HGST Ultrastar He8 Helium (HUH728080ALE601) 8TB 7200RPM 128MB Cache SATA 6.0Gb/s 3.5in Enterprise Hard Drive - 5 Year Warranty (Renewed)

[Oldschool297]

Is these disks something to recommend? 68$. I rather have WD but enterprise disks can be good as slaves when they stand more.

NOTE -These HDD is used by Datacenter Servers for about 3 Years period. HDD was refurbished and data wiped with DoD standard. It's fully tested & passed HGST factory diaGnose software test. Since this is heavy duty enterprise HDD with a 2.5M-hour MTBF ratinG.
8TB, 128MB Cache, 7200RPM, SATA III 6.0Gb/s - DesiGned for 24/7/365 Heavy Duty, 2.5M-hour MTBF
Works for Any SATA Server, NAS (network storaGe), RAID, PC, Mac, CCTV DVR, Surveillance System, CCTV DVR
Bare Drive Only, SinGle Pack, (No Screws, Cables or Accessories included) -Friendly Reminder- Please FORMAT HDD on system in order to be detected, shows on system.

 

[Solaris17]

Is these disks something to recommend?

Personal preference. I dont use certain things that are used or refurb. The list is pretty short.

- Peripherals
- Storage

For the use case of most users on TPU (5400rpm NAS queens that download movies) a refurb drive might be ok.

The ones you pointed our are DC drives, they will be loud and vibrate a lot. I also wouldn't put anything on it that's important. I also wouldn't trust the specs unless bought from anyone other than WD/HGST because I guarantee if it was opened they didn't put helium back in it. Which is part of the drive spec for heads and platters that close.

 

[eidairaman1]

Dude, no used hdds

 

[Oldschool297]

I have been lucky with used disks as slaves, some has work from 6 to 10 years 24h, and those was usual. Do these DC drives sound similiar like disks from 1999-2002?

 

[Shrek]

Might be fine for a game library where one just loses download time if it fails.

 

[Kursah]

I've been running six refurbished HGST 3TB HDD's in my spare budget-oriented homelab server RAID for about 4-5 years solid now...wow, shit, I hadn't thought about it having been that long already. No issues at all, surprisingly and I expected problems. But I chose HGST refurbs over white label hoping that getting enterprise grade HDD's, even if refurbed, would give me better mileage for the investment, so far, so good. Would I use them in a production server or client application? Nope.

I purchased 2 spares just in case. Converted an old Lenovo TS430 I snagged from work way back when with both 4-bay drive-bays, backplanes, etc. Replaced the mainboard, CPU, RAM, PSU. Added a TS440 front panel with USB3. Soldered some leads and clips for standard ATX front panel on the FP board for power, reset, NIC 1 activity, HDD activity.

But I have replication and backups as well, because whether its used or new, disk failure is a when, not an if.

Buying refurb'd or white label storage is a risk for sure. That's akin to buying used brake pads and rotors for your car. Sure, they might last tens of thousands of miles, or it might be backplate embedded in rotor out of the chute.

The only other application I've utilized used drives in is additional storage for my workstation and my kid's workstations so they have a catch-all for their OBS caps for their respective gaming YT channels and meme creation efforts. Things that if lost, won't be the end of the world.

Might be fine for a game library where one just loses download time if it fails.

Exactly right and with what others said, use it as low-priority storage that you can afford to lose or be without should the drive give up the ghost unexpectedly.

If you have important data and are not backing up, you might want to work on that first.

 

[Assimilator]

Never, ever, ever buy second-hand storage.

 

[chrcoluk]

HGST have a very good rep.

I think refurb from the actual manufacturer is fine, as usually they provide a warranty still and at least its a refurb done by the actual company that makes the drives.

 

[Shrek]

Never, ever, ever buy second-hand storage.

I just got a 2TB second hand SATA SSD for just over $40 with postage and it is now my boot drive; mind you I did start running an automated backup onto the hard drive it replaced (and that is on top of the hand backup I make to a thumb drive)

I'd say never ever trust storage, new or second hand.

At one point I had my backups running on a mirrored drive, but I got lazy and now just have an automated backup at work, at home and on a thumb drive.

Even a straight backup is not enough and one of them is time-machine on a Mac, in case any files get corrupted.

 

[eidairaman1]

I have been lucky with used disks as slaves, some has work from 6 to 10 years 24h, and those was usual. Do these DC drives sound similiar like disks from 1999-2002?

The larger they are the more to lose.

 

[Dr. Dro]

I'll pass on used drives, but those are good stuff. Buy a new WD Gold from Amazon if you want one like it, they are modern Ultrastar drives with better specs than that old one

 

[unwind-protect]

Good drives.

 

[Oldschool297]

ese DC drives sound similiar like disks from 1999-2002?

It's maybe better to buy a Seagate? https://www.amazon.com/Seagate-Barr...ad-9975-4f6234c0010e&pd_rd_i=B07H289S7C&psc=1

Or another https://www.amazon.com/HGST-Ultrast.../B07MK3T56C/ref=psdc_1254762011_t2_B09VBBKKMB

 

[Shou Miko]

The larger they are the more to lose.

This is true, I fear the day when my 16TB Exos says goodbye but I hope by that time I have a backup and dedicated server så it won't be that bad.

 

[Oldschool297]

Is that's why you should have a smaller drive as master? Or SSD.

 

[Oldschool297]

As they give 5 years warranty on a 5 years used server disk they must be reliable? It is the thing with helium but i guess it can stay in the disks for 10years?

 

[Selaya]

Never, ever, ever buy second-hand storage.

honestly depends

buying used qlc or seagate hdds, very bad idea
but something like an enterprise hgst drive or an 860 pro, yup sign me up - those are virtually indestructible

 

[JohnnyQuest1111]

I have been purchasing these drives for the past few years. When I receive them, I ALWAYS test them using the SMART short and long tests. The long test takes 8 to 10 hours. In my experience, its hit or miss. Sometimes I've received a DOA. Sometimes they pass the SMART short test but not the SMART long test. Sometimes they pass both tests and that's when I keep them. I've seen power-on hours from 25K to 55K with lower power-cycle and head load counts. When the drive fails my tests, the seller always replaces them. Sometimes I still get drives that won't pass the SMART tests. Sometimes I've been blacklisted by the eBay vendor when I return a drive. Like its my fault they sent a bad drive, eh?

Although I comprehend the comments about not buying used hard drives, I still am and they are being used in my home server. These HGST He8 drives have a 2.5M hour MTBF and with the 5 year warranty, I don't see a problem. I think the key to using these re-certified drives is to use them in RAID arrays, which is what the data centers do. So, RAID1, RAID5 or (preferably) RAID6 arrays. If one fails, you replace it, maybe under warranty, and let your RAID controller rebuild the array with the replacement drive.

My server is a DELL PowerEdge R510 with 8 bays and a PERC H700 hardware RAID controller running two virtual drives. One VD as a (3TB) RAID1 array, which contains the OS and some temporary storage. The other VD is a 6 drive (8TB) RAID6 (24TB total) with a hot-spare. In RAID6, I can have as many as two drives fail at once and still have a functioning array. The rebuild time on an 8TB drive is 8 to 10 hours.

I recently purchased a DELL PowerEdge R410 to be used as a backup for the R510. It has 4 bays and the same PERC H700 hardware RAID controller running one virtual drive as a RAID5 array. In this case, I have the OS partition on the RAID5 virtual drive. No hot spares and I can take a single failed drive and still have a functioning array. If I purchase some spares and not use them then the 5 year warranty clock is going to keep ticking anyhow. I'm hoping these drives will still be available or perhaps the 10TB will retire by then.

I saw a comment about the helium leaking out. I've seen no published data, so I'm not sure how much time that would take. But if you get 5+ years out of them then what's the problem?

I know its a calculated risk but some of these drives have been running 3 years with no problem.

Note: I currently have 3 "new" HGST 8TB He8 drives that came in. All were replaced under warranty because they didn't pass the SMART long test. I've been using the smartctl tool under Linux to run the test. I'm thinking about performing a firmware update and I apparently need to use the WD HUGO software to do so. The HUGO program is written by HGST and can initiate the short and long tests. These 3 drives won't complete the SMART long test as they fail shortly into the 90% remaining mark. They also will not log the test failure in the log. However, as I write this post, all three drives are currently running the long test initiated by the HUGO program. It's been well into 2 hours since I started the tests with HUGO and they are at 80% remaining. So why do they consistently fail the SMART long test with smartctl and not with HUGO?

I welcome any comments on my thinking with these drives and especially on the SMART/HUGO long test differences.

Peace and blessings,
JQ