lionelhutz

Well, I should have read up on RAID5 again. RAID5 doesn't use the parity data when reading back a disk so it wouldn't automatically recover from bit rot. However, if it did use the parity on reads it could recover from bit rot. I wonder it any of the better controllers could be set to check parity on reads?

I suppose bad sectors could be considered bit rot as it's defined. I have seen hardware issues posted here that caused bad data, but that wasn't really bitrot, or bad data caused by the HDD platter deteriorating over time. The bad data was caused by bad electronics not processing or transferring the bits correctly. All the parity error isues posted seem to be are caused by things like the previous paraghraph or were caused by things like the upgrade bug or things like hard-powering the server off. People with well-built stable servers don't seem to have any issues with an unexplained parity error or two just randomly popping up every so often. The other linked thread is wrong about RAID5. RAID5 can reconstruct from a bit error, since the data is stored in stripes, not disk by disk, and the stripe can be reconstructed. unRAID with the disk by disk protection could not reconstruct the error.

Yup, just turn-on user share and start the array. The shares will appear and you can then set them up.

Set the free space the same for every share. This will keep the disks that free, if it is allowed by the split level. Remember, split level over-rides all other settings as far as when to choose a different disk. You Music share should use split level = 1 to keep each artist together. And yes, move some of the things off disk1 and disk2 to disk3 to free-up space. I'd recommend any continuing TV seasons get moved first or else the disks will get filled again.

You have to tell the directory structre since the split level forces certain directories to stay on a single disk. You TV share could easily be set to keep a whole season on a single disk. So, no matter what other settings you change or how many reboots you try, the partially existing TV season that you're trying to complete will stay on the same disk it started to use. The high water does not use a fixed cut-off of 60gig. The 60gig number might have been given as an example but that is all. Not sure where you're reading but try the unofficial manual for a description of the high water. Free space is the free space which is left per disk. Use "Most Free" if you want the most free disk used whenever possible (when allowed by the split level).

Copying from one to another share will over-ride the split level and allocation method. The data will stay on the same array disk if the share is using that disk. In other words, if the files exist on disk2 before you copy them to the TV share they will stay on disk2.

Personally, I'd keep the shares seperate. No, there is not an include for each directory. Every top level directory will be a share. There is a spot where you can tell it which disks to use for each share. It sounds like your split level choices are OK. You added the extra Genre level in the music so to the split level to get the same result. In this case 2 would still keep each artist together on a single disk. The user shares won't cause data loss. If you don't like them then turn them off again. And yes, they are easier but you must understand what your settings mean and that certain settings can lead to full disks. For example, adding TV series to fairly full disks with a split level of 1 means the whole series stays on a disk. Start with 100gig free on the disk and add a series which is 120gig in total and you will get a disk full error. unRAID won't start using a new disk just because that one is full. You set it to keep the series together and it either does so or fails trying.

In this situation a split level of 2 will allow the Show Folder to be created on each of your data drives. What is causing your shows to get moved all over the place is your MOST FREE allocation method. With this method the show gets put on the drive that has the MOST FREE space. If you want to keep shows on one drive for as long as possible consider changed to HIGH WATER which will fill the drive to a pre determined point before rolling to the next drive. This is not true. Level 2 allows the show folder to be created on multiple drives but does not allow the season folder to be created on multiple drives. Therefore, each season should be together on a single drive. Once again. We need to know how the data ends up in the share??

I took a quick look and it appears the Dune can support multiple network drives. So, as a first step, make sure the directory name on each disk for each media type is exactly the same. eg, "TV Shows" on disk4 is spelt exactly like "TV Shows" on disk7. Then, just turn on user shares. You should get a new user share on the network for every existing level directory and the shares page should list the settings for each of these shares. Now, if you want to copy files to the shares you need to set some settings for each one. First step would be to assign the disks each one is allowed to do. Use the format "disk1,disk2,disk3" to include or exclude those disks. Music would use disk1,disk5". You don't need to fill out both, just the include is enough. Second would be the allocation method. I recommend either high water or most free. Fill-up can create problems. The split level would also need setting for the shares using more than one disk. Movies would use 1 and TV could use 1 or 2. 1 keeps the whole show together on a single disk and 2 keeps each season together on a single disk. Music use 1 to keep each artist on a single disk. If you continue to use the disk mapping to copy data to the server then ignore the above paragraph and just keep doing what you're doing. The share settings only matter if you copy data to the share. Try that and see how it goes. Monitor the disks as you copy the first data to make sure it goes where you want and how you want.

How did these episode files get on the drives?

OK, the main thing you seem to have missed is that the split level is dependant on the folder structure. You MUST describe the directory structure if you want to verify your split level setting. Without the folder structure we are guessing. You MUST fisrt know the directory structure you will use before picking a split level. The manual is telling you how the split level works for 3 different directory structure cases (examples). Each case has a directory structure illustration followed by an explanation of the possible split levels and their effects. The effect of the split level will change because the directory structure has changed. Basically, the split level does over-ride the allocation method. The split level will force a certain disk for certain files even if the allocation method calls for a different disk.

I have a 1.5T EADS drive that has been acting up. It threw 3 UNC sector errors during the last parity check. No pending or re-allocated sectors were in the SMART report. I cleared off the data and ran a preclear on the drive 4 times. #1 pass - showed 3 bad sectors different than the parity check and they were detected during the post read. #2 pass - showed 2 bad sectors same as above with the 3rd one disappearing. #3 pass - showed 0 bad sectors. #4 pass - remained at 0 bad sectors. In total, I picked out 6 different bad sectors being detected. Could the drive not be reporting that these sectors were re-allocated? I'm not sure what to do with this drive now because I don't trust it but there is no reason to RMA it. Has anyone else RMA'd a seemingly good WD and got a new one without any hassle? I can try the WD tools next I suppose but the drive appears fine at the moment so I don't expect any results to justify an RMA. I can post the syslog from all of the above if anyone wants to see it. I'm running on an older ECS AMD 740 chipset board with a Corsair 400W power supply and 8 x hotswap bays. About the only other issue I've had is some DMA or CRC (interface CRC errors) at one time that some new cables seemed to help. The server hasn't even been opened or moved for well over a year until I just blew the dust out of it.

Nice, I'm thinking you should have make a new rc4 board though to keep the issues of each release seperate.

I suggest you look at post 15 and run the commands there. You need to confirm the partition is correct before doing any file system checks. http://lime-technology.com/forum/index.php?topic=15385.15

Just to add another option. You could go to the user share page and modify the schedule. The last * means move every day. You can change it to a number 0 to 6 corresponding to days of the week starting with Sunday. So, since it's Saturday you could use 5 meaning it will only run on Fridays. Put it back to a * to get it to run every day again. No reboot required. http://en.wikipedia.org/wiki/Cron#Predefined_scheduling_definitions

The mover schedule is on the shares page when you have a cache drive.

I use a 500gig drive. I allow the shares to use the cache drive so I process and put all the files onto the cache drive first and then I have the mover set to move the files to the array once a week. I haven't come close to filling it yet. I seem to typically download between 150gig and 250gig a month. Looking at Newegg, it's like $5 more to step from say a 160gig drive to either a 250gig or 500gig drive. So, it doesn't make much sense to go smaller than 500gig. I guess you could also go with a 60gig SSD for about the same cost if you move directly to the array when processing. Otherwise, try to find a used or clearance disk and hope for the best. Maybe try a local computer shop and see if they have an old drive lying around they'd sell to you cheap.

FYI, moving share to share means the files stay on the same disk regardless of the allocation and split level settings on the share.

Told ya, as soon as you tried to get it working and then posted what you unsuccessfully tried that you'd get help. Peter

Skank I have no idea what plink is. Did you read the documents for it or look up any examples? You likely have to just answer yes and carry on. I believe you had a script or something running on that media player you are using (I don't recall the details and really don't care) which would do the same as pressing the sleep button on the unMENU interface. But, you complained that the sleep button didn't work right because your player was hanging up. So, try editing the script for the sleep button and using the line Joe gave you instead of what's there now. I thought this was suggested in the other thread where you were asking for it. I'm pretty sure the script is in the unmenu folder on the flash drive. The name should be fairly obvious. Be careful when editing the script to use unix EOF and not DOS ones. Try telneting to the server and type mc to use Midnight Commander to access the server files and do the editing. I hate to tell you, but people here (or anywhere for that matter) generally don't want to do the work for you when it appears you won't put in any effort yourself. I think your problem of not seeming to get help stems from that fact you were told what to change but it appears you never followed up on it or tried to change it. If you'd come back after making an effort with a question about what you had tried I'm sure you would have got more help. Peter

Agreed. If it's not due to another onboard controller then it appears some newer LGA1366 boards are having issues with them. But then you are having issues on an older P35 chipset board too. So far, no reports of AMD systems having trouble with it. Peter

SATA2 Serial ATA II PCI-Express RAID Controller Card (JMicron JMB362) Requirements: PCIe x1 or faster slot Cables: Standard SATA (1 port per cable), such as Where to get them: http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=280429470355 Special Instructions: None, should be plug'n'play. Thanks! Added to the first post. - Raj

My main reason for using 3.5" is that the largest size 2.5" drive has typically been about the same cost as a 3.5" drive that is double in size. Or, in other words, you pay about 2x for the same storage capacity. It then costs again for more SATA ports too. I'm sure with a cheaper power supply and swap trays this could balance out somewhat but the ports cost money and the drives are a large part of the server costs. Peter

OK, I just installed it and put cache_dirs -w in my go script and drives were still spinning up the same as before. So, I stopped it and then ran it at the prompt like - cache_dirs -F -v The below is the result; Executing find /mnt/disk2/Car_Music -noleaf Executing find /mnt/disk1/Movies -noleaf Executing find /mnt/disk2/Music -noleaf Executing find /mnt/disk1/New_Movies -noleaf Executing find /mnt/disk2/New_Movies -noleaf Executing find /mnt/disk2/Pictures -noleaf Executing find /mnt/disk2/Storage -noleaf Executing find /mnt/disk2/TV_Shows -noleaf Executing find /mnt/disk3/TV_Shows -noleaf Executed find in 0.287845 seconds, weighted avg=0.289340 seconds, now sleeping 6 seconds So, it appears to be going through all the directories and it did not spin up any drives the first pass when run on the comand line so it also appears the version running from the go script was working. I left it last night and the drvies all spun down sometime early this morning and stayed spun down while I stopped it and ran it from the command line. While it was running as above, the drives still spun up if I tried to navigate to almost any directory on them. I could go to the top level of the shares but that was about it. As I understand it, it will scan to a default maxium directory depth of 99 and I don't have anything beyond about 4 or 5 levels. I am using W7 Windows Explorer. I specifically set the default view to "List" which should not try to gather any specific info about the file or create a thumbnail of the file. I also turned off the "Details" pane and the "Preview" pane as well. Actually, the preview was off already. I then tried the -M 1 option and it seems a little better but still spins up disks.also I also tried -p 1 option but not much change. However, after both of these options it seemed to me there was a higher chance the disk would spin up when I pressed the back button while surfing directories. But, I still didn't notice much if any pattern to it. So, anyone have any ideas? Is this a Windows Explorer thing? Anyone else know if W7 Windows Explorer will still read file info no matter what. Peter

Has capacity increased more than reliability? Is he not using URE out of context? Say you replace a data drive and another drive (not the parity) fails during the 'rebuild' of the new drive. Can we put the old drive back in and then force Unraid to recognize parity as correct using the restore button and 'mdcmd set invalidslot 99'? Will this not then get us back to a position where the failed drive could be rebuilt? I'd guess the best way would be to move the data off the drive then write 0's to the drive you want to remove. Remove the drive and then use the restore button and 'mdcmd set invalidslot 99' to remove the drive from Unraid. That should keep the array protected during the whole process. Peter