TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

CSI: Backblaze – Dissecting 3TB Drive Failure

12 pointsby damian2000about 10 years ago

1 comment

trounceabout 10 years ago
I have 2 lightly-used home ZFS servers (OmniOS&#x2F;Napp-IT), each with 8x ST3000DM001 drives, for 16x drives total. The drives were purchased in mid to late 2012, all shucked from external units. It was by far the cheapest way to get storage at the time, given the Thailand flooding.<p>I&#x27;ve had to buy replacements for complete or partial failure, for 3 in 2013, 10 in 2014, and 1 in 2015 so far. 4 of the replacements have experienced failures too, including 2 RMA returns after I switched to buying internal drives with warranties. Only 4 of the original 16 remain.<p>These stats match my experience. The ST3000DM001 drives are by far the least reliable drives I&#x27;ve ever experienced in 20 years of using drive arrays. Fortunately, with two ZFS raidz arrays mirroring each other, I haven&#x27;t had any data loss.
评论 #9421865 未加载
评论 #9420586 未加载