[LU-3176] linux md raid-6 read-error correction in degraded state

• Type: Bug
• Resolution: Fixed
• Priority: Critical
• Affects Version/s: Lustre 1.8.9
• Labels: patch
• Environment: CentOS5.x/RHEL5.x, Linux MD Raid
• Severity: 3
• Assignee: Yang Sheng
• Reporter: Shuichi Ihara (Inactive)

Description

md raid6 is not doing correct read-error when single disk failed and going to be degraded mode for one more drive. I think original code only considered raid5, but did't work for raid6.
This is fixed in mainstream linux kernel.
http://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/commit/?id=87aa63000c484b