Sorry we lost some posts because of database errors!

*New 12.1 series Release:
2020-09-01: XigmaNAS 12.1.0.4.7728 - released

*New 11.4 series Release:
2020-08-27: XigmaNAS 11.4.0.4.7718 - released!


We really need "Your" help on XigmaNAS https://translations.launchpad.net/xigmanas translations. Please help today!

Producing and hosting XigmaNAS costs money. Please consider donating for our project so that we can continue to offer you the best.
We need your support! eg: PAYPAL

Warning: stay away from the latest HGST Ultrastar He drives (e.g. HC520 or He12)

Hard disks, HDD, RAID Hardware, disk controllers, SATA, PATA, SCSI, IDE, On Board, USB, Firewire, CF (Compact Flash)
Forum rules
Set-Up GuideFAQsForum Rules
Post Reply
m_seitz
Starter
Starter
Posts: 28
Joined: 01 Mar 2015 15:36
Status: Offline

Warning: stay away from the latest HGST Ultrastar He drives (e.g. HC520 or He12)

#1

Post by m_seitz »

Are you sensitive to unanticipated noises from your car, your hard-drives, ...?
Are you afraid of ransomware that encrypts all your data?
Do you have your NAS in your living room, sleeping room, ...., anywhere hard-drive noise can be distracting?

Then: Never even think about getting one of these drives!


I had 4 TB HGST drives before, and they were loud too. But my old drives were only noisy when they actually had to read or write data.
These new He drives come with an "awesome feature" called "Background Media Scan (BMS)" or something like that. As soon as a drive is idle, it starts scanning the surface. All of it. If/when the drive get interrupted, it continues where it left off as soon as it idles again. Once a drive finished a scan, it will wait for a while, and start a new scan eventually.

At first, I thought XigmaNAS was doing something in the background. A lot of googling later, I found a few complaints and some information from other people who experienced the same constant disk activity. I disconnected the SATA cables, powered on the NAS and those He12 drives greeted me with the same annoying chattering as before!

My NAS sounds as if there was some read/write activity ALL THE TIME. And it is driving me crazy :evil: :evil: :evil: :evil: :evil:
WD things it is a good idea to scrub their drives whenever they feel like it and we customers do not have a say in the matter. Apparently, there is no way to switch off the scrubbing. Sometimes I want to kick that crappy NAS of mine to hell :(

Btw., the drive model is "HGST HUH721212ALN604". From what I read on the interwebs, other HGST and WD drives have the same issue.
My Nas
MB: Asus M5A78L-M/USB3 RAM: 32GB unbuffered ECC (4x Kingston KVR16E11/8)
CPU: AMD Phenom II X2 550 Storage: 6x 12TB HGST (crap, don't buy!) HUH721212ALN604 (raidZ2, one single pool)
PSU: Cooler Master V450S (450W) UPS: CyberPower CP1300EPFCLCD (USB via usbhid-ups)
OS: XigmaNAS 12.1 x64-embedded

jlagermann
Starter
Starter
Posts: 15
Joined: 24 Dec 2018 02:15
Location: Just southwest of St. Louis, MO
Contact:
Status: Offline

Re: Warning: stay away from the latest HGST Ultrastar He drives (e.g. HC520 or He12)

#2

Post by jlagermann »

I've been looking at the exact same HDD (now WD UltraStar) and have been doing a lot of research on them. More importantly, trying to decide if I really need to add more storage yet. Back in February I posted a writeup about APM, AAM, and S.M.A.R.T. Configurations https://www.xigmanas.com/forums/viewtop ... 70&t=15082 and I talked about the Advanced Power Management feature settings. All WD SATA drives support APM, some you have to enable it with a jumper, others, like the Ultrastar drives, it's enabled in software. By default it's disabled, which means the drive will always run at full RPM's, even if there is no IO activity.

Below I've included some links directly to resources on Western Digital's site. The product manual list the general features on page 17, describes the different operating modes (active, idle_0, idle_A, idle_B, etc.) on page 24, and describes the Advanced Power Management feature set on page 67.

If you really want to spin down the drives to make them quieter when they are not in use, you would have to set APM to 'level 64' or 'level 1'. Just keep in mind, spinning the drive up and down too frequently will shorten the life of the drive. If the drives are part of a zpool, it does not take very much traffic to keep them spinning, especially with RAID-Z2 or RAID-Z3.

Link to the product manual for the Ultrastar DC HC520 https://documents.westerndigital.com/co ... m-spec.pdf
Link to the Western Digital Ultrastar DC HC520 Datasheet https://documents.westerndigital.com/co ... -hc520.pdf
Link to the part number selector table for Ultrastar DC HC520 https://documents.westerndigital.com/co ... -hc520.pdf
System: XigmaNAS 12.1.0.4.7091 (full version) RootOnZFS Mirror, Supermicro X11DPi-NT, Dual Xeon X3104, 32GB DDR4 ECC RDIMM, LSI SAS 9207-4i4e, CyberPower PR1500 UPS

256GB M.2 (nvd0p1 119G mainPool cache, nvd0p2 119G futurePool cache)
2x 120GB SSD - efi Mirror(p1 200M), freebsd-boot Mirror(p2 512K), Swap Mirror( p3 2G ), zroot Mirror(p4 90G), mainPool SLOG Mirror(p5 10G), futurePool SLOG Mirror(p6 9.6G)
6x WD Red 3TB, 1x WD Red 4TB (RAIDZ2 mainPool & spare)

m_seitz
Starter
Starter
Posts: 28
Joined: 01 Mar 2015 15:36
Status: Offline

Re: Warning: stay away from the latest HGST Ultrastar He drives (e.g. HC520 or He12)

#3

Post by m_seitz »

Thanks for this interesting read.
To answer your question: No, I don't want the drives to spin down :D (for the reasons you mentioned)
However, the undocumented Background Media Scan is so disturbing that I actually switch off the NAS when i don't need it for one or two days :cry:

The manual for the HC520 mentions ADM (Automatic Drive Maintenance) in the glossary, but it is not mentioned anywhere else. I have yet to skim through the whole thing ... 317 pages :geek:
My Nas
MB: Asus M5A78L-M/USB3 RAM: 32GB unbuffered ECC (4x Kingston KVR16E11/8)
CPU: AMD Phenom II X2 550 Storage: 6x 12TB HGST (crap, don't buy!) HUH721212ALN604 (raidZ2, one single pool)
PSU: Cooler Master V450S (450W) UPS: CyberPower CP1300EPFCLCD (USB via usbhid-ups)
OS: XigmaNAS 12.1 x64-embedded

jlagermann
Starter
Starter
Posts: 15
Joined: 24 Dec 2018 02:15
Location: Just southwest of St. Louis, MO
Contact:
Status: Offline

Re: Warning: stay away from the latest HGST Ultrastar He drives (e.g. HC520 or He12)

#4

Post by jlagermann »

I have unsuccessfully searched high an low for something that mentions or explains what might be happening. If you can share a source or article you read I would appreciate it. I don't fully understand how a stand a lone hard drive could scrub itself without the parity from the other drives. :?: It has to be doing something other than scrubbing.
System: XigmaNAS 12.1.0.4.7091 (full version) RootOnZFS Mirror, Supermicro X11DPi-NT, Dual Xeon X3104, 32GB DDR4 ECC RDIMM, LSI SAS 9207-4i4e, CyberPower PR1500 UPS

256GB M.2 (nvd0p1 119G mainPool cache, nvd0p2 119G futurePool cache)
2x 120GB SSD - efi Mirror(p1 200M), freebsd-boot Mirror(p2 512K), Swap Mirror( p3 2G ), zroot Mirror(p4 90G), mainPool SLOG Mirror(p5 10G), futurePool SLOG Mirror(p6 9.6G)
6x WD Red 3TB, 1x WD Red 4TB (RAIDZ2 mainPool & spare)

m_seitz
Starter
Starter
Posts: 28
Joined: 01 Mar 2015 15:36
Status: Offline

Re: Warning: stay away from the latest HGST Ultrastar He drives (e.g. HC520 or He12)

#5

Post by m_seitz »

A hard-disk may verify if it can read data without errors, but it cannot repair or recover destroyed data. I guess that corrupted data will be read a certain number of times. If you are lucky, the data can be read eventually and relocated.

I had a hard time figuring out what my disks are doing. All I could find was evidential information, but no hard facts. Apparently, WD/HGST and Seagate added two "features" to their (consumer grade) hard-disks: Some kind of wear-levelling and the Background Media Scan.

The wear-levelling seems to prevent the heads from hovering over the same spot for too long. You might hear your hard-disks moving their heads every 5 seconds. Mine did this during the first few days and then stopped doing it.
The Background Media Scan is the scheduled scrubbing when idle. Right now, my NAS is dead quiet (as quiet as you would expect without read/write activity but disabled spin-down). I expect the hard-disks to scan their data tomorrow or in a few days.

I might be wrong and several of the six hard-drives I bought might be dead/broken on arrival. This is unlikely but possible. SMART did not report any pending or relocated sectors, and MD5-checking the data copied from my previous hard-disks turned out fine without errors.

All I can give you, jlagermann, are a few links:
https://www.amazon.com/review/R22XBDD7OU4D85
http://web.archive.org/web/201901201814 ... edia-scan/
https://forums.anandtech.com/threads/ba ... n.2525454/
https://forums.unraid.net/topic/88548-h ... in-unraid/
http://www.santools.com/smart/unix/manu ... nction.htm

Edit: One hour later, and the hard-disks started doing their thing ... :cry:
My Nas
MB: Asus M5A78L-M/USB3 RAM: 32GB unbuffered ECC (4x Kingston KVR16E11/8)
CPU: AMD Phenom II X2 550 Storage: 6x 12TB HGST (crap, don't buy!) HUH721212ALN604 (raidZ2, one single pool)
PSU: Cooler Master V450S (450W) UPS: CyberPower CP1300EPFCLCD (USB via usbhid-ups)
OS: XigmaNAS 12.1 x64-embedded

m_seitz
Starter
Starter
Posts: 28
Joined: 01 Mar 2015 15:36
Status: Offline

Re: Warning: stay away from the latest HGST Ultrastar He drives (e.g. HC520 or He12)

#6

Post by m_seitz »

Today, I dismantled the HDD cage and put the HDDs in a straight line to find out which one was making access noises. The conclusion: all of them.

Here is a short video:
https://www.youtube.com/watch?v=5JhRE12VqqM
My Nas
MB: Asus M5A78L-M/USB3 RAM: 32GB unbuffered ECC (4x Kingston KVR16E11/8)
CPU: AMD Phenom II X2 550 Storage: 6x 12TB HGST (crap, don't buy!) HUH721212ALN604 (raidZ2, one single pool)
PSU: Cooler Master V450S (450W) UPS: CyberPower CP1300EPFCLCD (USB via usbhid-ups)
OS: XigmaNAS 12.1 x64-embedded

Post Reply

Return to “Hard disk & controller”