Live data from Hacker News

Seagate: 'new' hard drives used for tens of thousands of hours

tomshardware.com

191–200 of 227 posts

Re: Seagate: 'new' hard drives used for tens of thousands of hours

#191

Earlier quoted context omitted.

There was an early SMR talk by a WD (IIRC) engineer to a crowd of ZFS devs and sysadmins, and the engineer thought it should fit well because he misheard the default record size being 128MB rather than the actual 128kB, so thought it one record would fit one zone.

That seems like a pretty big mistake, are you sure that's real? ZFS has a lot of genuine potential to work well with SMR drives, but that potential depends on glue code that nobody has written yet. I think BTRFS is halfway there?

BTRFS can handle zoned disks. The doc claims that it was a natural fit with the way BTRFS works.

The problem is that you cannot buy host managed zoned disks.

Re: Seagate: 'new' hard drives used for tens of thousands of hours

#192
post #189

Years ago, a Seagate HD suddenly crashed. Wasn't particularly old. Then some point later, there was a good deal on Seagate HDs, and I bought another one. Also crashed "early" in its life. I never buy them again. The main PC building forum I used to frequent also have a pinned topic announcing that they've lost all hope with Seagate and will never recommend it for anyone asking for help on a build. That was 10 years a…

One of their drives was famously so unreliable that it has its own wikipedia page [1]. Guess who had two of these fail almost at the same time in his NAS a couple of years back. 1 - https://en.m.wikipedia.org/wiki/ST3000DM001

Had 4 of the 5000 version of the models. Quickly moved to other drives. Got lucky.

Re: Seagate: 'new' hard drives used for tens of thousands of hours

#193
post #61

You can read out the FARM logs of Seagate hard drives using smartctl -l farm /dev/sd They're supposed to be "more trustworthy" than the regular SMART stats. (My two "new" 16TB Exos drives had 0 hours (regular) and ~18k hours (farm) - DOM 04JUN2021 and 07JUN2021. Also, zfs refused to format the drive: 'already formatted as ddf_raid_member'.)

Wow, this is great. I don't care about "more trustworthy", but this has so much more information than regular smart data.

Incredible. Even stuff like range of voltages on voltage rails that the drive has seen, etc. I'm proud of my power supply, looking at the data. :D

Re: Seagate: 'new' hard drives used for tens of thousands of hours

#194
post #28
post #2

> the company launched an official eBay store that sells refurbished drives. [...] However, this store only sells in the US One could make a joke that seagate did start selling refurbs in EU, just without telling anyone. Why is it always seagate when there's something wrong with HDDs?

there are only a handful of companies that make HDD so it will be one of them. WD has a terrible reputation as well. Who else makes HDDs? (Toshiba is the only one I know of and they never seem to come up with someone asks for recommendations so I'm not sure if they are small, so bad nobody would think about them, or just hard to get retail)

I'm suspecting Toshiba might just be "a well-kept open secret" and people want to keep them for themselves. That's my only explanation for their absence in recommendations. Perhaps combined with that probably the other manufacturers actively market in ways that Toshiba doesn't care for.

Gone through a bunch of their MG/MN series drives over 5+y, and the 3/5y warranty was honored without BS on the one RMA case. You can also see them track well in Backblaze rankings.

Their N series are supposedly also great but never saw the point in paying the premium.

I have no idea why you would pay more or the same for same-sized drives from either Seagate or WD if you have the option.

Re: Seagate: 'new' hard drives used for tens of thousands of hours

#195
post #170

Has anyone had experience involving the police in matters like this? I reached out to the German news source that first reported on this issue, but they couldn’t help. The writer of the article mentioned that "The dealer will probably hide behind the supplier." Here's the response I received today from Böttcher AG, where I purchased the hard drives, after sending them the FARM logs and photos of the drives (translate…

I think in many countries you’d be better off reaching out to a consumer protection-specific agency or tribunal, making sure to cc the vendor or manufacturer. For instance, in the United States, you could contact the Federal Trade Commission or your state’s attorney general, or potentially file a case in small claims court if the matter isn’t quickly resolved.

You might get a better response from regulators if you wait until you do have issues getting a refund, but that probably also between jurisdictions. Good luck!

Re: Seagate: 'new' hard drives used for tens of thousands of hours

#196
post #6

Article is about reports from German customers (this is clear in the orignal title, but not the current HN title), and mostly with 16TB Exos HDDs. FWIW, I just bought a bunch of Seagate HDDs (US vendor) and my warranty periods line up with the DOMs on the packaging. You can check your warranty period/status at: https://www.seagate.com/support/warranty-and-replacements

Thanks! I just checked an 10TB Ironwolf bought to a German seller through Amazon.fr less than two months ago: Its warranty expires in a few days... :(

From my experience dealing with them, Seagate's default warranty info is based on drive's date of manufacture. So it is possible that this drive had sat in some Amazon warehouse for 5 years before you got it.. You can register the drive with receipt to get that data corrected on Seagate's side.

Re: Seagate: 'new' hard drives used for tens of thousands of hours

#197
post #61

You can read out the FARM logs of Seagate hard drives using smartctl -l farm /dev/sd They're supposed to be "more trustworthy" than the regular SMART stats. (My two "new" 16TB Exos drives had 0 hours (regular) and ~18k hours (farm) - DOM 04JUN2021 and 07JUN2021. Also, zfs refused to format the drive: 'already formatted as ddf_raid_member'.)

Anyone know how to check these drives on a Mac? I have a new (?) 8TB Seagate drive out in front of my house right now, just delivered.

If you have an ARM Mac and a USB enclosure for the drive, I had some success with smartmontools running in ARM Linux under VmWare. I also tried UTM (pretty much QEMU), but UTM's USB passthrough was not good enough.

Re: Seagate: 'new' hard drives used for tens of thousands of hours

#199

Earlier quoted context omitted.

The HGST Ultrastars, which were bought by WD and so now are WD Ultrastars, have a great reputation and despite being marketted as enterprise drives, are not much more expensive than consumer drives. My data point is that I've bought several of them and never had any trouble.

>My data point is that I've bought several of them and never had any trouble. This is, emphatically, not a good data point. You need thousands of drives running for years under normal loads to actually tease out outlier failure rates in a drive SKU. You need to literally do what Backblaze does.

I know, that's why I didn't say "you should buy these drives because I've never had any failures". I was just adding some context.

Re: Seagate: 'new' hard drives used for tens of thousands of hours

#200

If you're looking for cheap bulk storage, refurbished/used is a great way to go. It'd just be nice if they told you that's what you were getting :) Shout out to https://diskprices.com/

Some of the ebay sellers are super honest with their product they are moving too. I saw a photo of a wd red, it looked like someone tried to destroy it with a blowtorch. But they could have easily lied and put up the wd stock product photo.

Yeah esp with enterprise gear. Consumer stuff on eBay is more hit and miss
Post reply on HN