I'm curious if I'm alone in this unfortunate experience: has anyone else faced the issue of both mirrored disks in a RAID 1 array failing at the same time or very close to it? This has happened to me twice now! The whole idea of RAID 1 is to have protection against a single drive failure, yet I've ended up with zero redundancy in two different systems over the years. I'm reaching out to see if anyone else has faced this issue and what you think might have caused it. Was it a power surge, a problematic backplane, or maybe drives from a faulty batch bought together? Also, what's your top recommendation for RAID 1 hardware or drives? Am I really that unlucky, or is this a more common occurrence than people realize? Thanks in advance for sharing your experiences!
5 Answers
Yeah, I've seen this too. Drives from the same batch often fail together. I've had two drives die on me because of a UPS failure. Lately, I shifted to using ZFS because it provides more robust features like snapshots and better scrubbing capabilities compared to mdadm checks.
Both RAID 1 and RAID 5 have their downsides. If you're not reading your data often, bad sectors might develop unnoticed. This can cause issues when rebuilding after a failure. It's crucial to have regular checks—SMART tests and automated alerts can really help here. I'm also moving towards ZFS, as it seems more reliable overall for data integrity.
It's a known issue called common mode failure. The general advice is to avoid using drives from the same lot as they could fail around the same time, which defeats the purpose of redundancy. Also, keep in mind that RAID isn't a backup solution. Any drive can fail unexpectedly, including during degraded modes.
Just a heads-up—make sure your disks are actually bad before panicking. There are lots of factors to consider, like how the drives are connected and the type of controller you're using. If there's a communication issue, mdadm might preemptively fail a healthy drive. Plus, check your PSU and make sure it's functioning well; that can lead to unexpected drive failures, too!
Thanks for the insights! I’ll definitely investigate the connections more.
It's definitely not uncommon to have simultaneous failures. I've encountered it a few times myself. To combat this, I now use two disks from different manufacturers in my RAID setups to minimize risk.
Good point! I ran into issues with cabling before, so I always double-check the connections.