Waryle

@Waryle@jlai.lu

• ⁨Comment⁩ on ⁨used hard drive retailers⁩ ⁨⁨13⁩ ⁨hours⁩ ago⁩:

  Sure, in a ultra-consumerist world, everybody buy all things new. Now that hard drives cost a kidney, you buy refurbished ones, and you realize that it’s not that bad, especially if you use the right mitigations.

  And that’s a trend that may grow and spread to other things, considering that mondialisation will get increasingly difficult due to geopolitics troubles, and petrol becoming costlier over the next years/decades.

• ⁨Comment⁩ on ⁨used hard drive retailers⁩ ⁨⁨16⁩ ⁨hours⁩ ago⁩:

  Bought 4x10TB drives 2 years ago for 385€ on eBay, from a 20+ years old seller account with tens of thousands of good reviews with a long warranty on drives, to build a local media server + some other services for family and friends.

  One was dead on arrival: they just sent me another one, no question asked.

  I have them set-up in a RAIDZ1 array, so I can have one disk down and still have the system up.

  I have my services and critical data backed up daily (with about twenty snapshots that go back to months before) with Restic to both a Hetzner storage box and a local external drive (both tested for recovery).

  I have one disk that failed me a few months ago: sent the disk to the seller, they tested it, and sent me a new 12TB drive because they didn’t have 10TB anymore.

  Added it to my ZFS pool, and since then, everything is ok. My few users didn’t even notice anything going on: downtime was minimal, performance was ok.

  If more than one disk fails, I have my data on my external drive. If my external drive fails, I can download everything from Hetzner. In both situations, I wait for new drives, replace the old ones in the pool, and restore my snapshots. The downtime is annoying, but I’m not running a business so nobody is paying for anything, so it’s ok.