SYSTEM AND METHOD TO DETECT AND RECOVER FROM INOPERABLE DEVICE MANAGEMENT BUS

• US 20200133805A1
• Filed: 10/26/2018
• Published: 04/30/2020
• Est. Priority Date: 10/26/2018
• Status: Active Grant

• Current Assignee: Dell Products LP (Dell Technologies Inc.)
• Original Assignee: Dell Products LP (Dell Technologies Inc.)
• Inventors: Bisa, Rama Rao, Gupta, Chitrak, Gurram, Mohan Murali, Dhote, Komal, Vijaykumar, Suresh, Sahoo, Prasanna Kumar
• Granted Patent: US 10,853,204 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 11/0706 the processing taking place on a specific hardware platform or in a specific software environment
• CPC Class Codes: G06F 11/0751 Error or fault detection not based on redundancy power supply failures G06F1/30; network fault management H04L41/06
• CPC Class Codes: G06F 11/0793 Remedial or corrective actions recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40
• CPC Class Codes: G06F 11/2012 and using different communication protocols
• CPC Class Codes: G06F 11/3027 where the computing system component is a bus
• CPC Class Codes: G06F 11/349 for interfaces, buses
• CPC Class Codes: G06F 13/1668 Details of memory controller
• CPC Class Codes: G06F 13/4282 on a serial bus, e.g. I2C bus, SPI bus on daisy chain buses G06F13/4247
• CPC Class Codes: G06F 2201/805 Real-time
• CPC Class Codes: G06F 2201/81 Threshold
• CPC Class Codes: G06F 2201/88 Monitoring involving counting
• CPC Class Codes: G06F 2213/0016 Inter-integrated circuit (I2C)

Abstract

Embodiments are described for detecting and recovering from an inoperable device management bus. A remote management controller is configured to offload device management bus transactions that use a messaging protocol. The messing protocol transactions are offloaded to a bus protocol controller that is responsible for managing bus operations using the messaging protocol. The bus protocol controller updates a set of bus status counters stored in a shared memory based on the processing of the offloaded messaging protocol transactions. The remote management controller processes device management bus transactions that use a bus protocol and updates the bus status counters based on the status of the bus protocol transactions. The remote management controller determines the status of the device management bus based on the bus status counters in shared memory, if the device management bus is inoperable, resets the remote management controller.

Claims

1. A method for detecting and recovering from an inoperable device management bus by a remote management controller, the method comprising:

offloading bus transactions using a first protocol to a bus controller managing device management bus operations using the first protocol;

updating, by the bus controller, bus status counters based on a monitored status of the offloaded first protocol bus transactions;

processing device management bus transactions using a second protocol;

updating the bus status counters based on the status of the second protocol bus transactions;

determining the status of the bus based on the bus status counters in shared memory; and

if the bus is inoperable, resetting the remote management controller.

2. The method of claim 1, wherein the device management bus comprises an I2C bus.

3. The method of claim 2, wherein the bus protocol controller comprises an I2C co-processor of the remote management controller.

4. The method of claim 2, wherein the first protocol comprises MCTP (Management Component Transport Protocol) messaging.

5. The method of claim 4, wherein the second protocol comprises IPMB (Intelligent Platform Management Bus) bus transactions.

6. The method of claim 1, wherein the status of the device management bus is determined based on the bus status counters indicating consecutive errors in the first protocol bus transactions or the second protocol bus transaction.

7. The method of claim 6, wherein the status of the device management bus is determined based on the bus status counters indicating stalled progress in the first protocol bus transactions or the second protocol bus transaction.

8. A system for managing a device management bus coupled to a plurality of managed devices, the system comprising:

a shared memory storing bus status counters;

a remote access controller coupled to the plurality of managed devices via the device management bus, the remote access controller configured to;

offload bus transactions that utilize a first protocol to a bus protocol controller;

process device management bus transactions that utilize a second protocol;

update the bus status counters based on a monitored status of the second protocol transactions; and

determine the status of the device management bus based on the bus status counters in shared memory; and

the bus protocol controller configured for managing bus operations using the second protocol, the bus protocol controller further configured to;

process the offloaded transactions using the second protocol;

update bus status counters based on the status of the offloaded second protocol transactions.

9. The system of claim 8, wherein the bus protocol controller comprises an I2C co-processor of the remote management controller.

10. The system of claim 9, wherein the first protocol comprises MCTP (Management Component Transport Protocol) messaging.

11. The system of claim 9, wherein the second protocol comprises IPMB (Intelligent Platform Management Bus) bus transactions.

12. The system of claim 11, wherein the messaging protocol comprises MCTP (Management Component Transport Protocol) messaging.

13. The system of claim 8, wherein the status of the device management bus is determined based on the bus status counters indicating consecutive errors in the first protocol bus transactions or the second protocol bus transaction.

14. The system of claim 13, wherein the status of the device management bus is determined based on the bus status counters indicating stalled progress in the first protocol bus transactions or the second protocol bus transaction.

15. A remote management controller for detecting and recovering from an inoperable device management bus, the remote management controller configured to:

offload device management bus transactions using a first protocol to a bus protocol controller managing bus operations using the first protocol;

update bus status counters stored in a shared memory based on the status of the offloaded first protocol transactions;

process device management bus transactions using a second protocol;

updating the bus status counters based on the status of the second protocol transactions;

determine the status of the device management bus based on the bus status counters in shared memory; and

if the device management bus is inoperable, reset the remote management controller.

16. The remote management controller of claim 15, wherein the bus protocol controller comprises an I2C co-processor of the remote management controller.

17. The remote management controller of claim 16, wherein the first protocol comprises MCTP (Management Component Transport Protocol) messaging.

18. The remote management controller of claim 17, wherein the second protocol comprises IPMB (Intelligent Platform Management Bus) bus transactions.

19. The remote management controller of claim 15, wherein the status of the device management bus is determined based on the bus status counters indicating consecutive errors in the first protocol bus transactions or the second protocol bus transaction.

20. The remote management controller of claim 19, wherein the status of the device management bus is determined based on the bus status counters indicating stalled progress in the first protocol bus transactions or the second protocol bus transaction.