Grading Checklist: RAID Degraded Rebuild Latency

A good response must include:

• Checked /proc/mdstat to confirm array state, rebuild progress, and estimated completion time
• Reviewed dmesg / smartctl for signs of additional drive degradation
• Identified the rebuild speed limits (/proc/sys/dev/raid/speed_limit_min and speed_limit_max) and recommended tuning
• Considered adjusting stripe_cache_size to balance rebuild speed vs. application I/O
• Checked the I/O scheduler in use and evaluated whether a change would help
• Assessed the risk of a second drive failure during rebuild (RAID-6 tolerates it, RAID-5 does not)
• Proposed a plan to reduce production I/O load during rebuild (read replicas, failover, traffic shifting)
• Verified the replacement drive is healthy using smartctl
• Mentioned monitoring for URE (Unrecoverable Read Errors) during rebuild
• Documented a rollback or escalation plan if the rebuild fails or another drive drops
• Considered whether ionice or cgroup I/O throttling could help prioritize application I/O
• Communicated timeline and risk to stakeholders (DB team, application owners)