Did decide to go the used drive route just saving up and waiting for the right deal from a good seller. I did notice on eBay specifically that some sellers have tested drives for a great price per tb with a shit seller’s warranty (30 days to 1 year) but offer insurance for a significantly longer length of time for a few more bucks. I’m wondering if that’s a good alternative to having a long seller’s warranty? Just assume I will back everything up properly so drive failure won’t be a massive concern I just want the option of returning it or getting a replacement if need be.
If you “RAID and backup enough” any chepo used hard drive is a good and valid. :)
Generally speaking datacenter drives with manufacturer warranty will be a good option because they’re easy to get replaced by the manufacturer without much fuzz. For any second hard drive I look at bad blocks > count of starts > time they’ve been running and if it isn’t an absurd number they should be fine.
If you put the effort in to get a wide range of sources/batches AND monitor your disks regularly: Agreed
But if you get a few from the same manufacturing batch/source? Then there are shockingly decent odds that they will fail at roughly the same time. Which is a huge problem if you aren’t monitoring your disks and able to “recover” from however many failures before they take down the array.
I completely agree with you, no discussion there. Active monitoring is required. Backup in 3-2-1 would solve the same source batch issue. Some backups should be kept spun down / cold storage so the likelihood of failure at the same time is close to none. Still I would mix the production and backups HDDs to avoid that.
I guess my general approach in my personal life is that I would rather spend a bit more on some refurbs of server grade hdds from a reputable outlet. Because once I start assuming all of my drives will fail at any moment I need much more hierarchical storage and a lot more replacement drives and so forth. And my actual off site backups are a much smaller subset of my data that is in an encrypted volume in a cloud storage provider’s bucket.
And in my professional life: You are paying for the warranties and support contracts. If you can’t afford to run your own storage then you should just call Amazon and ask for a good deal.
And in my professional life: You are paying for the warranties and support contracts. If you can’t afford to run your own storage then you should just call Amazon and ask for a good deal.
Oh yeah. Or don’t have storage at all, because if you can’t afford it you most likely don’t need it :P
You are not logged in. However you can subscribe from another Fediverse account, for example Lemmy or Mastodon. To do this, paste the following into the search field of your instance: !piracy@lemmy.dbzer0.com
⚓ Dedicated to the discussion of digital piracy, including ethical problems and legal advancements.
If you “RAID and backup enough” any chepo used hard drive is a good and valid. :)
Generally speaking datacenter drives with manufacturer warranty will be a good option because they’re easy to get replaced by the manufacturer without much fuzz. For any second hard drive I look at bad blocks > count of starts > time they’ve been running and if it isn’t an absurd number they should be fine.
Yes and no.
If you put the effort in to get a wide range of sources/batches AND monitor your disks regularly: Agreed
But if you get a few from the same manufacturing batch/source? Then there are shockingly decent odds that they will fail at roughly the same time. Which is a huge problem if you aren’t monitoring your disks and able to “recover” from however many failures before they take down the array.
I completely agree with you, no discussion there. Active monitoring is required. Backup in 3-2-1 would solve the same source batch issue. Some backups should be kept spun down / cold storage so the likelihood of failure at the same time is close to none. Still I would mix the production and backups HDDs to avoid that.
Yeah.
I guess my general approach in my personal life is that I would rather spend a bit more on some refurbs of server grade hdds from a reputable outlet. Because once I start assuming all of my drives will fail at any moment I need much more hierarchical storage and a lot more replacement drives and so forth. And my actual off site backups are a much smaller subset of my data that is in an encrypted volume in a cloud storage provider’s bucket.
And in my professional life: You are paying for the warranties and support contracts. If you can’t afford to run your own storage then you should just call Amazon and ask for a good deal.
Oh yeah. Or don’t have storage at all, because if you can’t afford it you most likely don’t need it :P