Systems and methods to initiate device recovery
The methods and systems provide remote device recovery using a sideband communication path in response to an error occurring in the device. A management entity sends a sideband communication using the sideband communication path to a logic circuit in communication with the device. The logic circuit initiates the recovery of the device.
1 . A system, comprising:
a network interface controller (NIC) with a plurality of communication paths;
a controller communicatively coupled to the NIC via a first communication path, wherein the NIC remotely monitors the controller to ensure that the controller is operating properly; and
a logic circuit communicatively coupled to the NIC and the controller via a sideband communication path separate and apart from the first communication path to provide communication between the NIC and the controller, when the controller is not operating properly and, wherein the logic circuit is configured to:
receive a sideband communication from the NIC via the sideband communication path to initiate a control function on the controller, wherein the controller is malfunctioning, inoperable, or non-responsive to communications from the NIC via the first communication path;
receive, via the sideband communication path, a control operation for devices on the system;
initiate the control function on the controller; and
initiate the control operation on the devices.
2 . The system of claim 1 , wherein the first communication path is a primary communication path between the NIC and the controller, and the sideband communication path is a redundant path between the NIC and the controller.
3 . The system of claim 1 , wherein the controller is a baseboard management controller (BMC).
4 . The system of claim 1 , wherein the NIC is a network interface controller (NIC) with dual communication paths.
5 . The system of claim 1 , wherein the control function is a recovery function that initiates a recovery of the controller.
6 . The system of claim 5 , wherein the control function initiates a reset of the controller.
7 . The system of claim 5 , wherein the control function places the controller in a safe mode or returns the controller to a last known good state of the controller.
8 . The system of claim 5 , wherein the control function restarts the controller.
9 . The system of claim 1 , wherein the control function initiates a diagnostic of the controller.
10 . A system, comprising:
a network interface controller (NIC) management entity with a plurality of communication paths;
a baseboard management controller (BMC) communicatively coupled to the NIC management entity via a first communication path; and
a logic circuit communicatively coupled to the BMC and the NIC management entity via a sideband communication path that is separate and apart from the first communication path, wherein the logic circuit is configured to:
receive a communication from the NIC management entity via the sideband communication path to initiate a control function on the BMC, wherein an error occurred on the BMC;
receive, via the sideband communication path, a control operation for devices on the system;
initiate the control function on the BMC; and
initiate the control operation on the devices.
11 . The system of claim 10 , wherein the error causes the BMC to operate incorrectly.
12 . The system of claim 10 , wherein the error causes the BMC to become non-responsive.
13 . The system of claim 10 , wherein the BMC further comprises a reset pin; and
the logic circuit initiates the control function on the BMC by generating a signal on the reset pin to change a state of the reset pin and restart the BMC.
14 . The system of claim 10 , wherein the BMC further comprises firmware; and
the logic circuit initiates the control function on the BMC by generating a signal to revert the firmware, flash the firmware, or update the firmware.
15 . The system of claim 10 , wherein the management entity is a network interface controller (NIC).
16 . The system of claim 15 , wherein the first communication path is a primary communication path between the NIC and the BMC.
17 . The system of claim 15 , wherein the sideband communication path comprises an Inter-Integrated Circuit (I2C) communication path between the NIC and the logic circuit.
18 . A method, comprising:
receiving, at a logic circuit, a sideband communication from a network interface controller (NIC) management entity to initiate a control function for a baseboard management controller (BMC) on a sideband communication path that is separate and apart from a primary communication path between the NIC management entity and the BMC;
receiving, via the sideband communication path, a control operation for devices on the system;
initiating, by the logic circuit, the control function on the BMC; and
initiating, by the logic circuit, the control operation on the devices.
19 . The method of claim 18 , wherein initiating the control function on the BMC further includes:
generating a signal on a reset pin of the BMC to change a state of the reset pin and restart the BMC; or
generating a signal revert firmware of the BMC, flash the firmware of the BMC, or update the firmware of the BMC.
20 . The method of claim 18 , wherein the NIC management entity is a network interface controller (NIC), and the sideband communication path comprises an Inter-Integrated Circuit (I2C) communication path between the NIC and the logic circuit.