Remix.run Logo
dragontamer 3 days ago

If you care about long term storage, make a NAS and run ZFS scrub (or equivalent) every 6 months. That will check for errors and fix them as they come up.

All error correction has a limit. If too many errors build up, it becomes unrecoverable errors. But as long as you reread and fix them within the error correction region, it's fine.

csdvrx 3 days ago | parent | next [-]

> run ZFS scrub (or equivalent) every 6 months

zfs in mirror mode offers redundancy at the block level but scrub requires plugging the device

> All error correction has a limit. If too many errors build up, it becomes unrecoverable errors

There are software solutions. You can specify the redundancy you want.

For long term storage, if using a single media that you can't plug and scrub, I recommend par2 (https://en.wikipedia.org/wiki/Parchive?useskin=vector) over NTFS: there are many NTFS file recovery tools, and it shouldn't be too hard to roll your own solution to use the redundancy when a given sector can't be read

WalterGR 3 days ago | parent | prev | next [-]

What hardware, though? I want to build a NAS / attached storage array but after accidentally purchasing an SMR drive[0] I’m a little hesitant to even confront the project.

A few tens of TBs. Local, not cloud.

[0] Maybe 7 years ago. I don’t know if anything has changed since, e.g. honest, up-front labeling.

[0*] For those unfamiliar, SMR is Shingled Magnetic Recording. https://en.m.wikipedia.org/wiki/Shingled_magnetic_recording

justinclift a day ago | parent | next [-]

I have a homelab with a bunch of old HP Gen 8 Microservers. They hold 4x 3.5" hdds and also an ssd (internally, replacing the optical slot):

https://www.ebay.com/itm/156749631079

These are reasonably low power, and can take up to 16GB of ECC ram which is fine for small local NAS applications. The cpu is socketed, so I've upgraded most of mine to 4 core / 8 thread Xeons. From rough memory of the last time I measured the power usage at idle, it was around 12w with the drives auto-spun down.

They also have a PCIe slot in the back, though it's older gen, but you'll be able to put a 10GbE card in it if that's your thing.

Software wise, TrueNAS works pretty well. Proxmox works "ok" too, but this isn't a good platform for virtualisation due to the maximum of 16GB ram.

matheusmoreira 3 days ago | parent | prev | next [-]

> What hardware, though?

Good question. There seems to be no way to tell whether or not we're gonna get junk when we buy hard drives. Manufacturers got caught putting SMR into NAS drives. Even if you deeply research things before buying, everything could change tomorrow.

Why is this so hard? Why can't we have a CMR drive that just works? That we can expect to last for 10 years? That properly reports I/O errors to the OS?

code_biologist 3 days ago | parent | prev | next [-]

The Backblaze Drive Stats are always a good place to start: https://www.backblaze.com/blog/backblaze-drive-stats-for-202...

There might be SMR drives in there, but I suspect not.

wmf 3 days ago | parent | prev | next [-]

Nothing can really save you from accidentally buying the wrong model other than research. For tens of TBs you can use either 4-8 >20TB HDDs or 6-12 8TB SSDs (e.g. Asustor). The difference really comes down to how much you're willing to pay.

3np 3 days ago | parent | prev | next [-]

Toshi Nx00/MG/MN are good picks. The company never failed us and I don't believe they've had the same kinds of controversies as the US competition.

Please don't tell everyone so we can still keep buying them? ;)

dragontamer 2 days ago | parent | prev [-]

SMR will store your data, just slowly.

It was a mistake for the Hard Drive business community to push them so hard IMO. But these days the 20TB+ drives are all HAMR or other heat/energy assisted tech.

If you are buying 8TB or so, just make sure to avoid SMR but otherwise you're fine. Even then, SMR stores data fine, it's just really really slow.

ErneX 3 days ago | parent | prev [-]

I use TrueNAS and it does a weekly scrub IIRC.