Methods and systems for updating the firmware on a plurality of network-attached computing devices

• US 7,421,688 B1
• Filed: 02/03/2004
• Issued: 09/02/2008
• Est. Priority Date: 02/26/2003
• Status: Active Grant

• Current Assignee: American Megatrends International, LLC
• Original Assignee: American Megatrends Incorporated
• Inventors: Richardson, Brian, Righi, Stefano, Petree, Jerry
• Primary Examiner(s): Kiss; Eric B
• Application Number:
  US10/770,951
• Time in Patent Office: 1,673 Days
• Field of Search: 717/168, 717/171, 717/172
• US Class Current: 717/172
• CPC Class Codes: G06F 8/65 Updates security arrangements therefor G06F21/57
• CPC Class Codes: H04L 67/34 involving the movement of software or configuration parameters network booting or remote initial program loading [RIPL] G06F9/4416

Abstract

Methods and systems are provided for updating firmware on a plurality of networked computing devices via a distributed network. An instruction to update the firmware on a plurality of networked computing devices, such as nodes in a computer cluster, is sent from a manager computer to the computing devices monitoring for instructions. When the instruction is received, the computing devices transition to an operating system independent recovery state, receive a new firmware image over the distributed network, and update the firmware with the new firmware image.

Claims

1. A method for updating firmware on a plurality of computing devices over a distributed network comprising:

receiving from a manager computer over the distributed network at the plurality of computing devices, an instruction to execute a firmware maintenance procedure, the instruction received while executing an operating system;

in response to receiving the instruction, rebooting the network attached computers to an operating system independent operating environment;

receiving at the plurality of computing devices a new firmware image over the distributed network in the operating system independent operating environment; and

in response to receiving the new firmware image, updating a current firmware on each of the plurality of computing devices with the new firmware image in the operating system independent operating environment.

2. The method of claim 1, wherein the plurality of computing devices comprise network attached computers and wherein the instruction to execute the firmware maintenance procedure is received from the manager computer by an agent application executing on each network attached computer in an operating system dependent state wherein the instruction comprises a single request sent in parallel to the network attached computers and wherein the method further comprises:

in response to rebooting the network attached computers to the operating system independent operating environment, sending a notification of readiness to update from each network attached computer that rebooted to the operating system independent operating environment over the distributed network to the manager computer.

3. The method of claim 2, wherein updating the current firmware with the new firmware image comprises erasing the current firmware and copying the new firmware image to a memory location of each network attached computer.

4. The method of claim 2, wherein receiving the new firmware image comprises receiving the new firmware image in fragments over the distributed network, and wherein the method further comprises:

receiving a broadcast status request prior to updating the current firmware;

in response to receiving the broadcast status request, determining whether a rebroadcast of any fragment of the new firmware image is necessary;

in response to determining that the rebroadcast of one or more fragments is necessary, sending a request for the rebroadcast of the fragments over the distributed network; and

receiving the rebroadcast of fragments in response to sending the request.

5. The method of claim 4, wherein determining whether the rebroadcast of any fragments of the new firmware image is necessary comprises determining whether any fragments are missing or corrupted.

6. The method of claim 4, wherein the distributed network supports a user datagram protocol/Internet protocol and wherein a current firmware of each network attached computer is updated in parallel and wherein the request for rebroadcast includes numbers for the fragments requested or excludes numbers for the fragments requested.

7. The method of claim 2, further comprising monitoring a communication port of each network attached computer for the instruction to begin the maintenance procedure.

8. The method of claim 7, wherein each communication port of the network attached computers is monitored by the agent application that monitors instruction activity upon only one communication port and utilizes additional processor resources on the network attached computer only upon receiving the instruction.

9. The method of claim 2, further comprising:

in response to updating the current firmware with the new firmware image, sending a notification of the update to the manager computer.

10. The method of claim 2, further comprising:

determining whether a current firmware on each network attached computer is valid after being updated; and

when it is determined that a current firmware is valid then rebooting each network attached computer having a current firmware that is valid.

11. The method of claim 2, wherein the firmware within each network attached computer comprises a BIOS of the network attached computer.

12. The method of claim 2, wherein the network attached computers comprise servers within a server farm.

13. The method of claim 2, wherein the network attached computers each comprise a node in a computer cluster wherein the computer cluster operates as a single system and wherein the manager computer comprises a cluster master control computer.

14. A computer-controlled apparatus capable of performing the method of claim 1.

15. A computer storage medium comprising computer executable instructions which, when executed by a computer, cause the computer to perform the method of claim 1.

16. A system for updating firmware of a plurality of network attached computers over a network, the system comprising:

a first computer operative to;

send an instruction to update the firmware of the network attached computers over the network; and

send a new firmware image over the network to the network attached computers; and

at least a second computer and a third computer comprising the network attached computers wherein each network attached computer is operative to;

while in an operating system dependent state monitor a communications port for the instruction to update the firmware;

in response to receiving the instruction, transition from the operating system dependent state to an operating system independent recovery state;

in response to transitioning to the recovery state, receive the new firmware image while in the operating system independent recovery state; and

in response to receiving the new firmware image, update a current firmware with the new firmware image while in the operating system independent recovery state.

17. The system of claim 16, wherein the first computer is further operative to receive notifications of readiness for update from the network attached computers wherein sending the new firmware image is in response to receiving the notifications of readiness and wherein each network attached computer is further operative to:

send at least one of the notifications of readiness to the first computer over the network wherein sending the at least one of the notifications of readiness is in response to transitioning to the recovery state.

18. The system of claim 16, wherein the first computer is further operative to broadcast a single new firmware image to update at least the second computer and the third computer in parallel over the network.

19. The system of claim 16, wherein the network supports a user datagram protocol/Internet protocol, wherein the new firmware image is received in fragments over the distributed network, and wherein the second computer and the third computer are further operative to:

prior to updating the current firmware and in response to receiving the new firmware image, receive a broadcast status request from the first computer; and

in response to receiving the broadcast status request, identify any fragments of the new firmware image that are corrupted or missing;

in response to identifying one or more of the fragments that are corrupted or missing, send a request for the rebroadcast of the fragments identified over the network to the first computer; and

in response to sending the request for rebroadcast of the fragments identified, receive the rebroadcast of the fragments.

20. The system of claim 16, wherein the second computer and the third computer are further operative to in response to updating a current firmware with the new firmware image, send a notification of the update over the network to the first computer and wherein the system further comprises a display device wherein the first computer is further operative to display update status messages via the display device.

21. The system of claim 16, wherein the firmware comprises a BIOS of the second computer and a BIOS of the third computer and wherein update of the firmware is asynchronously executed in an OS independent state of each network attached computer.

22. The system of claim 16, wherein the second computer and the third computer comprise nodes in a computer cluster operating as a single computing system and wherein the manager computer comprises a master control cluster computer.

23. A computer implemented method for maintaining a BIOS of a plurality of networked computers via a distributed network wherein the networked computers comprise nodes of a computer cluster operating as a single computing system, the method comprising:

receiving via the distributed network at each node, an instruction to update a BIOS on each node wherein the instruction comprises a single request broadcast to the nodes;

in response to receiving the instruction, transitioning the nodes to a recovery state wherein the recovery state comprises an OS independent operating environment;

receiving at the nodes a new BIOS image broadcast over the distributed network as a one image broadcast to all the nodes wherein the distributed network supports a user datagram protocol/Internet protocol; and

in response to receiving the new BIOS image, updating a current BIOS on each of the nodes with the new BIOS image.

24. The computer implemented method of claim 23, the method further comprising:

monitoring a communication port of each network attached computer for the instruction to begin updating the BIOS.

25. The method of claim 24, wherein each communication port of the network attached computers is monitored by an agent application that monitors instruction activity upon only one communication port and utilizes additional processor resources on the network attached computer only upon receiving the instruction.