Controlling virtual endpoint failover during administrative SCSI target port disable/enable

Patent metadata

Field	Value
Publication number	US-9747180-B1
Application number	US-201514675396-A
Country	US
Kind code	B1
Filing date	Mar 31, 2015
Priority date	Mar 31, 2015
Publication date	Aug 29, 2017
Grant date	Aug 29, 2017

Abstract

Embodiments are described for systems and methods that facilitate control of virtual endpoint failover/failback during an administrative SCSI target port disable or enable operation. In this case, SCSI target virtual endpoints may failover to a secondary SCSI target port when the primary port fails. When the primary port is corrected and enabled by the administrator the failover method pulls virtual endpoints on secondary ports back to the primary port under administrator control; and if an administrator wishes to manually disable a SCSI target port the failover operation pushes (failover) all virtual endpoints currently using the port as a primary to a secondary port.

First claim

What is claimed is:

1. A method of controlling failover operation among virtual small computer system interface (SCSI) transport ports in a SCSI network, comprising:
creating a virtual port on a first physical port of the SCSI network and assigning the first virtual port a fixed world-wide port name (WWPN) using a SCSI peripheral library (PLIB) comprising a simple access library providing a low-level interface to peripherals in the SCSI network;
receiving a command from a system administrator to disable the first physical port;
removing the virtual port from the first physical port and recreating the virtual port on a second physical port with the fixed WWPN to provide a failover port for data transactions intended for the first physical port; and
receiving a command from the system administrator to enable the second physical port
and failover to the second physical port so that data for the first physical port is transmitted to the second physical port;
wherein the first and second physical ports and the virtual port are represented as endpoints comprising abstracted objects having attributes that are portable through a network including the networked device, the attributes specifying addressing and operating characteristics of an associated endpoint, and wherein the PLIB is extended from an originally defined size to a size that enables it to contain both physical and virtual ports that do not share a same port index.

2. The method of claim 1 wherein the failover comprises a failback, the method further comprising receiving a command from the system administrator to enable the first physical port and failback to the first physical port so that data is transmitted to the first physical port.

3. The method of claim 1 wherein the SCSI network comprises a Fibre Channel system implementing the SCSI transport ports, and wherein the data store is maintained by a server computer executing a data domain deduplication operating system, and further wherein the SCSI network comprises a storage area network, and wherein the first and second physical ports each comprise one of a host bus adapter or a storage device port.

4. The method of claim 1 wherein the failure is one of an operation failure of the port or an administrative command to disable the port.

5. The method of claim 4 wherein the disable command is transmitted by the system administrator in response or in anticipation of a failure of the first physical port, the method further comprising defining failover policies that govern execution of the failover operation and include parameters selected from the group consisting of: failure trigger conditions, wait times for initiation of port transfer, address and sequence of backup physical ports, and manual/automatic initiation of the failover operation.

6. The method of claim 5 wherein the failover operation comprises one of: a failover from a first data storage device to a second data storage device; and failback from the second storage device to the first storage device.

7. A method of controlling virtual small computer system interface (SCSI) transport ports SCSI in a network during failover operations, comprising:
receiving a disable command from a system administrator to disable a SCSI target port in the network, the SCSI target port comprising a first physical port having one or more associated virtual ports; and
re-mapping, in response to the disable command, the one or more associated virtual ports from the first physical port to a second physical port while maintaining same world wide port numbers for the one or more associated virtual ports using a SCSI peripheral library (PLIB) comprising a simple access library providing a low-level interface to peripherals in the SCSI network;
wherein the one or more associated virtual ports represent endpoints that are abstracted objects having attributes that can be moved around the SCSI network utilizing the transport ports, and wherein the PLIB is extended from an originally defined size to a size that enables it to contain both physical and virtual ports that do not share a same port index.

8. The method of claim 7 further comprising:
receiving an enable command from the system administrator to enable the disabled SCSI target port; and
mapping, in response to the enable command, the one or more associated virtual ports from the second physical port back to the first physical port.

9. The method of claim 7 wherein the disable command triggers a failover operation to provide a failover port for data transactions intended for the first physical port.

10. The method of claim 9 wherein the failover operation comprises one of: a failover from a first data storage device to a second data storage device; and failback from the second storage device to the first storage device.

11. The method of claim 10 further comprising defining failover policies that govern execution of the failover operation and include parameters selected from the group consisting of: failure trigger conditions, wait times for initiation of port transfer, address and sequence of backup physical ports, and manual/automatic initiation of the failover operation.

12. The method of claim 7 further comprising managing the endpoints through a multi-threaded SCSI target daemon process that sends commands to an operating system of the SCSI network to create and maintain multiple virtual ports for the physical port based on at least some of the endpoints.

13. The method of claim 12 wherein the SCSI network comprises a Fibre Channel system implementing the SCSI transport ports, and wherein the attributes for physical ports include firmware version, physical presence, and link speed; and further wherein the attributes for virtual ports include host initiator identifiers; WWPN addresses, world wide node name (WWNN) addresses and fibre channel protocol (fcp2)-retry state.

14. The method of claim 13 wherein the operating system comprises a data domain deduplication operating system, and wherein the network comprises a storage area network, and
wherein the mapping comprises an N_Port ID (NPIV) mapping for the Fibre Channel system.

15. A computer program product comprising a non-transitory computer usable medium having machine readable code embodied therein for controlling failover operation among virtual small computer system interface (SCSI) transport ports in a SCSI network, by:
creating a virtual port on a first physical port of the SCSI network and assigning the first virtual port a fixed world-wide port name (WWPN) using a SCSI peripheral library (PLIB) comprising a simple access library providing a low-level interface to peripherals in the SCSI network;
receiving a command from a system administrator to disable the first physical port;
removing the virtual port from the first physical port and recreating the virtual port on a second physical port with the fixed WWPN to provide a failover port for data transactions intended for the first physical port; and
receiving a command from the system administrator to enable the second physical port and failover to the second physical port so that data for the first physical port is transmitted to the second physical port;
wherein the first and second physical ports and the virtual port are represented as endpoints comprising abstracted objects having attributes that are portable through a network including the networked device, the attributes specifying addressing and operating characteristics of an associated endpoint, and wherein the PLIB is extended from an originally defined size to a size that enables it to contain both physical and virtual ports that do no

Assignees

• Emc Ip Holding Co Llc

Inventors

• Fair Robert

Classifications

• G06F11/2033 Primary

  switching over of hardware resources · CPC title

• G06F2201/815

  Virtual · CPC title

• G06F11/2005

  using redundant communication controllers · CPC title

• G06F11/201 Primary

  between storage system components · CPC title