wheel

Well, got everything in, managed to barely fit the internal cage, and I'm wondering what I'm going to do when I need more SATA power cables (PSU only came with 2 4-prong plugs), but my biggest headache is the red "Power Failure" LED 6 on the motherboard. I tried unplugging everything but the PSU and the motherboard (except the CPU, left that hooked up) - any other ways to tell if it's a bad PSU or something else? This would actually be my second dead Corsair straight out of the box (last one was my first UnRaid box, which was a painful and long RMA process... Really hope I don't need to repeat that one, but I guess I'd rather it be a simple fix than something worse!) RE: PSU, the fan doesn't even spin up - but it sounds like this RM650 won't even start spinning up until it reaches a certain temperature, so I don't know if that's a red flag or not.

And I'd FINALLY made it around to upgrading parity at 6TB for just a shade lower than $275... Story of my Unraid life Bring on the reasonably-priced 8TB Reds, pronto!

It's the mental aspect of it - like building a lego monorail set without the piece that gets the engine running. Staring at that lonely nigh-complete tower while my RAM spends its third full day sitting in a shipping facility in Kentucky and my 4TBs stand ready for preclearing would just eat me up inside. Really hoping I see some movement on tracking tomorrow at least, but something tells me newegg saver (which has almost always been more like 1-2 days than 4-7 days for me) is finally giving me the shipping speed it's promised, and I'll be getting my ram on Tuesday. Wish I could be cautiously optimistic on this one, but that damn tracking status...

Newegg saver on all my newegg items; everything (and I mean EVERYTHING) I need to build this sucker is already here... except my RAM, which has been sitting in the same shipping facility since Monday. I could get started, I know, but that inability to boot the damn thing up - I'm just going to wait. Hopefully posting a build update this weekend, but that's up to newegg saver...

Not sure if it's a record low for the product (outside of gold box madness), but it's low enough for a data drive plan I have on the horizon. Thought I'd post as some small measure of gratitude for all the help this forum has provided on my other two recent threads! http://www.newegg.com/Product/Product.aspx?Item=N82E16822178520 Limit 5 ESCAKNA226

That... actually makes a lot of sense. Sorry for the goofy question, and thanks RE: drives - they're all in hotswap cages, but I'll be replacing the cache drive soon, so when that happens the inside is getting a full cleaning (probably 3 years since I've been inside that beast thanks to the cages).

Yep! That's what I did with my first key(s); for the extra $~30, makes sense to prepare for more even when I'm thinking "surely I'll have all I need NOW!" ... Not sure the chicken or egg on this one, unraid vs. hoarding tendencies. Thanks again, garycase!

So now that I've finished ordering pieces for my new, fault-proof (ha) action machine, it's time to turn my attention back to salvaging what I can from my once-proud alpha box. Just making sure I'm clear on steps: (*) Preclear New 4TB x2 (1) Remove two red balled drives from array, replace with new 4TBs (same capacity as old) (2) Power up box, new config (builds new parity) (3) Run parity check (4) Reiserfsck red balled drives, move salvageable content to new 4TBs My super-paranoid question: if one of my other drives (i.e., a THIRD drive) is close to dying and dies in the parity build (probably unlikely, but this incident has me gun shy), would it be smarter for me to try rebuilding the parity WITHOUT the two disks (removing 8TB from the total data footprint), then once I know I have parity protection again, try adding the two disks? Longer process, which I'd like to avoid if possible, but losing a third disk would REALLY be a headache with two already down.

Trigger pulled! Thanks for the offer, mr-hexen, but decided to lay out the tower thusly: 4-drive internal cage = 3 drives = 6TB HGST 7200rpm Parity (new), 4TB WD Green (new data drive, hopefully last awhile), 1TB WD Black (new cache drive, hopefully last awhile) 1 SATA-to-eSATA backplane Remaining 10 SATA slots = 8 in the normal slots, 2 in the 5.25"s ...no real need for a cage now that I've found an internal setup I'm comfortable with (though knowing my luck, even thoroughly Precleared 4TB may end up dead in a year or two). C'est la HD. Everything listed a few posts above has been ordered, plus a $50 shipped and taxed fan for the internal cage. Probably going to spring for the two-key unraid license (never thought I'd build a third, so may as well prepare for a fourth), and total spend without that license and a new low-profile USB stick was (including shipping and tax) $1075.82. BUDGET SUCCESS, if I just did a single unraid license. Thank you to everyone for your help with this! I'll post results once I get everything in, up, and running.

So if I put the 3-in-2 in the 5.25" (Cache, Preclear Slot, and Empty for Now) and still have 8 slots up front, that puts me at 11, with one permanent gap in the internal cage that'll hold my parity and (eventually) the last two data disks I'd install. Still need to figure out a fan for the internal cage though, and find a 3-in-2 with quiet enough built in fan to not defeat the silent goal... Unless I'm better off with a different approach for the internals if I'm only using 3 slots at most? In theory, I won't be filling 2 of those 3 for a year or more, if ever, but I'd really like the parity drive to be "buried" and figure I may as well do that at the building stage - in 4 years of unraid, haven't needed to replace one yet. Of course, now that I've typed that...

Already may have answered one of my own questions. Based on the description of "almost hot swappable" in this thread (http://lime-technology.com/forum/index.php?topic=34096.0), it'll probably make most sense for me to go with the esata backplane, but just connect one to the MOBO and use the 5.25" as cache drive space (doubt I'll flip that one out insanely often, but Preclear is a different story). EDIT: Or maybe http://www.newegg.com/Product/Product.aspx?Item=N82E16817994143 for the 5.25" bay, and keep the cache and parity drives on the "unscrew everything to change disks" drive cage...

Based on this list, I think I'm at $1100 or a little over with tax and shipping: CASE: Fractal Define R4 ($120 + tax) MOBO: Supermicro MBD-X10SL7-F-O ATX LGA 1150 Intel C222, DDR3 ($250 + tax) CPU: Xeon E3-1271v3 ($350 + tax) RAM: 2x 8GB G.SKILL Ripjaws X ECC ($140 + tax) PSU: Corsair RM-650 ($120 + tax) INTERNAL CAGE: 4-Drive per garycase ($30 + tax + s/h) ...and the unraid license brings me up to my budget. Nice! garycase, thanks again for all of the incredibly helpful suggestions; this is how I'm thinking of laying things out: Start with internal 4-drive cage (any suggestions on a silent fan for it?), but only place the 6TB parity drive in there for now. Won't be filling the other 3 slots until I've maxed the case, unless that's a bad idea for some reason. Start filling up the 8 front slots with existing data disks (2, 3 TBs, have at least 6 good ones, maybe more) Use one of the 5.25" bay slots for a cache drive, and keep the other open as an easy-access Preclear slot. ...or is the 5.25" difficult enough to work with that the eSATA back plate makes more sense? I'm trying to minimize messing with the box once it's up and running, but it'd be really nice to have a dedicated Preclear slot or dock that's insanely easy to access. Any thoughts? I'm about to dig through other threads talking about the Fractal R4 and how best to work with it, then I'll head back here before ordering parts. Thank you again to garycase and anyone else who can help!

Thanks, Gary - you've been tons of help today! Some OCD part of my brain hears about the 14 slots and wonders if there's a way to do anything fun with the extras in a case like the Fractal - assuming there's not a bigger version of the R4 or 5 that's equally quiet and takes advantage. I'm thinking at least one could be used for a SATA-to-ESATA dock that could serve as an external Preclear slot when needed. Does anyone have any experience using this board for something like that?

Also, yes on the correcting check if errors were found (but I can't remember the last time errors were found before this - a year or more ago, at least).

Thanks, Gary - so if one disk was DISK_DSBL yet started displaying that temperature again once the other disk died beyond a doubt (no spin, nada), there's pretty much no chance at this point that DISK_DSBL was for anything other than a second (or, temporally, first) dead disk?

Thanks, Frank - I managed to save about 130gb (!) before transfers stopped and the Disk15 status light on my unraid dash went to redball. Weirdly enough, Disk 13, which had been * temperature the whole time I was reading data off of the Disk 13 folders that were still visible (most were missing, and almost everything on Disk 15 was missing, visibly, though windows still reported "full" drives) went to 29 degrees Celsius when Disk 15 was reported dead. It stayed at that temp (while Disk 15 went *) until I powered down the box. It has remained powered down since that day. I'm finally getting time to work on this again, and really appreciate your guidance and that of anyone else on the board who can help figure out whether Disk 13 had something other than dead drive occur, because it would be awesome if I could rebuild Disk 15. Really, really awesome. Huge pain if I can't, though I will survive it. Tried the UPS test; worked fine. I have a second UPS I'd been planning to replace the first one with, and will use that moving forward to be safe. I don't believe any plugins were running - all that I have on my cache drive is in "." invisible folders, so mover shouldn't be a factor, and though I've tried Sab and Sick in the past, I don't think either was running at the time, though I guess they could have been. It's been quite awhile since a shutdown. I'll wait to hear back from someone on here before proceeding, but I do have two 4TB drives ready to replace both Disk 13 and 15 if necessary. Thanks!

Today's a shopping day, so I figured I'd start this thread in case anyone has any suggestions as I go along. I have a little experience building unraid-specific boxes for myself (two down, this is third), but only by following builds and posts on here. I have zero experience with soldering or difficult modding for fans, so I'm looking for a box / drive cage (if necessary) combo that's already silent or near-silent (this needs to be a unit that fits in a room adjoining my viewing area, maybe 15 feet away through open space). I feel like 8 slots would be enough (parity, 2.5" cache, and six data drives), as I have plenty of room to grow in my other two unraid boxes, but if there's only a minor jump in cost or noise, it might be nice to have the extra space to make old 1 and 2TBs useful as I move them out of rotation; I just wanted to build another one specifically for "active" material (vs. the other two being archive), so that would just be icing on the cake. Silent and under $1200 are the main concerns. Accordingly, I'll want a "beast" system (to use Rajahal's terminology) that can handle the processor-intensive plugins, including speedy extractions, transcoding, and encoding. It would also be nice if I could skip the SATA add-on cards I've used in my other boxes and go with one of these new motherboards that handles 8-12 SATA connections, but it looks from a bunch of posts like these are often difficult boards for unraid. The $1200 does NOT include hard drives - I'm set on those. Really, my key requirement is silence or near-silence. A cool-looking box wouldn't hurt, but there's a decent chance it'll get stuffed in a ventilated, open area behind a cater-corner couch if it's quiet enough, so a silent ugly box wins over a good-looking near-silent box. I'll even go up to $13 or $1400 if that jump guarantees some sort of space-age negative-sound system, but I'd hate to jump up to $1500 on what should just be a fun project (and sanity saver, for reasons explained in my 5.x Support "salvage" thread on another of my boxes). Thanks to everyone in advance for any suggestions - I'll post my shopping progress on this thread as I go.

Unfortunately won't have a lot of time to work on this after the holiday, and something tells me I won't get a lot of response today, so here's my plan: -Pick up two 4TBs on my way home from work -Try Rebuild on Disk 13 No idea what will happen if it fails, but if at that point I still have access to at least some of the files on 13 or 15 (the few that I can currently see), I'll try to salvage those into external storage. At that point, I guess I have to try and start fresh with disks 13 and 15 (the two new 4TBs) and hope that I can identify any new problems with old disks before my next power fluctuation (which seem to be too damn frequent in my town). If I'm making a huge mistake with either one of these steps, I really hope someone advises me against them, but if nothing else I'm leaving a record of this nightmare with these posts.

NIGHTMARE SCENARIO After no response to this post and some searching that indicated those error messages were nothing to be worried about, I let the issue lie. Even threw a new 4TB on top of a 2TB (parity check before and after, no problem) Well, that 4TB decided to go DISK_DSBL on me at 3am last night during a power surge (yeah, I'm APC-protected; not sure what the hell happened to get it powered off, I only woke up to the beeping) "No problem," I think, "I'll grab another 4TB in the morning and do a rebuild over New Year's Eve." So this morning I'm half-awake and start the array (which takes a minute, and was probably the decision that killed me) Everything's fine, but I notice a whining and clicking noise coming from my tower. Then I notice a * next to the temperature on Disk 15 in UnMenu. Then I notice 13 errors on Disk 15. Not the disabled Disk 13, that one's "fine" error-wise post-array start. 13 errors on Disk 15, the * temperature drive. So I go to check the folders that were in Disk 15. Two of them are fine, but one - containing almost a terabyte of data - appears completely gone. "Oh, shit." Here I am now, heading to get that 4TB still, but absolutely lost at what's the best next step to take. I assume that with two drives down (one disabled for sure, the other spitting out errors), my rebuild attempt will be worthless. How can I best salvage data from here on out? Is it too late for BOTH of those damn disks now? Please advise ASAP, and THANK YOU FOR SAVING MY SANITY! syslog.txt

Running unraid v5.0.5 for about a year now; box itself is 3 years, 3 months old, no hardware replacements other than drives. Mobo: SUPERMICRO MBD-X8SIL-F-O CPU: Intel i3 540 3.06 GHz 4MB LGA1156 CPU BX80616I3540 RAM: Kingston 4GB DDR3 1333 PSU: CORSAIR Enthusiast Series TX650 V2 650W Case: Antec 1200 Hot Swap Drive Bays: Norco SS-500 x 4 SATA Expansion Cards: Supermicro AOC-SASLP-MV8 8-Port SAS/SATA Add-on Card x 2 Cables: 0.5m 30AWG Internal Mini SAS 36pin (SFF-8087) Male w/ Latch to SATA 7pin Female (x4) Forward Breakout Cable x 4 Just started hearing a two-tone beep from the system, and when I checked the system log, I noticed this right up front: ...and the Machine check events freaked me out enough, but then I scrolled up and started seeing errors like these: ....and these: .... See attached syslog for the full context, but both of those blocks are freaking me out a bit, as I have no idea what they mean, nor have I ever seen them before. On Oct. 27, I replaced a 2TB with a 3TB (non-correcting parity check before and after rebuild); no problems, or so I thought. Parity check started late afternoon on Oct. 27 local system time and completed on Oct. 28 at 8am local system time. I've been running a non-correcting parity check monthly; no problems. Unraid has been blissfully perfect since my 5.0.5 upgrade, but now I'm wondering how long these late-night errors that don't seem to draw any attention unless regularly checking syslog have been showing up (or damaging my parity). Also, I'm not sure if this is a factor, but I have a feeling my cache drive is on its last legs, age-wise. I've been meaning to get a replacement for backup's sake if not instant replacement the past few weeks, but haven't managed it yet. Any advice on how best to proceed safely? syslog.txt

Hi, all: after weeks of digging through threads on this case and its Lian-Li counterpart, I'm leaning this direction, but could use some advice and maybe validation on this design before I start pulling triggers. I plan to eventually use all 8 hotswap slots and one of the 2.5" internal slots (for a cache drive, to be installed later). I also plan to add a 4x SATA card at some point to maximize those slots. Finally, I want to make sure I'm keeping this sucker powerful enough to run the "usual suspects" CPU-intensive background apps mentioned in countless other posts. My number one concern is cost control, but not at the expense of the goals listed above. Any thoughts on areas I could cut? Also, any suggestions on a cheaper motherboard that could still work with this SFF (and expansion plan)? Or if I'm spending this much on a motherboard anyways, should I just go for that new one with the 12x SATA onboard and skip the SATA expansion card altogether? Thanks in advance for any advice you can provide! I'm thinking about pulling the trigger on some of this today. I'm buying all parts in the United States, from California. CASE: Silverstone DS380 MOBO: ASRock E3C226D2I Mini ITX Server Motherboard LGA 1150 Intel C226 DDR3 1600/1333 (am I going to need specially angled SATA cables for this board in this case?) CPU: Intel Core i3-3225 PSU: Silverstone SFX 300W ST30SF RAM: (still deciding; leaning towards the cheapest DDR3 4GB stick I can find around town unless I'm advised otherwise) Thanks again for any advice and guidance!

Scratch that, just nerves - webgui FINALLY reloaded. Never seen one take that long...

Starting to get concerned: been over 20 minutes since I started the "start Data-Rebuild, and then expand the file system" phase. How long should I wait before forcing a power down and replacing the old 2TB, or is that a bad choice now? Ran parity check today, everything was fine before upgrade.

Just like it sounds, didn't see anything on search - I've run out of slots, and am planning to ask a friend at a repair store whether they have any old boxes lying around they'd be willing to let go of for a song. Does anyone have any recommendations on bare minimums to keep in mind if the ONLY thing I want to do is slap in a drive and run Preclear?

Thank you again for the advice, everyone - the "new config" is protected (parity check before and after new sync), and the new drive is working its way through two preclear cycles. Next step: upgrading to 5.0 final!