of a NVRAM battery failure in one or more of the plurality of node controller computing devices.