8TB WD Red Plus 3.5" 7200 RPM NAS Hard Disk Drive Expired

tl;dr version - most "enterprise" drives are equivalent to the WD Red drives, and just fine for use in a NAS. The WD Red drives are not "better". Just make sure they're conventional (CMR), not shingled (SMR).

long version:
Calling the WD Red drives "NAS" drives is a bit of a misnomer. They're just regular drives. WD made them "NAS" by deleting from all their other drives a feature which used to be important for most NAS applications.

When files are written to a drive, a checksum stored with it. This is a number generated by an algorithm which results in the checksum changing if the file changes slightly. Whenever a drive reads a file, it calculates this checksum, and compares it against the checksum which was stored when the file was originally written. If the checksums don't match, it knows it encountered a read error and the copy of the file it just read is not the same as what was originally written. (Bits are not written as 1s and 0s on a HDD - they're written as a waveform of magnetic field strength, with a ton of error correction coding. The read and write process is analog, so what you get is a little different each time you read it.)

The drive retries reading several times trying to get the correct checksum. If it fails to get a clean read within a certain amount of time, you get the "Abort, Retry, Ignore?" message you may have occasionally seen. On a desktop drive, this error isn't a problem. The computer just stops everything it's doing, until you tell it what to do about this drive read error.

Drives in a NAS are usually configured as part of a RAID array - the information for each file is spread across multiple drives. And crucially, these drives are hidden behind the RAID software so that they look like a single drive to the computer, and with redundancy so a read error doesn't slow it down. If there's a read error, it can't stop to ask you what to do about it.

Worse yet, a read error used to be interpreted by RAID software as the drive failing. It will drop the drive from the array, mark it as bad and needing replacing, and continue to operate in degraded mode (without the "failed" drive). With RAID 1 or RAID 5 (single-drive redundancy), if a second drive fails while in degraded mode, all your data on the RAID array is gone.

This is obviously a very bad thing to have happen. So it should only happen if the RAID array is absolutely sure the drive is actually bad. The number of read retries before an "Abort, Retry, Ignore?" query was judged too short for RAID. So drives were given a special command to increase the number of read retries before declaring the drive as bad. If you used the drive in a desktop, it would give up trying to re-read a file after a few tries. If you used it in a RAID array, the RAID software would toggle this command, and the drive would try longer to make sure the drive had really failed, before marking it bad. Each manufacturer had a different name for this command, but WD called it TLER.

Here's the thing. It used to be that every WD drive had the TLER command. But one day WD decided to charge users extra money for drives used in RAID arrays (since they're used primarily by businesses). So they deleted TLER from all their drives, and introduced the Red series of drives which were priced higher and still had the TLER command. From then on, if you wanted to use WD drives in a RAID array, you had to use the Red drives or risk a few single-bit read errors destroying all the data in your RAID array.

No other drive manufacturer does this. You can use any Seagate or Toshiba drive in a RAID array and still have access to this command. It's only WD which crippled most of their drive lineup to prevent you from using cheaper drives in RAID arrays. The whole point of RAID is to safely store data on cheap drives, relying on parity and redundancy to protect you from drive failures..(Initially RAID used to mean Redundant Array of Inexpensive Disks, although nowadays most definitions substitute Independent for 'I', which is kinda silly since a drive is by definition independent, and a RAID drive doesn't necessarily have to be a single independent drive.) And WD came up with this Red scheme as a way to prevent people from using cheap drives in their RAID arrays.

Furthermore, most RAID software has grown beyond relying on TLER or similar commands. They no longer immediately mark a drive as failed the moment it generates a read error. They will track such errors (they can get the correct read from parity), and use their own internal algorithm to decide if a drive has really failed before marking it as bad. So in most cases nowadays, you don't even need the TLER command. You can use any old drive (except shingled) without issue. If you're worried about the drive being powered on 24/7, you might want to buy enterprise drives instead of consumer drives. But honestly, being left powered on 24/7 is actually less stressful for electronics than being turned off and back on several times a day.

Why do I keep bringing up shingled drives?

Because shingled drives (SMR) are death for NAS performance. A SMR drive overlaps write tracks. The drive's read head is much smaller than the write head, so you can move the write head a fraction of its width, call that a different track, and the read head can still read the small non-overlapped portion just fine. This works fine for reading, but if any data needs to be erased or re-written, the entire block of overlapping tracks needs to be erased and the unchanged data re-written. This obviously takes a lot of time, and will kill performance on an array of drives like in a NAS.

So you don't want SMR drives in a NAS. Unfortunately, WD screwed the pooch here and released several SMR drives as WD Red drives (even though they're totally inappropriate for RAID), and didn't disclose to buyers that they're SMR. They've apologized and now disclose which drives are SMR. But they still sell SMR drives in their Red lineup. So buying a Red drive doesn't guarantee you good performance in a NAS, which was kinda the whole point of the Red branding.
https://www.tomshardware.com/news...techNOLOGY

Toshiba and Seagate always disclosed if their enterprise drives were SMR. So you're reasonably safe buying those (still need to check, but the info is out there). But they didn't always reveal if their consumer drives were SMR. So you have to be careful buying those if you plan to stick them in a NAS. Here's Toshiba's list of SMR drives:
https://toshiba.semicon-storage.c...428-1.html

Most NAS manufacturers have responded by putting out lists of "approved" drives for use in their NASes. But these are inclusive lists (drives known to work), not exclusive (drives you should avoid). So drives not on the list may still work.
https://www.synology.com/en-globa...ge_log_p=1

But it's $132.

Even $155 for WD Red Plus is a good deal. they're $200+ most places. These are NAS drives, not regular bulk storage. To be clear these are not SMR drives (which is good, this edit is in response to the very long and very informative comment about shingled drives - which this drive is not)

never been lower on The CamelSite nor anywhere else I can see. PCPP tracker also has never seen them this low. *shrug*