问题原文
原帖链接 · kontextify · CC BY-SA 3.0
I have an Ubuntu server with two HDD drives in a software RAID1 array. These are identical drives, and I'm assuming they're seeing the same level of read/write activity over their lifetimes (correct me if I'm wrong).
Does that mean it's reasonable to expect these drives to fail with 1-2 weeks of each other? Or am I missing something about RAID1 and disk failure?
已采纳解答
原帖链接 · Grant · CC BY-SA 3.0
Expect? Not necessarily.
Plan for it happening? Definitely.
Disks from the same batch put under the same workload can all die around the same time. Or one can die early and the other last 10 years. All depends on tiny defects in the disks and when they start causing issues.
RAID is not a backup solution. If both disks die you restore from your backups. Make sure you have backups.
RAID's purpose is to make a single disk failure not take out a server/service immediately. If one disk dies you swap it out as soon as possible and everything keeps running. If the other disk dies a week later, you repeat the process.
摘自公开归档,保留原文语言,版式经整理,外部媒体与邮箱地址已移除。
