초록 ▼

A system and method for clustered failover of storage appliances is provided. In the event of a failover of one storage appliance to a surviving storage appliance, the surviving storage appliance assumes the identity of the failed storage appliance. This assumption of the identifier is accomplished

A system and method for clustered failover of storage appliances is provided. In the event of a failover of one storage appliance to a surviving storage appliance, the surviving storage appliance assumes the identity of the failed storage appliance. This assumption of the identifier is accomplished by initializing a Fibre Channel Host Bus Adapter on the surviving storage appliance with the NodeName and PortNames of the failed storage appliance.

대표청구항 ▼

What is claimed is: 1. A method for transport-level failover of a first device to a second device in a Fibre Channel (FC) network, comprising: configuring the second device with a data structure having an entry containing a port name and a node name of the first device, wherein the second device in

What is claimed is: 1. A method for transport-level failover of a first device to a second device in a Fibre Channel (FC) network, comprising: configuring the second device with a data structure having an entry containing a port name and a node name of the first device, wherein the second device includes at least a first port and a second port, and where the first port on the second device receives requests directed to the second device; detecting, by the second device, a failure in the first device; initializing, on the second device, the second port connected to the FC network; configuring the second port with the port name and the node name of the first device, using the entry in the data structure listing the port name and the node name of the first device; asserting ownership, by the second device, of a set of disks owned by the first device in response to configuring the port name and the node name of the first device; and servicing, by the second device, data access requests directed to both the first and second device. 2. The method of claim 1 wherein the step of detecting a failure further comprises detecting, by the second device, a lack of a heartbeat signal from the first device. 3. The method of claim 1 wherein the first and second devices are FCP devices. 4. The method of claim 1 wherein the step of detecting a failure further comprises: initiating by a user a failover command. 5. The method of claim 1 wherein the step of initializing, on the second device, a second port connected to the Fibre Channel network further comprises: activating a second host bus adapter on the second device. 6. The method of claim 1 wherein the step of initializing, on the second device, a second port connected to the Fibre Channel network further comprises: activating an additional port on a host bus adapter on the second device. 7. The method of claim 1 wherein the step of configuring the second port with the port name of the first device the first device further comprises: setting a node name of the second port to the node name of the first port of the first device. 8. The method of claim 7 wherein the second port's node name is a world wide node name. 9. The method of claim 1 wherein the second port's port name is a world wide port name. 10. The method of claim 1 wherein the first device is a storage appliance and the second device is a storage appliance. 11. The method of claim 1, further comprises: initiating a giveback operation, by a user; ceasing to service requests for the first device by the second device; and configuring the port name of the second device to a standby name. 12. A computer readable medium, including program instructions executing on a second FCP device in a Fibre Channel network, the Fibre Channel network having a first FCP device, the computer readable medium including instructions for performing the steps of: configuring the second FCP device with a data structure having an entry containing a port name and a node name of the first FCP device, wherein the second FCP device includes at least a first port and a second port, and where the first port on the second FCP device receives requests directed to the second FCP device; detecting, by the second FCP device, a failure in the first FCP device; initializing, on the second FCP device, the second port connected to the Fibre Channel network; configuring the second port with the port name and the node name of the first FCP device using the entry in the data structure listing the port name and the node name of the first FCP device; asserting ownership, by the second FCP device, of a set of disks owned by the first FCP device in response to configuring the port name and the node name of the first FCP device; and servicing, by the second device, data access requests directed to both the first and second device. 13. The method of claim 12, further comprises: initiating a giveback operation, by a user; ceasing to service requests for the first FCP device by the second FCP device; and configuring the port name of the second FCP device to a standby name. 14. A computer cluster operating over a network, the computer cluster comprising: a first computer operatively interconnected with the network, the first computer having a first port for communicating with devices connected to the network and the first computer controlling access to a first set of disks; a second computer operatively interconnected with the network, the second computer having a first port on the second computer for communicating with devices connected to the network and the first port on the second computer adapted to receive requests for a second set of disks, and a second port for communicating with devices connected to the network, the second port adapted to assume a network identification of the first port when a failover event occurs, wherein after a failure occurs, the second computer is configured to process data access requests directed to both the first and second computer; and an operating system on the second computer configured to store in a data structure a port name and a node name of the first computer for the second computer to use as the network identification of the first port. 15. The computer cluster of claim 14 wherein the network is a Fibre Channel network. 16. The computer cluster of claim 14 wherein the failover event is an user activation of a failover command. 17. The computer cluster of claim 14 wherein the failover event is an error condition of the first computer. 18. A computer for use in a computer cluster in a Fibre Channel network, the computer comprising: means for configuring the computer cluster with a data structure having an entry containing a port name and a node name of one computer; at least one port for communicating over the Fibre Channel network; means for adapting at least one of the ports to assume a network identity of a port of a second computer in the Fibre Channel network in response to the entry of the port name and the node name; means for acquiring control of a set of storage devices associated with the second computer; and means for servicing data access requests directed to the assumed network identity of the second computer and the computer by the computer. 19. The computer of claim 18, further comprises: means for initiating a giveback operation, by a user; means for ceasing to service requests for the first device by the second device; and means for configuring the port name of the second device to a standby name. 20. A method for transport-level takeover of a first device by a second device in a Fibre Channel (FC) network, comprising: configuring the second device with at least two ports, where the first port on the second device is configured to receive requests directed at the second device; configuring the second device with a data structure having an entry containing a port name and a node name of the first device; activating, by a user, a takeover command; initializing, on the second device, a second port connected to the FC network; configuring the second port with the port name of the first device in response to the entry of the port name of the first device; asserting ownership, by the second device, of a set of disks owned by the first device; and servicing, by the second device, data access requests directed to both the first and second device. 21. A method for operating a first device and a second device in a Fibre Channel (FC) network, comprising: configuring the second device with a data structure having an entry containing a port name and a node name of the first device, wherein the second device includes at least a first port and a second port, and where the first port on the second device receives requests directed to the second device; detecting, by the second device, a failure in the first device; initializing, on the second device, the second port connected to the FC network; configuring the second port with the port name and the node name of the first device in response to the entry of the port name and the node name of the first device; asserting ownership, by the second device, of a set of disks owned by the first device in response to configuring the port name and the node name of the first device; servicing, by the second device, data access requests directed to both the first and second device repairing the first device; and using a give back command to transfer ownership back to the first device. 22. The method of claim 21 wherein the step of detecting a failure, further comprises: detecting, by the second device, a lack of a heartbeat signal from the first device. 23. The method of claim 21 wherein the step of detecting a failure, further comprises: initiating by a user a failover command. 24. The method of claim 21, wherein the step of configuring the second port with an the port name of the first device, further comprises: setting the second port's node name to the node name of the first port of the first device. 25. An apparatus for operating a first device and a second device in a Fibre Channel (FC) network, comprising: means for configuring the second device with a data structure having an entry containing a port name and a node name of the first device, wherein the second device includes at least a first port and a second port, and where the first port on the second device receives requests directed to the second device; means for detecting, by the second device, a failure in the first device; means for initializing, on the second device, the second port connected to the FC network; means for configuring the second port with the port name and the node name of the first device in response to the entry of the port name and the node name of the first device; means for asserting ownership, by the second device, of a set of disks owned by the first device in response to configuring the port name and the node name of the first device; means for servicing, by the second device, data access requests directed to both the first and second device means for repairing the first device; and means for using a give back command to transfer ownership back to the first device. 26. The apparatus of claim 25 wherein the means for detecting a failure, further comprises: means for detecting, by the second device, a lack of a heartbeat signal from the first device. 27. The apparatus of claim 25 wherein the means for detecting a failure, further comprises: means for initiating by a user a failover command. 28. The method of claim 25, wherein the means for configuring the second port with an the port name of the first device, further comprises: means for setting the second port's node name to the node name of the first port of the first device. 29. A computer cluster operating over a network, the computer cluster comprising: a first computer operatively interconnected with the network, the first computer having a first port for communicating with devices connected to the network and the first computer controlling access to a first set of disks; and a second computer operatively interconnected with the network, the second computer adapted with an operating system for storing a data structure containing a port name and a node name of the first port connected to the first computer, and having a second port for communicating with devices connected to the network, the second port adapted to assume the port name and the node name of the first port when a failover event occurs to receive requests directed to both the first computer and the second computer. 30. A method for operating a first device and a second device in a Fibre Channel (FC) network, comprising: configuring the first device with a first port to receive one or more requests, where the requests are directed to a first set of disks owned by the first device; configuring the second device with a first port to receive one or more requests, where the requests are directed to a second set of disks owned by the second device; configuring the second device with a data structure having an entry containing a port name and a node name of the first device; determining a failure occurred to the first device; configuring a second port on the second device with the port name and node name of the first device; asserting ownership, by the second device, of the first set of disks owned by the first device in response to configuring the port name and the rode name of the first device to allow the second device to receive and process requests directed to first and second device and; servicing, by the second device, data access requests directed to both the first and second device. 31. The method of claim 30, further comprising: repairing the first device; and using a give back command to transfer ownership back to the first device. 32. The method of claim 30, further comprising: configuring the first device with a data structure having an entry containing a port name and a node name of the second device. 33. A computer cluster operating over a network, the computer cluster comprising: a first computer operatively interconnected with the network, the first computer having a first port for communicating with devices connected to the network and the first computer controlling access to a first set of disks; and a second computer operatively interconnected with the network, the second computer having a first port for communicating with devices connected to the network and the second computer controlling access to a second set of disks, the second computer further configured with an operating system for storing a data structure containing a port name and a node name of the first port connected to the first computer, where upon failure of the first computer the operating system on the second computer is configured to initialize a second port on the second computer with the port name and node name of the first computer to allow the second computer to receive requests directed to both the first computer and the second computer. 34. The computer cluster of claim 33, wherein the first computer further comprises a data structure containing a port name and a node name of the second computer and an operating system to configure a second port on the first computer with the port name and the node name of the second computer if a failure occurs to the second computer.