Multiple block error correction in an information handling system
An information handling system includes a first memory and a baseboard management controller. The first memory stores a first firmware partition and a second firmware partition. The baseboard management controller includes a second memory. The baseboard management controller begins execution of a DM-Verity daemon, and performs periodic patrol reads of the first firmware partition. The baseboard management controller detects one or more block failures in the first firmware partition, and stores information associated with the one or more block failures in a message box of the second memory. In response to the entire first firmware partition being scanned, the baseboard management controller switches a boot partition from the first firmware partition to the second firmware partition, and initiates a reboot of the information handling system.
1. An information handling system comprising:
a first memory to store a first firmware partition and a second firmware partition; and
a baseboard management controller including a second memory, the baseboard management controller to:
begin execution of a DM-Verity daemon;
perform periodic patrol reads of the first firmware partition;
detect one or more block failures in the first firmware partition;
store information associated with the one or more block failures in a message box of the second memory;
in response to the entire first firmware partition being scanned, switch a boot partition from the first firmware partition to the second firmware partition;
initiate a reboot of the information handling system; and
upon the reboot:
if a flag is set in the message box, then perform a full rescan of the first firmware partition; and
in response to the flag being set:
read the information in the message box of the second memory; and
based on the information read from the message box, determine the block failures in the first firmware partition.
2. The information handling system of claim 1 , the baseboard management controller further to:
read data from a block of the second firmware partition, wherein the block is associated with one of block failures in the first firmware partition; and
write the data from the block of the second firmware partition to the associated block failure of the first firmware partition.
3. The information handling system of claim 2 , wherein block of the second firmware partition is a cryptographically verified good block of backup data for the block failure of the first firmware partition.
4. The information handling system of claim 1 , wherein in response to data of all block failures of the first firmware partition being replaced by data in corresponding blocks of the second firmware partition, the baseboard management controller further to:
switch the boot partition from the second firmware partition to the first firmware partition; and
initiate the reboot of the information handling system.
5. The information handling system of claim 1 , wherein the baseboard management controller further to:
determine whether an error threshold of the message box has been exceeded; and
in response to the error threshold being exceeded, set a flag in the message box.
6. The information handling system of claim 1 , wherein in response to the reboot of the information handling system, the baseboard management controller further to:
utilize forward error correction to correct all of the block failures of the first firmware partition.
7. A method comprising:
beginning, by a processor of an information handling system, execution of a DM-Verity daemon;
performing periodic patrol reads of the first firmware partition;
detecting, via the DM-Verity daemon one or more block failures in a first firmware partition of the information handling system;
storing information associated with the block failures in a message box of a memory of the information handling system;
in response to the entire first firmware partition being scanned, switching a boot partition from the first firmware partition to a second firmware partition;
determining whether an error threshold of the message box has been exceeded;
in response to the error threshold being exceeded, setting a flag in the message box;
initiating a reboot of the information handling system; and
upon the reboot:
if a flag is set in the message box, then performing a full rescan of the first firmware partition; and
in response to the flag being set:
reading the information in the message box of the second memory; and
based on the information read from the message box, determining the block failures in the first firmware partition.
8. The method of claim 7 , further comprising:
reading data from a block of the second firmware partition, wherein the block is associated with one of block failures in the first firmware partition; and
writing the data from the block of the second firmware partition to the associated block failure of the first firmware partition.
9. The method of claim 8 , wherein block of the second firmware partition is a cryptographically verified good block of backup data for the block failure of the first firmware partition.
10. The method of claim 7 , wherein in response to data of all block failures of the first firmware partition being replaced by data in corresponding blocks of the second firmware partition, the method further comprises:
switching the boot partition from the second firmware partition to the first firmware partition; and
initiating the reboot of the information handling system.
11. The method of claim 7 , wherein in response to the rebooting of the information handling system, the method further comprises:
utilizing forward error correction to correct all of the block failures of the first firmware partition.
12. A method comprising:
beginning, by a processor of an information handling system, execution of a DM-Verity daemon;
performing periodic patrol reads of the first firmware partition;
detecting, via the DM-Verity daemon one or more block failures in a first firmware partition of the information handling system;
storing information associated with the one or more block failures in a message box of a memory of the information handling system;
in response to the entire first firmware partition being scanned, switching a boot partition from the first firmware partition to a second firmware partition;
initiating a reboot of the information handling system; and
upon the reboot:
if a flag is set in the message box, then performing a full rescan of the first firmware partition; and
in response to the flag being set:
reading the information in the message box of the second memory; and
based on the information read from the message box, determining the block failures in the first firmware partition.
13. The method of claim 12 , further comprising:
reading data from a block of the second firmware partition, wherein the block is associated with one of block failures in the first firmware partition; and
writing the data from the block of the second firmware partition to the associated block failure of the first firmware partition.
14. The method of claim 13 , wherein block of the second firmware partition is a cryptographically verified good block of backup data for the block failure of the first firmware partition.
15. The method of claim 12 , wherein in response to the rebooting of the information handling system, the method further comprises:
utilizing forward error correction to correct all of the block failures of the first firmware partition.