garycase

There's no reason you couldn't do that; but why? Turbo-write is generally used to get the quickest possible writes directly to the array, so your data is immediately fault-tolerant and doesn't have to go through the cache. Since you're caching all your writes, then the amount of time it takes to do the actual writes to the array is "hidden" from you anyway, since from the user's perspective that data has already been written to the array. In most cases, those writes are also automatic ... that's the whole idea of the mover script. ... nevertheless, you can simply turn on turbo-write; initiate the mover; and then later just turn off turbo-write.

I'm surprised you're having heat issues with the CPU's in particular => dual fans on those Noctua heatsinks should provide PLENTY of cooling for your CPU's. According to Noctua's dissipation chart, the heatsinks you have, when used with a 2011 mount, should be able to dissipate 140w of heat from the CPUs ... and since your CPU's only have an 85W TDP that should be PLENTY -- I'm surprised you can get them to thermally throttle unless you're either significantly over-clocking them (seems unlikely); or there's an issue with the thermal compound not providing a good bond between the CPU and heatsink. I'd remove the heatsinks; thoroughly clean both the CPU and the heatsink [use Artic's ArctiClean kit: https://www.amazon.com/ArctiClean-Thermal-Compound-Remover-Purifier/dp/B001JYQ9TM/ref=pd_lpo_vtph_23_tr_t_2?_encoding=UTF8&psc=1&refRID=KTYAKS51HYQCT94RK4BH ]; and then reapply some good thermal compound [e.g. Arctic Silver] and remount the heatsinks. How are your drive temps? Three 120rpm fans should provide adequate cooling for the drive cages IF the fans have adequate static pressure and move a reasonable amount of air. There's a big big difference between a low-noise fan that's only moving 40CFM, a moderate speed fan that moves 80+ CFM; and a server-grade high-rpm (= noisy) fan that can move 120 - 150 CFM (e.g. Delta fans that run at 2900-3900 rpm). I'd want fans that move at least 80CFM. I can't tell from your picture just how well ventilated the case is -- i.e. are there adequate air intakes and exhausts that your fans can move the air they're designed to move? I suspect this is likely a combination of these factors -- re-applying good thermal compound and increasing the CFM will likely resolve it.

A 650w would be fine, but 550w is really all you need. Even if you bumped up to a 1070i video card a 550w PSU is still sufficient ... and if you changed to a more modern motherboard with an i7-8700 you'd actually be using LESS power than your i7-2600. [The TDP of an i7-8700 is 65w compared to 95w for the i7-2600; and the newer motherboard would also use less power than the older chipset with the 2600]

There are a variety of tools that work well for testing disks. If you have a Windows box with a spare SATA port or an eSATA or USB3 dock you can easily test any new disks on that. I do this using WD's Data Lifeguard utility (works fine with any manufacturer's drives) ... but you can also use the testing utility from the manufacturer of your disk if you prefer. Note that pre-clear can still be used for this purpose as well if you want to use your UnRAID system as your testing box. But as already noted, pre-clear is NOT needed anymore for the primary purpose it was developed (over a decade ago !!) ... which was to eliminate the down time you would have on a server while it was clearing a new disk. v6 of UnRAID clears the disk automatically BEFORE it is added to the protected array; so the array functionality is not impacted at all by adding new disks to it. When Joe L developed pre-clear, he did include a bunch of activity to test the drives; but the primary reason was to have a new drive zeroed before it was added to the array -- and he coordinated with Tom to include a "pre-clear signature", so UnRAID would know this drive could be quickly added. That functionality still works; but is simply not needed any more. Other than Data Lifeguard, you could use Seagate's SeaTools, SamSung's HUtil, HGST's Drive Fitness Test, or any of a variety of 3rd party tools like HDDScan, DiskCheckup, HD Tune, Crystal Disk, etc.

I suspect that most newer controllers don't have the issue that this resolves; but for those of us who still have older systems it's a very useful feature to eliminate any "stutter" in media playback caused by another drive spinning up on the same controller as the one we're streaming from. It's easy enough to avoid that -- either spin all drives up when streaming media; or don't use the array for anything else when you're streaming ... and the issue this resolves won't ever happen. But unless there's a good reason to eliminate the feature, I'd prefer to see it retained.

As aleady noted, since you didn't have any hash values for your files, there's nothing this plugin or any other checksum utlity can do to help identify corrupted files. You're also not likely get any notification of errors when copying files from the disk. For those files you have backups of, the best way to confirm you have good copies is to do a binary comparison between the files on the UnRAID server and your backups. Any files that don't match, you'll need to manually check to see if you can tell which is the good one -- and then replace the other one. Note that since you don't have checksums for either set of files you have no way of confirming computationally which is actually the correct file. Doing the comparison is easy -- just any any good comparison utility that does a binary compare. I use FolderMatch (not free), but there are several free utilities that can do this as well. [ http://www.foldermatch.com/ ] If there are any mismatches, it's clearly a bit more tedious, but hopefully you won't have any (or at least not many) of those. Since your backups are apparently not local, I'd download a bunch at once onto a PC; then do the comparisons for those; then repeat as needed until you've checked everything. Adding a UPS was the most important improvement you made -- I've told folks for decades this is MANDATORY accessory for EVERY PC ... a power-protected PC that never undergoes an unexpected "yank" of the power cord (which is basically what a power failure is without a UPS) has FAR fewer hardware issues than an unprotected PC. And for a server it's even more important. Hopefully you bought a quality UPS unit -- as a minimum it should have AVR, and ideally you want a true sine wave output. One other thing you can do to give you some idea of whether there's definitely corruption or not is to see if parity is good on the system. Do a parity check and see if there are any sync errors detected. Note that many would suggest doing a non-correcting check, but I see no reason to do that. You WANT parity to be good, so even if UnRAID's assumption that the error is in the parity isn't correct, that's really the only place it can be corrected, since you have no way to knowing which other disk it might be on. If there are sync errors, you'll know for sure you had some corruption, and you need to then do the testing outlined above r.e. file comparisons. Of course if you don't find any, then all you'll know is that the corruption was either on the parity disk (that's good, as it will be resolved by the parity check); or it's on a file you don't have backedup -- in which case there's no way to identify that file. Unfortunately, even if there are NO parity errors, that doesn't absolutely mean you didn't have some corruption, as a write sequence may have been completed immediately before the power loss but before all file operations were completed. But if that's the case, there's likely very few files involved.

ALL drives can fail -- whether they're in warranty or not has little to do with that. If your concern is that you don't want to buy another 8TB drive when the next drive fails, then you could buy a couple larger drives now to upgrade parity; and then defer the purchase of larger data drives until a drive actually fails. In fact, doing that would also give you one (or two if you have dual parity) spare 8TB archive drive(s) (the old parity drives) you could use in case of another drive failure. The disadvantage of doing this is, of course, that you would be limited to the maximum current drive size for your new parity drives -- which would also limit the size of any future data drives; whereas if you wait you could likely move to 14TB or 16TB drives. But better to buy a pair of 12TB units now instead of buying ALL new drives.

Most drives last FAR longer than their warranties. Unless you need additional capacity, I don't see any reason to buy new drives just because your current drives are beyond their warranty period. And it doesn't seem you're adding significant capacity -- if you have a single parity system, 5 12TB vs 6 8TB only adds 8TB of capacity; if it's a dual parity system, then the drive switch only adds 4TB. I'd at least wait until you actually have a drive fail before buying all new drives. You also, of course, don't have to replace them all at once ... although I confess I recently did exactly that with my oldest server -- replacing all of the old 1.5 & 2TB drives with 8TB HGST NAS drives when Newegg had a very nice sale on the HGST units ($210). But I needed the extra capacity. FWIW NONE of the drives I replaced were in warranty ... 'nor had they been for several years. And in two of my other servers, nearly every drive is several years past their warranty.

That's definitely not a drive I'd keep in the array with that SMART data. The 8TB Red is a good choice to replace it.

Actually there's an update available which will let the script work with 6.4.1:

A couple of additional comments .... r.e. "... Most people won't have an e-sata dock like you have - e-sata was never a very popular technology ..." => Actually, all you need is a SATA connection and a power cable. You could simply run a SATA cable from any available SATA port out the back of the PC and leave it available for connecting a spare drive when needed -- or you could connect it to an eSATA dock if you wanted to make it more convenient. Ditto a power cable (which you wouldn't need with a dock, since it would provide power). Or you could use a USB3 docking unit and avoid a SATA connection altogether. r.e. "... one could use a hot swap bay like this one ..." => An excellent choice, and with the tray-less unit Frank suggested it's just as easy as plugging a drive in an external bay. r.e. "... Could I pass a new drive through to a VM and test? " ==> Certainly. Just set up a Windows VM, pass a SATA port through to it; and you can do the testing via your VM. Of course since you'll already have the drive installed in your UnRAID server in this case, you could simply use pre-clear -- depends on whether you want the added flexibility of the multiple tests you can choose with the various other testing utilities. r.e. "... f the preclear script can be made to work in unRAID 6.4.1 I really don't see the harm. " ==> I certainly didn't intend to imply otherwise. There's no reason not to use pre-clear; I was simply noting that it's not really needed, since there are plenty of other testing utilities to test a new disk; and the original advantage when it was written of not having your server down while a new disk was being cleared is no longer a factor.

Compatibility with testing software isn't an issue -- any Windows system has a plethora of options (as I listed above); and there are several testing utilities for Macs as well. It's true the system used to test a drive does indeed have to be on for the duration of the test -- whether it's your UnRAID server or another system. As for the advantage of doing the test on your server -- that's a very minor advantage. It takes perhaps 20 seconds to pop a drive in my eSATA dock and start the testing. You do, of course, then have to install the drive in whatever system it's destined for after the test. If it's a drive destined for one of my UnRAID servers I suppose you can argue that's 20 seconds wasted -- but I think most would agree that's not much of a factor .

I have an eSATA port on my desktop connected to one of these docks: https://www.newegg.com/Product/Product.aspx?Item=9SIA1DS28Y2007&cm_re=esata_dock-_-9SIA1DS28Y2007-_-Product You don't really need the docking unit -- just an eSATA cable and power cable would work fine. You can also run it with a USB dock, although it takes a long time with USB v2 (v3 docks work just fine). And yes, Data Lifeguard works fine with every drive I've tried it with -- WD, Seagate, HGST, Maxtor, Toshiba, Fujitsu, Samsung, etc.

Rather than "fiddle" with pre-clear issues, I simply no longer use it. Note that the utility was written for two purposes: (1) to eliminate the long downtime in an array when you added a disk -- which was not useable until it had cleared the new disk (this was arguably Joe's primary motivation for writing it); and (2) to test new disks to be sure there weren't any "infant mortality" issues. #1 is completely obsolete => v6 clears new disks without any impact on array functionality. #2 is easily done with any of the manufacturer's diagnostic utilities. I use WD's Data Lifeguard, but SeaTools, HGST's WinDFT, Toshiba's Storage Diagnostic, or any of the various 3rd party disk testing utilities (HDTune, Crystal Disk, etc.) can do this ... and these have the advantage that you can test your disk on another system without impacting the server. When I get a new disk, I attach it to a spare SATA port on my desktop; then run Data Lifeguard and do a Short Test, Long Test, full Erase, and then repeat the short and long tests. ANY errors and I return the disk -- otherwise I add it to whatever system I bought it for (whether UnRAID, an HTPC, or one of our desktops).

Agree -- that's what I do ... store the drives in WiebeTech DriveBoxes [ https://www.amazon.com/DriveBox-Anti-Static-Storage-3-5-inch-10-pack/dp/B004UALLPE ] and keep them in a fireproof, waterproof, data-rated safe.

It has been a VERY long time since any controller didn't have 48-bit LBA support. But this has nothing to do with the 2TB limit, which is a different issue altogether.

I'm reasonably certain an X7SBE motherboard supports drives > 2TB. As long as that's true, there's no limit on how large those drives can be => 8TB, 10TB, 12TB, or larger will all work just fine. Simplest test is to just attach any drive > 2TB and confirm it's shown correctly in the BIOS. As long as that's true, you don't have any size constraints.

To find the GUID for any flash drive: Just download the USB creator for Mac, and make the USB flash drive you want to use bootable. Boot to it; then click on "Flash" on the main screen. This will show you the GUID for that flash drive. HOWEVER, if you have a flash drive that's already been licensed, then plug that drive into a computer and copy the license key from the flash drive to a "safe" place (a folder on your PC or Mac). This will be a file named Pro.key, Plus.key, etc. THEN you can download the USB creator (for Mac or Windows as needed); make the flash drive a bootable UnRAID flash drive; and then copy the key file back to the Config folder on the flash drive, and you can then boot UnRAID from it. BTW, upgrading to 6.4.1 should not have had any impact on your UnRAID flash key's ability to boot unless you were using a flash drive that you have moved the key from. If that's the case, then you need to use the flash drive you transferred the license to -- NOT the old one you transferred it FROM. That drive is no longer valid, since the license has been transferred.

Yes, you'd see a nice bump in write speeds. First, the Ironwolf drives are 7200rpm, vs 5900rpm for the Archive units. In addition to the higher rotational speed, they also have faster seek times and a higher areal density (1.5TB platters vs 1.33TB for the archives). Each of these provides some improvement ... collectively they should make a very noticeable difference. Since every write requires 4 disk operations (2 reads, 2 writes), the improved seek times and faster rotational speed are the key factors. I'd expect roughly a 20-25% improvement in write speeds with a switch from the archive units to the 12TB Ironwolfs.

Thanks Johnnie -- that's nice to know ... and much better than disabling the monitoring of the attribute. In fact, I had done that quite some time ago for a disk with reallocated sectors, but had forgotten you could do it

CRC errors aren't all that unusual. I have one drive with quite a few (number hasn't changed in years); and one other drive with 1 CRC error that has been there since I added the drive a few months ago (a new 4TB WD Red in one of my older arrays). I don't like the "error indicator" on the Dashboard for these, so I simply went to disk settings and unchecked the "UDMA CRC error rate" attribute so it's not reported ==> easy enough to just look at SMART data periodically to see if anything's changed; and if there were real problems you'd almost certainly see other errors anyway (i.e. an uncorrectable error).

As you've noticed, it's possible that an XFS volume won't be able to hold as much as a Reiser volume of the same size due to file system overhead differences. Assuming you did a COPY (not a MOVE) and all the data is still on disk #1, you can simply delete all of the data from #14 and then copy everything to disk #13 (which clearly won't run out of space since it's a full TB larger than #1. Not sure why the array is sluggish -- a reboot should indeed resolve that.

As I noted, the meaning was clear as long as you thought about it.

The meaning is clear, but the numbers are reversed -- the file system uses about 1GB per TB. [It would be difficult to allocate a TB for each Gb ]

Nice bump in parity check speeds; but it's interesting that you're getting "sub-par" transfers. What kind of transfer speeds are you seeing? What's your disk complement? While I'm not interested I can confirm this is a rock-solid little board that draws very low power (idles under 20w) and although I wouldn't recommend it for a new UnRAID server, if somebody just needs a basic NAS, doesn't need the fastest transfer speeds, and doesn't care about the long parity checks, then it's not a bad choice. FYI the most recent "sold listings" for this board on e-bay were $42.09 and $59.00, so there IS at least a bit of a market for it.