System and method for recovery of sideband interfaces for controllers
Periods of interoperability of sideband buses prevent effective management of managed devices by a remote access controller. Embodiments avoid periods of inoperability of sideband buses and recover the sideband bus without resetting the managed devices or the IHS (Information Handling System). The remote access controller configures timer and transmits the timer to a managed device. The managed device monitors the sideband for messages for the remote access controller. If no messages are received before expiration of the timer, the managed device resets its sideband bus endpoint, without resetting the managed device.
1. A system for managing a plurality of managed devices via a sideband bus, the system comprising:
a remote access controller configured to:
configure, for a first managed device of the plurality of managed devices, a first timer to expire after a first duration;
transmit a timer command to the first managed device via the sideband bus, wherein the timer command specifies the first timer;
transmit temperature requests to the first managed device via the sideband bus at periodic intervals;
the first managed device coupled to the remote access controller via the sideband bus, wherein the first managed device comprises a sideband bus monitor configured to:
start a first timer for the first duration based on receipt of the timer command from the remote access controller;
monitor the sideband bus for the temperature requests from the remote access controller;
if a temperature request from the remote access controller is detected prior to the expiration of the first duration, restart the first timer and report collected temperature sensor readings to the remote access controller; and
if no temperature request from the remote access controller is detected prior to the expiration of the first duration, reset the sideband bus without restarting the first managed device; and
wherein the remote access controller is further configured to determine a second duration until the first managed device resets the sideband bus, if the remote access controller does not receive the temperature sensor readings before beginning a next periodic interval.
2. The system of claim 1 , wherein the first duration is selected for the first timer by the remote access controller to be greater than the periodic interval at which temperature requests are transmitted by the remote access controller.
3. The system of claim 1 , wherein the first managed device is a host bus adapter coupled to a plurality of storage devices.
4. The system of claim 3 , wherein, in response to the detection of a temperature request prior to the expiration of the first duration by the host bus adapter on a periodic, the host bus adapter collects the temperature sensor readings from the plurality of storage devices.
5. The system of claim 4 , wherein the host bus adapter relays the temperature sensor readings from the plurality of storage devices to the remote access controller via the sideband bus.
6. The system of claim 1 , wherein the remote access controller is further configured to signal closed-loop cooling using the temperature sensor readings from the first managed device if the temperature sensor readings are received prior to the expiration of the second duration.
7. The system of claim 1 , wherein the remote access controller is further configured to signal fail-over cooling if the temperature sensor readings are not received prior to the expiration of the second duration.
8. A method for managing a plurality of managed devices via a sideband bus, the method comprising:
configuring, by a remote access controller, a first timer for a first managed device of the plurality of managed devices, wherein the first timer is configured to expire after a first duration;
transmit a timer command from the remote access controller to the first managed device via the sideband bus, wherein the timer command specifies the first timer configuration;
transmit temperature requests to the first managed device via the sideband bus at periodic intervals;
starting, by a sideband bus monitor of the first managed device, the first timer for the first duration based on receipt of the timer command from the remote access controller;
monitoring the sideband bus for the temperature requests communications from the remote access controller;
restarting the first timer and reporting collected temperature sensor readings to the remote access controller, if a temperature request from the remote access controller is detected prior to the expiration of the first duration, and
reset the sideband bus without restarting the first managed device, if no temperature request from the remote access controller is detected prior to the expiration of the first duration;
detecting that the remote access controller does not receive the temperature sensor readings before beginning a next periodic interval; and
responsive to the detecting, determining, by the remote access controller, a second duration until reset of the sideband bus by the first managed device.
9. The method of claim 8 , wherein the first duration is selected for the first timer by the remote access controller to be greater than the periodic interval at which temperature requests are transmitted by the remote access controller.
10. The method of claim 8 , wherein the first managed device is a host bus adapter coupled to a plurality of storage devices.
11. The method of claim 10 , further comprising:
collecting, by the host bus adapter in response to the detection of a temperature request prior to the expiration of the first duration, the temperature sensor readings from the plurality of storage devices.
12. The method of claim 11 , further comprising:
relaying, by the host bus adapter via the sideband bus, the temperature sensor readings from the plurality of storage devices to the remote access controller.
13. The method of claim 8 , further comprising:
signaling, by the remote access controller, closed-loop cooling using the temperature sensor readings from the first managed device if the temperature sensor readings are received prior to the expiration of the second duration.
14. The method of claim 8 , further comprising:
signaling, by the remote access controller, fail-over cooling if the temperature sensor readings are not received prior to the expiration of the second duration.
15. An Information Handling System (IHS) comprising:
a remote access controller configured to:
configure, for a first managed device, a first timer to expire after a first duration;
transmit a timer command to the first managed device via a sideband bus, wherein the timer command specifies the first timer configuration;
transmit temperature requests to the first managed device via the sideband bus at periodic intervals;
the first managed device coupled to the remote access controller via the sideband bus, wherein the first managed device comprises a sideband bus monitor configured to:
start a first timer for the first duration based on receipt of the timer command from the remote access controller;
monitor the sideband bus for the temperature requests from the remote access controller;
if a temperature request from the remote access controller is detected prior to the expiration of the first duration, restart the first timer and report collected temperature sensor readings to the remote access controller; and
if no temperature request from the remote access controller is detected prior to the expiration of the first duration, reset the sideband bus without restarting the IHS, and
wherein the remote access controller is further configured to determine a second duration until the first managed device resets the sideband bus, if the remote access controller does not receive the temperature sensor readings before beginning a next periodic interval.
16. The IHS of claim 15 , wherein the first duration is selected for the first timer by the remote access controller to be greater than the periodic interval at which temperature requests are transmitted by the remote access controller.
17. The IHS of claim 16 , further comprising:
a plurality of storage devices, wherein the first managed device is a host bus adapter coupled to the plurality of storage device.
18. The IHS of claim 17 , wherein the host bus adapter relays temperature sensor readings from the plurality of storage devices to the remote access controller via the sideband bus.
19. The IHS of claim 18 , wherein the remote access controller is further configured to signal closed-loop cooling using the temperature sensor readings from the first managed device if the temperature sensor readings are received prior to the expiration of the second duration.
20. The IHS of claim 15 , wherein the remote access controller is further configured to signal fail-over cooling if the temperature sensor readings are not received prior to the expiration of the second duration.