System recovery using WoL

1. A computer-implemented method for resetting a management device of a system, comprising:

receiving, at a network interface (NIC) of the system, a wake-on-lan (WoL) message via a switch of the system;

receiving, at a monitoring device of the system, status information of a management device of the system;

in an event that the management device is frozen according to the status information, causing the management device to be reset through the monitoring device; and

in an event that the management device is not frozen according to the status information, shielding the management device from the WoL message to prevent the management device from being reset.

2. The computer-implemented method of claim 1 , further comprising:

determining that the management device is not frozen based upon the status information; and

shielding the management device from the WoL message to prevent the management device from being reset.

3. The computer-implemented method of claim 1 , wherein receiving, at the NIC of the system, the WoL message comprises:

receiving, at a physical layer hardware (PHY) of the system, the WoL message over a network; and

receiving, at the switch, the WoL message from the PHY.

4. The computer-implemented method of claim 3 , further comprising:

receiving, at the switch, a network packet via the PHY;

determining that the management device recovers from being frozen; and

causing the network packet to be forwarded directly from the switch to the management device.

5. The computer-implemented method of claim 3 , wherein the switch communicates with the PHY via a first serial gigabit media independent interface (SGMII) and communicates with the NIC via a second SGMII.

6. The computer-implemented method of claim 3 , wherein the switch communicates with the management device via a reduced gigabit media independent interface (RGMII).

7. The computer-implemented method of claim 1 , further comprising:

receiving, at a delay circuit, the WoL message from the NIC; and

causing the NIC to be reset by the delay circuit.

8. The computer-implemented method of claim 7 , wherein causing the NIC to be reset by the delay circuit comprises:

withholding the WoL message at the delay circuit for a threshold time period; and

causing the WoL message to be passed on to the NIC at the end of the threshold time period.

9. The computer-implemented method of claim 7 , wherein the delay circuit is a LC circuit or an RC circuit.

10. The computer-implemented method of claim 1 , wherein the monitoring device comprises a single-stage watchdog circuit or a multistage watchdog circuit.

11. The computer-implemented method of claim 1 , wherein the monitoring device comprises a first input pin to receive the status information of the management device and a second input pin to receive the WoL message.

12. The computer-implemented method of claim 11 , wherein the status information is received directly from the management device or from a separate device monitoring status of the management device.

13. The computer-implemented method of claim 11 , wherein the monitoring device is connected to the management device via an output pin of the monitoring device and a power-on reset pin of the management device.

14. A system, comprising:

a processor; and

a computer-readable medium storing instructions that, when executed by the processor, cause the system to perform operations comprising:

receiving, at a network interface (NIC) of the system, a wake-on-lan (WoL) message via a switch of the system;

receiving, at a monitoring device of the system, status information of a management device of the system;

in an event that the management device is frozen according to the status information, causing the management device to be reset through the monitoring device; and

in an event that the management device is not frozen according to the status information, shielding the management device from the WoL message to prevent the management device from being reset.

15. The system of claim 14 , wherein the instructions, when executed by the processor, cause the system to perform operations comprising: determining that the management device is not frozen based upon the status information; and shielding the management device from the WoL message to prevent the management device from being reset.

16. The system of claim 14 , wherein receiving, at the NIC of the system, the WoL message comprises: receiving, at a physical layer hardware (PHY) of the system, the WoL message over a network; and receiving, at the switch, the WoL message from the PHY.

17. The system of claim 16 , wherein the instructions, when executed by the processor, cause the system to perform operations comprising: receiving, at the switch, a network packet via the PHY; determining that the management device recovers from being frozen; and causing the network packet to be forwarded directly from the switch to the management device.

18. The system of claim 14 , wherein the instructions, when executed by the processor, cause the system to perform operations comprising: receiving, at a delay circuit, the WoL message from the NIC; and causing the NIC to be reset by the delay circuit.

19. The system of claim 18 , wherein causing the NIC to be reset by the delay circuit comprises: withholding the WoL message at the delay circuit for a threshold time period; and causing the WoL message to be passed on to the NIC at the end of the threshold time period.

20. The system of claim 14 , wherein the monitoring device comprises a first input pin to receive the status information of the management device and a second input pin to receive the WoL message, and wherein the monitoring device is connected to the management device via an output pin of the monitoring device and a power-on reset pin of the management device.