garycase

I do similar things with my new drives. I run WD's Data Lifeguard and do a quick test; then an extended test; then write zeroes to the full drive; then repeat the quick & extended tests. If it's a drive I don't need right away, I then stick it on my "testing box" [A spare old PC I use ONLY for testing drive and data recovery] and do a full Level 4 run of Spinrite on the drive (this takes a LONG time). I've actually never had a drive that passed the Data Lifeguard tests fail the Spinrite run, but a Level 4 run of Spinrite is about as intense as you can get -- multiple data patterns; every block re-written multiple times; and if it did find any read errors it would do a statistical analysis of the errors and attempt to fix the block before doing a reallocation. Spinrite, by the way, DOES track the drive temps and will both warn you and stop the test if the temps get out of line.

If the drive tests okay, and reformats okay, then it's a good drive to use for some of your backups. Generally you only write data to a backup disk once; then "tuck it away" for use in an emergency .... so it doesn't get nearly the use of an operational drive in the system. Several of my backup drives are either RMA replacements or drives that I just elected to replace because I didn't like the reallocated sector counts.

Clearly you don't have a habit of over-engineering things. That's might be enough in my book! Might need to give some thought to floor reinforcements... I absolutely over-engineer the daylights out of things. ... but 1,000 pounds is clearly enough [My wife wanted some shelves in the laundry a few years ago -- I built the shelves out of 2 x 12's with 2 x 12 supports. My wife and I could have climbed up and jumped on them and they wouldn't have budged We've long since sold that house ... but everything I build is engineered along those same lines.

By the way, depending on just how high the casters need to be to keep the radio off the ground, those casters come in 3", 4" and 5" sizes. The larger sizes also have higher weight limits -- so if you're concerned about weight, you may want to use the 4" ones. The 3" have a 110 lb/caster limit -- so with 4 casters you'd be able to manage 440 pounds (I'd think that's PLENTY, but weight can indeed add up. The 4" have a 250 lb/caster limit -- so 4 would hold 1,000 pounds !! [absolutely enough ] If you're paranoid (or just need the extra height), the 5" units have a 350 lb rating -- or 1,400 pounds with 4 of them !!! They also all come both with or without locks -- you may not want the locks ... or may only want them on a couple of the wheels.

I suppose it would slightly defeat the "as original as possible" look ... but you could always add some non-marking casters [These work very well -- I've installed them on quite a few pieces of furniture around the house: http://www.homedepot.com/p/Everbilt-3-in-Swivel-with-Brake-Non-Marking-Rubber-Caster-4036345EB/203672428 If you don't want to add them because of the screw holes they would add, these guys also work very well: For carpet: http://www.homedepot.com/p/Shepherd-4-in-Round-Slide-Glides-Non-Adhesive-4-Pack-9316/202531725 For hardwood: http://www.homedepot.com/p/Everbilt-3-1-2-in-Felt-Bottom-Round-Slider-4-Pack-4713444EB/203672360?MERCH=RV-_-RV_nav_plp_rr-2-_-NA-_-203672360-_-N

Yes, I was looking at your last parity check date => from your comment above about it "being alive" and the parity check temps, I thought you had just run that.

You might want to set the time and date

This is REALLY starting to look good !! I wouldn't bother with a Reset button ... just a power button is all you really need. Personally, I also wouldn't connect the hard drive LED output to anything ... those flashing LED's can be a bit annoying. Definitely looking forward to pictures of the finished unit

Definitely overkill => and in addition it will be running well below the 80+ certification levels, so will be relatively inefficient. With only 5 drives I'd probably buy a 450w unit, which would easily support 3-5 more drives. If you want some "future proofing" I'd go with about 650w.

Cache_Dirs will indeed do a LOT of reading when you first boot the server; but will normally finish after all the directories are cached. In you have a LOT of files on disk1; and if the pre-clear activity is using enough of your memory that all of the directories can't be cached; then Cache_Dirs could be constantly attempting to update the cached information, and never actually finish, since everything isn't fitting in memory.

Constant reads? ... or is it just spinning? On some controllers, if one drive is active, other drives on the controller will also be active => so if the drive you're pre-clearing and disk1 are both on a controller with these properties, that would explain it. But it shouldn't be causing actual read activity from the drive. Do you have any plugins that may be using the disk?

I'm sure that's exactly what they do. With fault-tolerant systems it's not all that big a deal ... if a drive fails, whether it's new or not, you just replace it. But I, like you and many others here, prefer to identify those drives that have infant mortality issues BEFORE I actually use them => and my experience mirrors yours ... I've had VERY few drives actually fail in use if they pass my rigorous initial testing; but have had quite a few drives (although not a high %) that didn't make it through the initial tests I subject them to.

The power supply isn't an issue -- just use an SFX power supply. The Silverstone units are excellent choices (either the 300 or the 450w unit).

Yes, that's easy to overlook. This is also a feature of the Gold & Platinum MasterCard and Visa cards. It's been around for a LONG time ... at least 25 years, and perhaps longer. Originally it was "double your warranty" ... then it was restricted so it can't add more than a year [i.e. a 3 year warranty would have originally been doubled to 6 years; then after the change it was covered up to 4 years]. Finally, they all added a maximum total warranty time of 5 years, which is what most of them now provide. ... but for hard drives that works fine ==> since most have warranties between 1 & 3 years, you simply get one additional year. If the drive fails in the actual warranty period, you get a free replacement from the manufacturer; if it's during the year after the warranty expired, you get your money back from your credit card One important caveat: The credit card extended warranties work well (I've used them several times), BUT they have very precise filing procedures for claims, so be sure you keep (a) your receipt for the drive; (b) the credit card statement that shows that specific charge; and © a copy of the manufacturer's warranty in effect at the time you bought the drive. As long as you have those, it'll be very simply to get reimbursed if you should have a problem.

Absolutely agree. And in addition to that you get a drive that uses less power; runs cooler; and is optimized for vibrational control. Definitely worth a few extra $$ in my opinion. Ditto. I certainly agree I'll never replace my older, smaller drives when notably larger sizes are available; but it's certainly reasonable to expect you won't need to do this every year. 3 years is a good interval for this. One other thing I've noted: It is VERY likely that if you have a drive fail after a couple years and RMA it for replacement that you will receive a LARGER drive in return for it. Both WD and Seagate do this routinely; and I assume others do as well. So a longer warranty not only means you won't have to buy a new drive for that slot for longer; but it may also mean that if it DOES fail you'll get a larger drive without the need to buy one

Close cousins for sure ... but NOT twins. WD did make some improvements in vibration control and tolerance; and modified the firmware to implement NAS-specific optimizations. In addition, the Reds use less power than the Greens. I think there's no doubt that the Reds are worth the modest price difference for the nominal power savings; the NAS-related optimization and vibration control technologies; and the longer warranty.

I REALLY doubt that. WD isn't a fly-by-night company, and I just don't think they'd resort to that level of deceptive marketing. I think it's much more likely that the statistics are based on drives from early in the 3TB Red production cycle; and it's likely that things have improved a good bit since then. It'll be interesting if NEXT year's stats show this to be the case.

Not only is it not a bad things, but any UnRAID'er should have at least one, if not two, drives on the shelf as spares for the inevitable day when a drive fails. If the spares "age" to the point where they're not large enough anymore (due to increased drive sizes), just use them as backups Indeed, there are often some really good buys during those few weeks. I'm definitely watching the pricing for 6TB Reds -- may just have to buy a few if they drop under $240

I really don't think that's likely -- the 3TB WD Reds have 3 platters; the 4TB have 4 platters. I very much doubt there's any quality difference in the platters themselves, or, for that matter, in the electronics, actuators, head assemblies, etc. It MAY be true that this failure data is based on relatively early units in the production cycle -- since BackBlaze had them long enough to get statistically valid data. If that's the case, then later 3TB units, and the 4TB units (which came out later, after I'd assume production improvements were already in place) should be more reliable ... but we'll need NEXT year's report to see if that's the case.

I'm also a bit surprised at the relatively high failure rate for the Reds. I've certainly not seen that with the ones I've purchased -- both for myself and others. They've actually been exceptionally reliable ... I hesitate to jinx myself, but I'll say it anyway: Out of perhaps 40 drives in 2 years, I've seen ZERO failures of those in service. I HAVE had a couple that failed my initial testing that I returned for replacement, but none of the drives that passed my initial testing [WD Lifeguard short test; extended test; write zeros to the whole drive; then repeat the two tests] have had any issues since they've been put in service. And those that I had to have replaced may have been damaged by poor packaging during shipment, as I've not that issue in over a year (since Newegg started packing their drives much better). Nevertheless, I just ordered a 3TB HGST to add a bit of storage to my wife's computer (This is the first non-WD Red drive I've bought in a while excluding SSDs)

Good info -- looks like I need to start buying HGST drives

Agree that high-fly writes should be considered an "experimental" parameter. WD doesn't bother to report these, and I've seen numbers all over the charts on Seagate drives ... with NO issues as a result. I suspect they can simply be ignored altogether, but it won't hurt to monitor them to see if the behavior starts to change over time.

SNAP works well, especially if the amount of data you need to backup is no more than the capacity of the external drive you're mounting. With modern high-capacity drives, that can be up to 6TB, so it's likely not a problem with music collections (but easily could be with video/movie collections).

Those 6" slides are a nice find -- I've looked for very short slides for a couple of projects ... these will work nicely. Don't need any at the moment, but definitely bookmarked the page !

... by the way, WeeboTech's utility to catalog all the files, generate checksums, verify them, etc. sounds like a VERY nice addition !! Looking forward to when this is completed, primarily because running verifications natively on the UnRAID server will be appreciably faster than across the network.