garycase

A Pico PSU is almost certainly enough -- the only real question is spin-up current for 6 drives ... but with such a low power CPU the 160w Pico version should be plenty. On the other hand, the 300w Silverstone SFX unit is a very nice unit -- and although you'd likely be below its 80+ certification level (20% of 300w = 60w, which is likely more than you'll draw), it should still be reasonably efficient. I'd be inclined to use it just for the extra "headroom." But if you want the absolute lowest power draw, use the Pico unit.

Both are superb units -- you won't go wrong choosing either. Personally, I'd buy the Seasonic, but that's just personal preference.

Definitely agree. While none of us like to acknowledge that the value of our nice, shiny new hardware deteriorates like a new car when you drive it out of the showroom, the facts are that it does. Yet it's not at all uncommon for a poster to ask for nearly new prices when advertising gear on the forum. As bjp noted, INFORM yourself - check the new prices; ask for SMART reports on any drives you're considering AND get the serial #'s and do a warranty check if they're advertised as "still in warranty". Also, if you're buying older used drives, compare the cost/TB with a new, higher capacity drive -- which would also have a nice new warranty. There ARE some very good values advertised, and some hard-to-find parts as well -- so DO take a look at what's offered, and don't try to "steal" it ... the sellers DO deserve a fair price. Just be sure that you've compared the cost to what you'd pay for new equipment -- I've seen several cases where you could buy a new system with BETTER specifications than what was offered for the same price.

It's a superb power supply => but WAY overkill for UnRAID ... and in fact would be running well below it's peak efficiency at the low wattages an UnRAID system would likely require.

Good choice. I suspect that the drive's fine with no HPA, but in any event, even if there is, it's an insignificant amount of space ... and a trivial workaround if you later get a 5TB drive with a few more bytes. I'd do exactly that -- just "tell" MyMain it's fine, and then the warning won't show up anymore

You posted that earlier -- that's where I got the number

Forgot the link: http://www.hdat2.com/

4,883,770,552 x 1024 = 5,000,981,045,248 ... which is > 5TB by disk drive size standards, so it's likely that's the correct number. I suspect this is a display anomaly in MyMain, which doesn't "know" standard values for 5TB or 6TB drives. I suspect bjp will be updating this in the not-so-distant future (we need to establish the correct values for these - I suspect you have provided that for 5TB units) ... but meanwhile, just use the process bjp noted earlier to "tell" MyMenu that your drive is okay. Worst case: If it actually has a tiny HPA, you could eventually learn that when you try to add a 5TB drive that doesn't have one -- UnRAID wouldn't allow it (since it would be larger than your parity drive). You'd simply have to use the new 5TB unit as parity, and use this one as a data drive. If you really want to be certain, attach the drive to a spare system (or just unplug all the drives except this one); boot to a DOS CD with HDAT2; and see if it shows an HPA. If so, you can remove it with the SetMax function. You don't actually have to have ONLY this drive in the system - I just always do that as a paranoid precaution against accidents when I'm using a tool like HDAT2.

Obviously it depends on when you purchased your 1430SA's ... or, more precisely, when they were made and what firmware revision was on them when you got them. The updated BIOS was released on 8 Nov 2010 (Release 2507). The Release Notes for this update include the following: 4. Major Functional Changes: -------------------------------- 1. Added support for drives larger then 2TB in size. Your testing thread is dated March of 2011 ... I suspect you purchased/borrowed a 1430SA that was fairly new at the time ... so it already had the Nov 2010 update. But a lot of folks using UnRAID are likely to have controllers earlier than that -- and they need to apply the update if they want to use this controller with v5/v6 and drives > 2TB. I've seen a LOT of motherboards that won't support drives > 2TB. No new boards are going to have this issue -- so it depends on what you consider "... new enough to be considered for unRAID ..." I agree it's very unlikely anyone building a system today would be likely to have this problem ... but if you're planning to use a board from an older system, it's a good idea to check compatibility with the larger drives. With a bunch of older systems being sidelined by those who don't want to use XP anymore due to end of support, there may be some who want to repurpose that hardware for basic UnRAID systems ... and some of these could easily have motherboards that don't support the larger drives.

Yes, a lot of folks are using these with > 2TB drives, so clearly that's not an issue. Sounds like your upgrade plans are just fine ... just wanted to be sure you'd checked, since you're not currently using any drives above that threshold.

Based on the response from Asus to a question r.e. supporting drives > 2TB on that motherboard, it appears the motherboard ports don't have any problem with that. The other question is whether or not any add-in controller you might be using supports them. For example, the Adaptec 1430SA often used in UnRAID systems a couple years ago DOES have the ability to support the larger drives, but only if you flash it with updated firmware.

I presume you've already upgraded to v5 (or 6) ... but a more compelling question (given that you've currently got all 2TB drives) is whether or not you've confirmed that the system supports > 2TB drives ...

8) Somehow I suspect Toshiba has sold a few others (Although this may indeed be one of the first used in UnRAID)

An HPA is a "Host Protected Area" -- an area reserved on a hard disk for a variety of purposes. It's most commonly known in UnRAID as a Gigabyte-related issue, as Gigabyte motherboards used to save copies of the BIOS in an HPA they would create on an attached drive. It's also been used by Dell to store Media Direct files; by Lenovo/IBM to store recovery data; and by others for various purposes. The issue that prompted this thread was WD using an HPA to store some info unique to its MyBook drives. When you create an HPA, you effectively reduce the size of the hard drive by the size of the HPA. The issue that creates here is that any drive with an HPA will be slightly smaller than the same size drive without an HPA. Since parity must always be >= the largest data drive, a drive with an HPA can't be the parity drive if there are other drives of the same capacity in the system that don't have HPAs. It's not a problem for a data drive -- you lose a few KB, but that's the only issue. But clearly you should be aware of this if you're harvesting drives from external enclosures.

This was Tom's post where he made that quote: http://lime-technology.com/forum/index.php?topic=32859.0;topicseen Very interesting (and apparently mostly true) account of how Pixar nearly lost Toy Story II !!

Tom hits the nail on the head ...

Note that the guy in this recent case rebooted his array with the wrong USB flash drive, which had incorrect configuration information on it -- I'm not at all surprised this caused some issues. I agree that in this case it would make a lot more sense for UnRAID to give an "Invalid Configuration" message and simply stop ... but what's not clear is whether or not the bad configuration data was based on the same set of disks -- in which case the serial #'s would have been okay, and it may have simply started a parity check based on a drive that WAS the parity drive in an earlier config (e.g. before it was replaced with a 4TB drive). So it's not at all clear that what UnRAID did was "wrong" -- but it IS clear that the user messed up big time. And, of course, if he had backups ... I will NEVER understand why anyone who has data they don't want to lose doesn't back it up.

That is indeed a VERY good price for a modular HX series !! Just ordered one as a spare PSU ... I'm sure I'll find a use for it "one of these days"

As already noted, you may not want to use that card due to its 2TB HDD limit; but the more general answer to your question r.e. card size is well documented in the Silverstone manual. From page 17 of the manual: *There is a 6 inch limitation for standard width graphics card *If the card width is less than 2.35 inches, then there are no length limitation Page 18 details how much these restrictions relax if you replace the stopper plate on the 3rd drive module -- but if you do this, you can only install 7 drives, so that's not applicable for your use.

?? The case should have about any card you want. The drive cage has a section missing on one of the bays, so if you need that space for a long add-in card (e.g. a video card) it will fit just fine (it does, however, reduce the useable drive count to 7 in the hot-swap bays). If you're referring to how many ports you may have to add to do a specific configuration, that's a function of what motherboard you use -- NOT a question related to the case.

A cache explains it ... I had looked at your config, but "... small bunch of drives ..." isn't particularly descriptive

Are you saying you're getting 70MB/s writes to a parity-protected array ?? Which write mode do you have UnRAID set for?

Understand. I agree there are (rare) cases where a non-correcting check is useful. In my case, if I'd had the same issue, and run a correcting check, I'd have run a checksum validation on the drive, identified the corrupted files, and simply replaced them from my backups. The real solution to this issue is dual parity ... then a parity error can be corrected with no "guesswork" as to where the error is

I don't have any problem with the idea of running non-correcting checks. It's simply that there's very little you can do if you find errors, other than just run a correcting check afterwards. If there were better tools [i.e. a "potentially corrupted files" tool that would show which files on every disk were potentially corrupted by a specific bit location's parity being wrong", then it would make more sense to do the non-correcting check -- PROVIDED, of course, that you have a way to USE the information. For example, if you knew that files a, b, c, d, e, and f were potentially corrupted due to a parity error, do you have the ability to verify which ones are good or not? If you don't have checksums or backups (or both), this information isn't of much utility. But I agree that if you CAN answer that question reliably, then a non-correcting check would let you identify those times that the error was actually on a data disk, although then you have the issue that there's no "rebuild this file from existing parity" tool (you could, of course, rebuild the entire disk).

I'd go ahead and install UnRAID on both drives (you have to install it to determine the GUID, which is needed to get your license). Then get your two licenses, and add them to the appropriate keys. You now have two flash drives that will boot UnRAID. Now set the 2nd drive aside in a safe location ... and configure your UnRAID server with your primary key. Set up a routine backup to copy the complete flash drive EXCEPT the key file (Pro.key or Plus.key) to a folder on one of your PC's. You can have this auto-update as frequently as you wish (Mine does it every night). Then ... if something goes awry with your key, you simply copy the backup folder to your 2nd flash drive; insert it in your UnRAID server, and boot