lionelhutz

It appears you believe we all are clairvoyant or something like that. No one can help you unless you tell us what you want to do with your file structure.

FYI, if you switch the parity mode to reconstruct-write then the moves would likely go a little quicker. All drives have to be spun-up though.

It has to do it that way to ensure that the parity is correct when a new disk is added. If you add a new 4T data disk the parity may not be valid in the area you didn't let it complete.

Ya, it appears it is a 12 bay case. Those parts look OK but they don't match. The CPU is LGA1155 and the board is LGA1150. Also, Gigabyte puts what's called a HPA onto a hard drive to save the BIOS and the HPA messes with unRAID. You have to either ensure that feature can be turned off or just use a different brand. Biostar and ASRock seems to be fairly popular for home PC class boards or Supermicro for server class boards.

Compass in the second link. $199. The case is bare. You have to buy a motherboard and a processor. As a minimum, make sure the processor has built-in graphics. For example, the Celeron G1820 processor. Then, make sure the motherboard supports the processor and has video output. After that, sort the motherboards by number of SATA ports and PCIe slots and pick a suitable board. You can get 8 SATA expansion cards so a motherboard with 4 x SATA ports and a PCIe x16 slot would pretty much work to connect all 12 slots in the Roswell case. As for memory, pick say 4gig of memory that matches what the motherboard uses. For the power supply pick a Corsair or PC Power and Cooling that has a single +12V power rail. Here's a good example. http://www.newegg.com/Product/Product.aspx?Item=N82E16817139028 Finally, if you're not sure, pick the parts you think and post them up and someone will help with suggestions.

That Rosewill 14 bay hot-swap case looks like a nice unit. You can't buy a tower and add the hot-swap bays to it for any cheaper than that costs. It would give you up to 48T using 4T drives and even more storage with larger drives in a few years. unRAID doesn't support hot-swap while the array is online but it does support hot-swap for changing drives and you can plug-in and preclear new drives without even stopping the array or powering down the server. Then, once the drive is ready you just stop the array, add it and start the array again. You don't really need a SSD just for file transfers. The network will be the speed limitation if you use a new HDD so the extra speed of the SSD would be a waste. I personally would never use an Atom. It's buying into a limited processor. A lower end i3 or i5 would idle at about the same power but they have much more processing capability available if you need it. You can also swap just the processor or motherboard if you find you need to. The price difference really isn't that much either.

I would first try keeping the config directory except the plugins subdirectory and keep the go file from the new install zip. Try keeping it all and if that doesn't work then delete parts and try again. I think just keeping super.dat, disk.cfg, share.cfg and the shares directory would keep the basic setup intact. replace the ones you don't keep with the defaults from the zip. To try and answer your questions about starting over. Yes, you'd have to assign the disks to the correct slots if you wanted them to be in those slots. It doesn't matter if you mix the disks up, but make 100% absolutely positively you DO NOT assign a data disk to the parity slot. That is the only mis-assignment that will cause a big problem. You can first assign all the disks to be data disks and start the array. If you get one disk that shows up as unformatted then that is the parity disk. The top level directories will create user shares since the top level directories are user shares. They would have default settings so you'd have to change those settings.

Thanks for the add but those were all at the end of last year and they are in service now and I won't need another one for a while. I sold my old ones off except one I put in a USB case so I got no drives left to test with. If I had a drive to test I've give it a run and help you out. I'm just following along because the preclear is so heavily used that any improvement is most excellent to see happening.

FYI, I had 2 Seagate 1T drives and they would both log a double link reset every few days. They never went offline though. I have hot-swap bays so I've never touched the internals of the server but I've replaced those HDD and I haven't found a single link reset in the log since so it was the Seagate drives. Unfortunately, I just don't know what to suggest for your system but I can see it's kicking your ass. It would help if you could summarize with an up to date equipment list and pictures of the server showing the overall build and others with details of the power cables between the PS and the HDD and the SATA cables between the motherboard and the HDD.

You probably installed a plug-in and gave it the location /mnt/user/disk3/apps as the location to install. If you don't first fix that then it will continue to be a problem. Your "disk3" is located on the cache disk. I suggest you go to the cache disk. The easiest would be to share it from the main tab by clicking on the cache name and setting it to be exported. Then, you will find a "cache" share on your PC. Move the apps out of the disk3 directory to the root of cache. Delete the disk3 directory. Fix your plug-in that was using /mnt/user/disk3/apps to be /mnt/user/apps or /mnt/cache/apps. Stop and start the array to make sure the changes take effect. Then, go to the shares and you should find a new apps share so set the use cache disk setting to only.

I can edit. Are you trying to edit a specific page or does it not work anywhere?

You guys do know the preclear script creates a report on the flash drive giving the 3 times and speeds? Here's examples from 2 different Toshiba 3T and a Hitachi 3T == Last Cycle's Pre Read Time : 6:43:49 (123 MB/s) == Last Cycle's Zeroing time : 5:48:03 (143 MB/s) == Last Cycle's Post Read Time : 13:52:32 (60 MB/s) == Last Cycle's Total Time : 26:25:32 == Last Cycle's Pre Read Time : 6:33:50 (126 MB/s) == Last Cycle's Zeroing time : 5:38:29 (147 MB/s) == Last Cycle's Post Read Time : 13:35:50 (61 MB/s) == Last Cycle's Total Time : 25:49:19 == Last Cycle's Pre Read Time : 6:38:14 (125 MB/s) == Last Cycle's Zeroing time : 5:36:23 (148 MB/s) == Last Cycle's Post Read Time : 13:49:19 (60 MB/s) == Last Cycle's Total Time : 26:05:05

Here's a package. Put it into /boot/packages and it'll install when you boot. I believe this is still installed if you use Influencers SABnzbd packages, and possibly with other versions of that package as well. http://code.google.com/p/unraid-greenleaf-repository/downloads/detail?name=par2cmdline_tbb-0.4-i486-1kh.tgz&can=2&q=

Here's an overview of the new security model. I don't use security but I believe this is all still valid. http://lime-technology.com/forum/index.php?topic=7047.0

It sounds great and there will be many people who will love the quicker speed you've found. Now I wonder if there was a good reason you disabled the randomized IO that you've forgotten about?

Saving 10+ hours would be for a 4T disk? Interesting. The last 3T drive I precleared took 13h 35m to post read. The drive wrote zero's at 147 MB/s and my typical parity check speeds are this fast too (I have all the same drives) so I suspect this would be the fastest average sequencial read/write speed to read the whole drive. It took 5h 40m to do this clearing, so I'd expect any post read improvement to have a similar read time as a minimum. It would appear just under 8 hours of time could be saved at the most, which is a good improvement. However, if you add random seek testing then I'd expect the average read speed to go down and the time saved to be less. FYI, pre-read was 6h 33m.

Lots of good info but I'm quite positive smb-extra.conf won't fix the OP's problem....

Make the share with a minimum free space and fill-up allocation method could work. Just know it will fill the lowest assigned disk first. You can also initially assign a single disk and then add another disk if the current one is running out of space.

Good catch. That one messes with people all the time.

Go to the settings tab and check the global share settings and make sure you don't have include or exclude settings there. If that is good try stopping and starting the array.

It appears there was a disk error when making the reiserfs so I think your cache drive is hosed. As a simple test manually create the share directory on the cache drive and see what happens. You might also want to run a reiserfsck on the drive. You need to fix the problem. There is no need to be using smb-extra.conf or a period in front of the directory name to perform a simple task like creating a cache only user share.

itimpi is wrong. It will never split between the 2 disks. If you do as you wrote then the whole movie directory you create will be stored on a single disk. Basically, split level 0 means that sub-directories can only be placed onto the disks which contain the existing top-level directories. Or, to put it another way, unRAID will never create a directory on another disk once it exists. So, when you copy a movie to the share and it creates the new sub-directory, that sub-directory will never be duplicated on another disk. It's actually a very good way to do a mixed level share. Just manually create the top level structure for the different file types on all of the disks and then the new sub-directories and files you place will always stay on a single disk.

Look at the description I put in the unofficial unRAID manual. http://lime-technology.com/wiki/index.php/Un-Official_UnRAID_Manual#Allocation_method Let us know if there is something you don't understand. Remember, it fils disk by disk so the highest number disk will be the last to "catch up". As others have noted, the split level takes precedence. As an example, assume you start in the fall with single shows for each season. If you use a split level that keeps complete TV seasons on a single disk then the allocation method just might start all these new TV seasons on a single disk since that's the disk presently being filled. Once that happended then all the episodes go to that disk.

Yes, pretty easy to build the system and then hot-swap your disposable drives onto each backplane and check that they are recognizable. Boot up any version of unRAID 5 (you can use a flash drive and the free version) and then you can check that the drives are available to assign as you swap them between bays. Knowing the cases working fine the majority of the time will be of little consolation if you do have a big issue.

Yes, you can hook it to another PC and check the disks. You could do it with your backtrack linux PC if it has reiserfsck on it. Also check the SMART reports to ensure the disks aren't failing. OYou could also put the 2T drives back in, initialize the array again and start over at the point where you begin to replace those drives with the 3T drives. Before going to far I would investigate the system stability and the condition of the 3T drives. Having one red-ball indicates some kind of hardware issue that will continue to cause problems unless you eliminate it.