Stuck choosing between 5400 and 7200 RPM drives!

Longevity = Warranty and do not plan on the drive lasting a day longer. Hopefully it will not fail before then but that is always a possibility, and it does happen, not often thankfully. Also find out if WD will do an advance RMA to the Republic of Georgia. I live in Georgia, USA, kind of close :slight_smile:

I think you have already made up your mind… two 12TB drives and use the old 8TB drives as well. It will also be easier to migrate your data over form the current drives. Much simpler.

Question: Is there anything wrong with your drives right now? What do the SMART tests show (I know you are doing them routinely). If you are unsure what you are reading, just post the SMART data and we can give you an analysis. At least let you know if one of the drives is on the way out the door or if they both look great right now.


no issues in operation so far, not sure about what smart say though…

yeah, RMA process from my place is a huge pain in the… :d

lol give me your street address I’ll pick you up so we can go drink some beer :slight_smile:

1 Like

The data above all looks good as you said. It does not state anything about SMART Long Self-test passing or not. If you are not doing so, I recommend a weekly Long test and a daily Short test. Right now, your drives will likely last a few more years. I like WD drives in general and the only complaint I had was them sneaking in SMR in the RED lineup. Jerks.

Well time for dinner in this time zone.

1 Like

Ended up buying 2x 12TB drives (WD120EFBX)

Fingers crossed. I’ve seen mixed pos/neg reviews, mostly pos, but will see…

Make sure to burn them in prior to deployment - that way you mitigate the chance that you’ll be part of the negative reviews :slight_smile:

Meaning? I don’t want to burn my brand new, shiny hdds when they arrive :ddd

An old guide, but still relevant. Burning your brand new shiny hdds is 100% what you always want to do unless putting them into production asap is more critical than making sure they are actually tested reliable instead of just being new :slight_smile:

Summary of the guide; smart tests to make sure it isn’t DOA, then badblocks to write & read several FULL passes of the HDD to make sure that every sector is actually good prior to writing any of your important data on it.

Expect burn-in to take a few DAYS on 12tb drives. You could skip it, but it sucks finding out 8 months later that your shinny new drives have bad sectors & need to be replaced asap, your data is now at risk since you only have 2 drives on the pool, & now you have to deal with manufacturer warranty instead of retailer & have to pay for shipping, waiting on said back & forth shipping, & hoping they don’t send you another dud.

CAUTION WARNING CAUTION; make sure to triple check you’re executing badblocks on the correct drive if you have any with data you want to keep on the machine you’re running it on.

Got it! Thanks a lot! I’ll definitely test them once they arrive! But have to say testing for few days is a pain.

Agreed - there are a few tuneables with badblocks to cut down a few hours. My old notes have the following (someone more knowledgeable can correct what I wrote a few years back)

Following sets –c for multiple blocks tested at once (it won’t go past what drives allows but may use more ram, gives some speed benefits over default). –b for blocksize of the disk (4096 >2TB, don’t use for for <2TB, high volume at like 20tb could be higher)
• badblocks -c 2048 -b 4096 -wvs /dev/diskname

To be honest, I have never tested new drives deeply like you describe. Only SMART test prior installation. Funny I can somewhat compare this with how I’m testing digital camera sensors for bad/dead pixels :slight_smile: for example my K1 camera had a few dead pixels right after unboxing, brand new, and it’s considered as normal as there are millions them, and manufacturers never guarantee that there will be no such cases. They even implemented a special feature in the menu where you can disable/repair these dead pixels.

Having said that, I’ll be a bit disappointed if I find bad blocks in my brand new shiny pretty red drives.

Agreed, but also important. And yes, it’s possible to speed up badblocks a bit, but the time in testing also, hopefully, pushes you further into the “bathtub curve,” such that if a drive’s going to fail early, it’s going to happen during testing rather than once it’s in production.

I have 6 x 18 TB drives on the way. I’m not looking forward to a week or more of badblocks testing before I put them into production, but if that helps me weed out a marginal drive, it’s more than worth it.

1 Like

ogh man, backing up those drives must be a real challenge. I’m curious if there is some affordable tape backup drives, because last time I was checking they were so expensive.

More spinners would be much cheaper than tape. I like the idea of tape, but I’ve never found a way to make it economically feasible.

1 Like

I would recommend using my script to run the burn in. Not only does it take care of such things it also works around some gotchas that come with using badblocks with larger drives then it was designed for (12tb qualify). Do the burnin run before adding the disks to any pool and following the directions and yes, definitely be sure you have the right drives before pressing go.

3 Likes

Thanks, I’ll do that, but what do you mean by saying right drives? Not all drives are compatible with this script?

If you specify a drive that is part of your array (ie the wrong drive) it will corrupt it.

Ergo, double check which drive you are testing and make sure you use “the right drive”

4 Likes

ok got it, thanks.

No it doesn’t. This is the infamous “5400 rpm-class”.

1 Like

Indeed, there’s been articles written about it.

1 Like