ibixat

I had some issues with an improper shutdown and am running a parity check, the reads for the pairty drive and first 6 disks in the array are all staying very close in number, the last 4 disks are showing 1/3 or less the number of reads, all the drives are the same size , 6 are connected to the motherboard connections and 6 are connected to 3 different 2 port pci expres sata controllers. here's a list of the drives and their reads partway through, I zeroed it out and waited a bit then took this snapshot. I've listed where each drive is hooked up as well Temp Size Free Reads Writes Errors Parity (sdh) 28°C 2 TB - 421925 0 0 Motherbaord Disk 1 (sdi) 27°C 2 TB 132.81 GB 420526 0 0 Motherbaord Disk 2 (sdj) 26°C 2 TB 197.19 GB 419893 0 0 Motherbaord Disk 3 (sdk) 27°C 2 TB 116.5 GB 419486 0 0 Motherbaord Disk 4 (sdl) 27°C 2 TB 168.32 GB 423743 0 0 Motherbaord Disk 5 (sdm) 27°C 2 TB 187.37 GB 421847 0 0 Motherbaord Disk 6 (sde) 28°C 2 TB 981.58 GB 418824 0 0 PCIE2 Disk 7 (sdf) 26°C 2 TB 879.83 GB 131932 0 0 PCIE3 Disk 8 (sdc) 27°C 2 TB 1.41 TB 121068 0 0 PCIE1 Disk 9 (sdb) 27°C 2 TB 2 TB 145137 0 0 PCIE1 Disk 10 (sdg) 26°C 2 TB 2 TB 123429 0 0 PCIE3 Cache (sdd) 28°C 1 TB 1 TB 0 0 0 PCIE2 Does this seem normal due to the way the drives are connected or is something odd happening?

I never said I unassigned the cache disk, I disabled it on the one share that I was about to transfer a large amount of data to. It was still enabled on 2 or the 3 other shares. It's safe to ignore paragraph 2 for the most part, as I said that was venting from being frustrated at issues I'm having, the real issue is stated explicitly in paragraph 1 and reiterated in 3.

I'm not sure if anyone else is seeing this behaviour but it's not hard to replicate. If you stop your array and restart it without rebooting the machine does it properly remake the user0 mount point? I had stopped mine and restarted and the user0 mount point was gone, I didn't think too much of it as I didn't need to use it right then and I had a memory upgrade coming so I was rebooting soon, but then mover didn't run for me. I stopped the array and rebooted and the user0 was back. The next day comes and I have a large quantity of rips I need to move from my pc's HD to the array, I want to disable the cache drive for the share I'm using so I disable it but last time I did that the change didn't take effect till restarting the array. So I stopped the array again and restarted and once again the user0 is missing, the log shows it being removed but never recreated. Long story short I had some issues at that point with drives saying for free space "resizing" after stopping and restarting array and they wouldn't change to normal and they were causing problems so I had to bring the system down improperly, machine is back up and running user0 is back my parity check is running and hitting a bunch of out of sync problems, no doubt from my bad shutdown, the data I recently wrote to the array seems to be in good shape so I'm not sure what's going on there but that's not really the issue I'm posting about. I kinda wish sync errors during a parity check could also report the file on each disc that could possibly be affected so i could verify they are not problematic etc. I'll run the file system checks and another parity once this one finishes. Anyway, the thing I wanted to bring up was the stop and start array, for me it does not recreate the user0 mount point, can anyone else please check and verify if the same issue happens to them. They sync errors and whatnot are my own screw up, so the middle there was just me venting =P thanks for listening. Brian

I don't have experience with this specific case but I did pick up a Xigmatek Utgard for 50 bucks from newegg, it is similar to the elysium with only 9 5.25 bays and one of the internal 4x3 drive cages included. It's currently out of stock for that version but there are two more variations they sell for 10 bucks more each. They sell the internal cages for about 25 bucks each so I grabbed 2 of them as well. I think it worked out nicely in the end, the drives stay a decent temperature even when the room gets warm. I Imagine that their higher end case will be at least as good or better than their mid tower and I was very pleased with the quality level of the Utgard given the low price, it feels nice and solid and all the edges were smooth etc, no gaping wounds from simply assembling it. I'd definitely buy another Xigmatech case if the price range was right, I'm not sure why the 12 bay elysium costs so much though. I'm running 12 drives in my Utgard now and most run between 25-30 C, the 7200 rpm cache drive usually 2-3 warmer than the rest. Without the 4 in 3's the drives were hitting 34-37 on the top end, once I put the 4 in 3's in the drives dropped down to the high 20's low 30's at their hottest. Have some pictures of how it looks inside and what the front of the 4 in 3's look like with the front of the case off, I'm wish I had better cable management skills but it seems ok so far.

I'm guessing he's looking for pictures of the antec 300 with cages, and info on what cages are used in the pics.

I don't know how much the shipping to the UK would be but monoprice has a very inexpensive pci express 2 port sata card. http://www.monoprice.com/products/product.asp?c_id=104&cp_id=10407&cs_id=1040702&p_id=2530&seq=1&format=1#largeimage I wonder if when shipping and whatever else is added how much that would end up costing you. Hmm nevermind, I put in a london postal code to get an idea and shipping was kinda high, brought the total for the card + shipping to around 26pounds, it appears that the shipping doesn't go up by adding some cables too so the breakout cables you need added on brings it up to 38ish pounds for the 2 port card and 2 of those forward breakout cables. No clue if you'd get dinged for Vat or taxes.

at this point it's sounding like a safe bet, either the CPU or the MB itself must have taken the hit.

As is expected, once I set up unraid to store my media collection my collection rapidly expanded to fill all available space. Now I'm ready to slap a few more drives into my main system but I'm curious about the preclears. When I setup this system I was only able to do a few drives at a time as they were in use or being transferred over etc, so in the end I had precleared 3 drives transferred information over and then precleared the next 3. I never really was using the system too hard while the second batch of preclears ran. I have 5 more drives coming in this week (for those of you wincing at 5 drives in one order I 2 WD 2 Samsung and 1 hitachi to reduce the chance of a bad batch of one type wreaking havoc) but I was curious what opinions there were regarding preclearing multiple drives while using for normal media serving duty. Would you just take a system with 7 drives currently (1 parity 5 data 1 cache) and preclear the 5 drives all at once in that machine, would you limit the number to preclear at once to 2 or 3, or would you consider hooking them up in another machine and clearing them there? Just looking for some suggestions from those who have more experience with this than I do so any tips or suggestions would be welcome. Thanks Brian

my last power surge (possibly lightning) based pc disaster was the power supply. I could turn it on the fan in the PSU would spin the lights would come on but the machine wouldn't post. Hooked up a new power supply and that machine worked until I replaced it. I'd very much second Joe's thoughts on the PSU at this point.

Does that rig play Blu-ray, DVD-A and SACD ISOs? It will play high def content no problem. I do not thing that the current release of XBMC supports Blu-ray ISO files. It would be best to check there site and read up on the xbmc.org wiki. Blu-Ray ISO is in the early stages, the main movie with audio and subtitle selections + chapters is working for most ISO's but there are still issues. Full Blu Ray Menu support is not even on the to-do list at this point and likely never will. It can be done on windows using an external player and appear seamless, but of course that requires you have windows on the machine and run XBMC on top of that.

one of the best write ups I've seen about xbmc is here http://www.avsforum.com/avs-vb/showthread.php?t=1294647#8 i'd read through that and see if it answers any questions you have about it.

There isn't any reason why it wouldn't work, just make sure you have it secured so that the cables and or cage are not able to be jostled about too much, I'd suggest at minimum locking sata cables and hopefully no pets or kids around who like to mess with stuff like that. It will expose cabling to possible outside elements (pets and children) that normally don't mix well with cables, but if that's not an issue and you don't mind the appearance it shouldn't really hurt anything.

The bitstreaming Lossless HD formats from XBMC is a bit of a nuisance right now. I believe that audioengine (the Next version of XBMC's Audio processor) will fix most of the issues, not necessarily by bitstreaming them (hardware limitations on a lot of cards prevent this) but by decoding them and sending uncompressed pcm etc. Several ATI cards can bitstream the HD formats, as can a few of the Nvidia 4xx cards, the Nvidia ION (1 and 2) can't, I believe there are some onboard intel solutions that can do it. Whatever you do just make sure you research it thoroughly so you don't get buyers remorse, but I think your post here is an indication that you are already doing that =) Now as far as XBMC goes the customization it can offer and the different skins etc really do make it the gold standard like DoeBoye said. Plus being open as it is anyone can write a plugin or add support for some random bit, you're not stuck waiting on a company to possibly do it. For example if you're into comics at all, or even if you're not, check out this thread for a good example of what can be done with XBMC if someone puts their mind to it, http://forum.xbmc.org/showthread.php?t=48191. In the end from everything I've read about them online I don't think you'd be disappointed by a Dune or PCH c200, as far as your question on cost, you can put together a VERY reasonably priced HTPC to run openelec for around the same price I believe.

When the new audioengine is finished for xbmc and it can bitstream the lossless formats over hdmi or decompress and send them as pcm will that be considered "accurate" by audiophiles or will there be some bias against using it still?

Quick note, you can hit shift page up and shift page down to scroll up and down from the command prompt to see messages you may have missed The buffer is pretty large so you should be able to catch what you need, you can also type from anywhere "cd /boot/preclear_reports" then "cat preclear_rpt*" and it will spam your screen with all the preclear final reports for each disk you've run, and use shift page up and down again to scroll through to check the results. This is of course if you want to do it on the unraid machine itself, it is easier to just browse the files in notepad2 or Notepad+ from the share.

After having some mushkin and ocz memory crap out on me this month I ordered replacements and went with Kingston this time. Out of curiosity since everyone here seems to have oodles of experience, who is your preferred memory manufacturer? And if you've ever rma'd with them how was the experience? I've built my own pcs since I was around 14 or 15 years old from spare parts obtained through my moms work at first then by way of part time jobs in high school. I don't even know how many computers I've built but when it comes to memory i draw a blank when it comes to preference. I always just bought whatever was a good price and up till recently had never really had any issues. For reference I'm 36 now so I've not been doing this quite as long as some of you . Anyway was just wondering the above as I sit watching memtest86 run on 2 machines to make sure I remove all the bad modules. Brian

They shipped my the next day after putting it on backorder, so I got them both at the discount. Have both my PC and the Unraid + modem and switch hooked into the one, the other one is my Av receiver, HTPC, TV and Tivo. MAN does my TV draw power, I love the display on the front showing the current load, makes me think a new TV is in order so I can save money on electricity...that's totally my story and I'm sticking to it.

I was getting a consistan 25 MB/sec using MC transferring from a USB drive to a array of Samsung F4 drives with parity active. I also thought that was kind of slow given the network access speeds I usually get but I just ignored it and let it run overnight. Maybe MC has some limitation in it's speed?

after all the testing and reseating the DIMM's it copied the 1.5 TB without a single error. Prior to that it was getting them all the time, and after reseating the memory never failed another memtest.

Rather than run memtest first (I know I know) I decided to just finish copying over the first set of data, teracopy was catching any errors before and since the array was basically brand new if it was still problematic I could just wipe it out and start over after the memtests. End result is I copied over 1.15 TB consisting of 2670 files and so far the test phase of teracopy hasn't found one error in the first 2100 copied over. Definitely looks like I had a badly seated dimm. I'm still going to fire up memtest after everything is copied over, just to be on the safe side. Brian

I was getting CRC errors when transferring large files in teracopy, after much frustration since I just built the machine I started a memtest86 and was getting errors on test 5 and test 5 only, I've removed 1 of the sticks and test 5 passed just fine with the single stick in the machine, the then swapped the two sticks to test the other one and it also passed test 5 when it's in the first socket. I then tried both sticks in the socket I initially removed a stick from to see if that was it and they both passed test 5 in that socket. I let each test run through 2 pass cycles, so then I put both sticks back in their original locations. And let it go through 10 cycles of the 5th test, and one full cycle of all tests and now I'm not getting any errors anymore. So would the prevailing theory on this one be that I was just a lamer and didn't seat one of the dimms very good the first time around and randomly picked it as the one to remove first? Should I let memtest run for a solid 12-24 hours maybe before I start copying files, I'm somewhat wary to skip to copying and relying on teracopy to find any errors at this point though I am hoping to get this system in place sooner rather than later. After 24 hours of preclearing, 10 hours of the initial party sync I'm anxious to get the ball rolling, but data integrity has to come first of course. Thanks Brian S.

Thanks for the heads up on this deal, I picked up 2, one for the new (and my first) unraid server that I'm building and the other for my htpc and tivo (Tivo takes far far too long to reboot when power goes out), I was on the fence and naturally my power dropped for about 3 seconds while I was looking at the page considering the purchase... Well I did order 2, and when I checked the order online it was fine, now several hours later it's showing me as backordered instead of the "expected to ship in 1 to 3 days" that I had earlier, no email from officemax letting me know that the order status changed either, hope they still fill my order at the advertised price.