April 24, 201412 yr Use /dev/md4 to keep parity in sync. You can run reiserfsck or an equivalent recovery utility on the image file. rieserfsck requires the image be mounted first. Thanks for the response!!! So you are suggesting using reiserfsck or something equivalent on /dev/md4 (virtual device in UnRAID) as opposed to /dev/sdb1 (The partition on the new 3TB backup drive where the block level copy of my troubled 2TB drive is)? The only reason I ask is I wasn't sure if the backup partition I made of the troubled disk on the new 3TB would be known as /dev/md4 as It only shows as /dev/sdb1 in Partimage. On the server use /dev/md4 whenever possible. Reading from /dev/sdb1 is also ok but make writes to /dev/md4 if possible. The image can be worked on in unRAID but using a Windows or Linux system will be easier and allow more options. When using DD, I did not create an image but just copied the entire partition from the troubled drive to the new 3TB backup drive. That shouldn't matter, correct? I am assuming it's as easy as just mounting the backup partition. This is an image. Thank you again for all of your help with this! I just want to make sure I don't do the wrong thing and mess up what I have done so far. Your help has been greatly appreciated!!! NP.
April 24, 201412 yr Author On the server use /dev/md4 whenever possible. Reading from /dev/sdb1 is also ok but make writes to /dev/md4 if possible. The image can be worked on in unRAID but using a Windows or Linux system will be easier and allow more options. Okay great! I hate to be a pain, but I just want to make sure that i'm doing everything okay. So right now I have the 3TB drive with the backup image hooked up to my computer, which is currently booted up with SystemRescueCd. The image I made onto the 3TB drive is on the partition "/dev/sdb1". Should I run "reiserfsck --rebuild-tree --scan-whole-partition against this partition to perform recovery? Also, I think I may know why this whole mess happened, and if it is why, then I am truly stupid. I recently sold one of my 2TB drives in my server that I wanted to replace since it was getting old. Before selling it, I copied all of the data over to an empty 2TB drive in my UnRAID server that had been pre-cleared. After removing the disk I sold, I booted up UnRAID with the array unstarted, and then set this drive that I'm now having trouble with to the location the old disk was at (Disk 4). I (stupidly) did not run a parity check at this point, and forced UnRAID to accept the parity as valid since I figured the data was going to be exactly the same, just a different physical hard drive. I'm not sure if that would have caused this whole debacle of that drive becoming disabled after the non-correcting parity check, but if so, then I would imagine the data on that drive is still okay (hopefully). Again, I'm just throwing this out there in case I am wasting time doing all of these recovery steps for no reason. If that is the case, then I wonder if I've messed anything up by running "reiserfsck --rebuild-tree --scan-whole-partition" on MD4. I feel like I did not do this right, and maybe that is why I'm having this problem in the first place? I should have mentioned this first, but I've been so worried about losing data, coupled with a few stressful weeks at work, that I forgot all about this...
April 25, 201412 yr On the server use /dev/md4 whenever possible. Reading from /dev/sdb1 is also ok but make writes to /dev/md4 if possible. The image can be worked on in unRAID but using a Windows or Linux system will be easier and allow more options. Okay great! I hate to be a pain, but I just want to make sure that i'm doing everything okay. So right now I have the 3TB drive with the backup image hooked up to my computer, which is currently booted up with SystemRescueCd. The image I made onto the 3TB drive is on the partition "/dev/sdb1". Should I run "reiserfsck --rebuild-tree --scan-whole-partition against this partition to perform recovery? Absolutely not. reiserfsck --rebuild-tree --scan-whole-partition image.file. See here: http://www.cgsecurity.org/wiki/ReiserFS_File_Undelete_HOWTO Also, I think I may know why this whole mess happened, and if it is why, then I am truly stupid. I recently sold one of my 2TB drives in my server that I wanted to replace since it was getting old. Before selling it, I copied all of the data over to an empty 2TB drive in my UnRAID server that had been pre-cleared. After removing the disk I sold, I booted up UnRAID with the array unstarted, and then set this drive that I'm now having trouble with to the location the old disk was at (Disk 4). I (stupidly) did not run a parity check at this point, and forced UnRAID to accept the parity as valid since I figured the data was going to be exactly the same, just a different physical hard drive. Parity was trashed at this point. Parity is still trashed. You can pretend that the parity is is not present for the rest of this discussion. I'm not sure if that would have caused this whole debacle of that drive becoming disabled after the non-correcting parity check, but if so, then I would imagine the data on that drive is still okay (hopefully). Again, I'm just throwing this out there in case I am wasting time doing all of these recovery steps for no reason. If that is the case, then I wonder if I've messed anything up by running "reiserfsck --rebuild-tree --scan-whole-partition" on MD4. I feel like I did not do this right, and maybe that is why I'm having this problem in the first place? I should have mentioned this first, but I've been so worried about losing data, coupled with a few stressful weeks at work, that I forgot all about this...
April 26, 201412 yr Author Absolutely not. reiserfsck --rebuild-tree --scan-whole-partition image.file. See here: http://www.cgsecurity.org/wiki/ReiserFS_File_Undelete_HOWTO Thanks again as always! Here is where I am at now after all of your help and suggestions... [*]I created an image of the troubled 2TB drive to the 3TB backup drive by typing "dd if=/dev/sda1 of=/mnt/recovery/reiserfs.dd conv=noerror" in terminal. [*]I ran reiserfsck against the "reiserfs.dd" image I created by typing "reiserfsck --rebuild-tree --scan-whole-partition" in terminal. I attached the reiserfsck log that was created after the check was done, and also inserted it as code below. I'm not sure what to make of these results. It looks like it did correct some things, but again before I move forward with anything, I will wait to see what the expert opinion is root@sysresccd /root % reiserfsck --rebuild-tree --scan-whole-partition /mnt/recovery/reiserfs.dd reiserfsck 3.6.24 ************************************************************* ** Do not run the program with --rebuild-tree unless ** ** something is broken and MAKE A BACKUP before using it. ** ** If you have bad sectors on a drive it is usually a bad ** ** idea to continue using it. Then you probably should get ** ** a working hard drive, copy the file system from the bad ** ** drive to the good one -- dd_rescue is a good tool for ** ** that -- and only then run this program. ** ************************************************************* Will rebuild the filesystem (/mnt/recovery/reiserfs.dd) tree Will put log info to 'stdout' Do you want to run this program?[N/Yes] (note need to type Yes if you do):Yes Replaying journal: Done. Reiserfs journal '/mnt/recovery/reiserfs.dd' in blocks [18..8211]: 0 transactions replayed ########### reiserfsck --rebuild-tree started at Fri Apr 25 21:41:36 2014 ########### Pass 0: ####### Pass 0 ####### The whole partition (488378638 blocks) is to be scanned Skipping 23115 blocks (super block, journal, bitmaps) 488355523 blocks will be read 0%.. left 445389858, 34705 /se left 0, 30505 /sec 694 directory entries were hashed with "r5" hash. "r5" hash is selected Flushing..finished Read blocks (but not data blocks) 488355523 Leaves among those 480586 Objectids found 700 Pass 1 (will try to insert 480586 leaves): ####### Pass 1 ####### Looking for allocable blocks .. finished 0%....20%....40%....60%....80%....100% left 0, 69 /sec Flushing..finished 480586 leaves read 480584 inserted - pointers in indirect items pointing to metadata 85 (zeroed) 2 not inserted non-unique pointers in indirect items (zeroed) 1266 ####### Pass 2 ####### Pass 2: 0%....20%....40%..vpf-10260: The file we are inserting the new item (659 663 0x1 DRCT (2), len 1000, location 3096 entry count 65535, fsck need 0, format new) into has no StatData, insertion was skipped ..60%....80%....100% left 0, 0 /sec Flushing..finished Leaves inserted item by item 2 Pass 3 (semantic): ####### Pass 3 ######### ... os/Films/Songs from the Second Floor (2000)/songs.from.the.second.floor.mkvvpf-10680: The file [664 666] has the wrong block count in the StatData (5445120) - corrected to (8245304) Flushing..finished Files found: 504 Directories found: 192 Pass 3a (looking for lost dir/files): ####### Pass 3a (lost+found pass) ######### Looking for lost directories: Looking for lost files:6 /sec vpf-10680: The file [632 634] has the wrong block count in the StatData (28934096) - corrected to (16) Flushing..finished Objects without names 1 Files linked to /lost+found 1 Pass 4 - finished done 480571, 27 /sec Deleted unreachable items 1801 Flushing..finished Syncing..finished ########### reiserfsck finished at Sat Apr 26 08:59:02 2014 ########### I truly can't thank you enough for all of your assistance! You've been more than helpful and I really do appreciate it! reiserfsck_report_-_2014-04-26.txt
April 27, 201412 yr Now mount the disk and examine its contents as described on the cgsecurity page. Repeat the process on /dev/md2 and the results should be the same. Then run 2 parity checks.
April 28, 201412 yr Author Now mount the disk and examine its contents as described on the cgsecurity page. Repeat the process on /dev/md2 and the results should be the same. Then run 2 parity checks. Thanks again! So I mounted the backup image and it looks like there is a "lost+found" directory, as noted in the reiserfsck report I posted in my previous post... ####### Pass 3a (lost+found pass) ######### Looking for lost directories: Looking for lost files:6 /sec vpf-10680: The file [632 634] has the wrong block count in the StatData (28934096) - corrected to (16) Flushing..finished Objects without names 1 Files linked to /lost+found 1 Now, when you say repeat the process on /dev/md2, did you mean to say md4 as that is the troubled device is in my UnRAID setup? Just clarifying to see if you meant md4, or if you really did mean md2. I hope you don't think I'm second guessing, I'm just trying to make sure I'm doing everything right Much more importantly though, I am not sure the exact process for mounting the aforementioned device in UnRAID to check its contents, as that is what I am interpreting to do from your instructions. I did find a previous thread (http://lime-technology.com/forum/index.php?topic=26952.msg236340#msg236340), but I don't think this is what I'm looking for at all. Then, you can see if the first partition on the disk can be mounted. Either reboot and see if the disk mounts, or type mkdir /mnt/disk4 mount -t reiserfs -r /dev/sde1 /mnt/disk4 (note the device is /dev/sde1 with a "1" at the end of the name denoting the first partition) If the mount is successful, you can then reboot by typing reboot and see if unRAID can mount it. (No need to stop the array before the reboot this time, since it is already stopped) You've been so helpful and it looks like I'm getting close to the end of this! I know I keep saying it, but I'm very thankful for all of your help. I look forward to hearing back whenever you have the time
April 28, 201412 yr Run it on md4. See Check Disk filesystems in my sig but run the command with rebuild-tree and scan-whole-partition.
April 29, 201412 yr Author Run it on md4. See Check Disk filesystems in my sig but run the command with rebuild-tree and scan-whole-partition. Okay great! I remember you telling me to run "reiserfsck --rebuild-tree --scan-whole-partition /dev/md4" a couple of weeks ago in UnRAID while I originally created an image (See my reply with the MD4 reiserfsck report and UnRAID syslog if helpful). Should I run it again? Is there any risk in doing so? Just want to make sure before doing anything irreversible. Thanks again!!
April 29, 201412 yr Has the same process been run on both /dev/md4 and reiserfs.dd? Are the contents of reiserfs.dd and md4 identical, currently? If so, we are done.
April 29, 201412 yr Author Has the same process been run on both /dev/md4 and reiserfs.dd? Are the contents of reiserfs.dd and md4 identical, currently? If so, we are done. Well, I noticed from the reiserfsck report after the check on MD4 in UnRAID that there were no "lost+found" files created, as was seen after the reiserfsck on the backup image, or at least from what I can surmise. Also, there are a bunch of entries that came up during pass #2 on MD4 that did not come up during the check of the image. Not sure if this is a problem? I did not get a full log of the reiserfsck on MD4 because I forgot to turn on logging in PuTTY, but I did snag what was on screen before closing the session. I've attached both reports to this post for comparison if at all helpful. I guess the best thing would be to manually check the contents of MD4 to see. Is there an easy way to browse MD4 in UnRAID? Is it even possible to do so? Thank you again for all your help! I couldn't have done any of this without your assistance! 2014-04-15_-_reiserfsck_report_MD4.txt 2014-04-26_-_reiserfsck_report_Backup_Image_.txt
April 29, 201412 yr Author md4 is disk 4. Share disk 4. Thanks! This should probably be obvious, but does this mean I need to put my original drive back into my server, assign the disk back to Disk 4, and then start up the array? I'm very sorry for being so difficult
April 30, 201412 yr I guess the best thing would be to manually check the contents of MD4 to see. Is there an easy way to browse MD4 in UnRAID? Is it even possible to do so? md4 is available as the disk4 share (unless you have disabled that share).
April 30, 201412 yr If 'md4' is physically in the array, and you just want to step through the files, itimpi's correct. In unRAID a 'drive share' is done from the main GUI page. click on the drive, then on SMB Security Settings, set the option to 'Export = Yes' then to access it from your PC, something like "smb://192.168.1.50/disk4". (that's my Mac 'Connect to Server...' string. Use your tower address. On a PC, i think 'disk4' will show up in the network drop down on the left of windows explorer window.) ummm...but then I got to thinking about your question...and I realize I don't really quite understand where you're at. You asked, 'Should I put the old md4 back into the array...' and 'manually check the contents of md4'... If you have a drive physically OUTSIDE OF THE ARRAY and you just want to see what's on it, you can mount it with a regular PC...you just need the Reiserfs driver loaded on the PC. BE CAREFUL TO NOT WRITE TO THE DISK...if you write to it, you'll destroy any hopes of using that disk as part of a parity recovery operation.
April 30, 201412 yr Author Thanks everyone! I guess the confusion here is due to the fact that in the UnRAID GUI, Disk 4 is just marked as "Disabled, no disk present" since the disk is no longer connected to my server. I'm not sure if there is even a possibility of accessing the share now that it is marked like that. Checking the files on the physical disk might not tell me anything new as the reiserfsck was run on MD4 in UnRAID while that physical disk was not present. These are the steps I have taken so far. [*]Purchased new 3TB Western Digital Green HDD to backup image of my disabled disk to [*]Plugged both the new drive and the problem UnRAID drive into my HTPC [*]Created bootable USB drive for SystemRecoveryCD and booted into "StartX" [*]Opened up GParted and formatted the new WD 3TB drive as ResierFS file system [*]Created a folder within /mnt called "recovery" (/mnt/recovery/) and mounted it to the reiserFS formatted partition on the new 3TB drive via the line "/dev/sdb1 /mnt/recovery" in Terminal [*]Created an image of the troubled 2TB drive to the 3TB backup drive by typing "dd if=/dev/sda1 of=/mnt/recovery/reiserfs.dd conv=noerror" in terminal. [*]The image was created and can be found in "/mnt/recover/reiserfs.dd". [*]I ran reiserfsck against the "reiserfs.dd" image I created by typing "reiserfsck --rebuild-tree --scan-whole-partition" in terminal.". [*]Mounted the backup image and it looks like there is a "lost+found" directory, as noted in the reiserfsck report. This leads me to believe there was a file that was recovered outside of its original directory. [*]The results of the reiserfsck --rebuild-tree --scan-whole-partition" on both MD4 in UnRAID and the backup image I created of the physical disk do not seem to be identical. This is where I would like to mount the MD4 (Disk 4 in UnRAID) so I can see what changes the reiserfsck --rebuild-tree --scan-whole-partition" on MD4 made to the virtual device. This is where I am currently stuck. I hope that I've completed all the proper steps correctly so far, but this is all so new to me that I've no real experience with such an issue
May 1, 201412 yr Is the array started? It will run with one missing or disabled disk. Start the array. Click on the disk 4 link on unRAID main and enable sharing.
May 1, 201412 yr Author Is the array started? It will run with one missing or disabled disk. Start the array. Click on the disk 4 link on unRAID main and enable sharing. Wow, I feel stupid as it was because I had the array in maintenance mode (which does not mount disks). Anyway, I navigated to Disk 4 in Windows Explorer and it looks like it has one "lost+found" directory, and it is completely empty. From what I can tell, all of Disk 4 is empty after completing the "reiserfsck --rebuild-tree --scan-whole-partition" on MD4. The entire file tree structure is also gone, which I have attached an image of the UnRAID GUI to illustrate this. I'm not sure what this means, but it does not seem good as I'm assuming it should reflect the results of the "reiserfsck --rebuild-tree --scan-whole-partition" I ran on the backup image. I am assuming I will also need to do some more intensive recovery, but I will again wait to see what the experts here suggest as you've all been so helpful!
May 1, 201412 yr Author What about the rest of disk4? What is NOT inside the "Lost and Found" folder? The "Lost and Found" folder is actually the only folder on the entire disk. I used to have a share on the disk called "Media" with a number of sub-folders inside it, but now it's strictly only this "Lost and Found" folder at the root of the disk. I've attached another image to illustrate it in Explorer. UnRAID GUI is also stating that Disk 4 has 94.7 MB Used and 2TB Free, so it sounds like it's empty
May 1, 201412 yr OK, that is unexpected, but since the data has been recovered, not worth persuing. Mount the image file that you created in unRAID and copy the data to an array disk.
May 1, 201412 yr Author OK, that is unexpected, but since the data has been recovered, not worth persuing. Mount the image file that you created in unRAID and copy the data to an array disk. Seems easy enough! Not sure why the data is not there, but I'm not worried about it. I guess my main question is regarding the best way to mount this image in UnRAID to copy the data over. Currently the image of the original drive is located on the backup 3TB drive I purchased. Is there an easy way to add the drive to UnRAID without making it a part of the array in order to copy over that data? Also, I do have an empty 3TB drive in my UnRAID array, but instead of putting a lot of reads/writes on it, would it be possible to just write to the 2TB drive that I was originally having issues with? Another question I have is since the data all appears to be intact on the original 2TB drive when I browse it in my Windows PC using DiskInternals Linux Reader, would it be possible to just put the 2TB drive back in my array and assign the drive back to Disk 4 while retaining that data? Since the data on MD4 is already seemingly gone, I just wanted to throw this out there to see if it is not only a possibility, but wise to do so. Sorry for all the questions... I know I am full of them! I figure there are probably multiple ways to approach this, but I will do whatever is suggested as I am no expert when it comes to this. Thank you again so much... I know this has been on-going for weeks and I apologize for clogging up the support forums!
May 2, 201412 yr Start unRAID with the 3T disk installed. Mount the disk: http://lime-technology.com/wiki/index.php/Copy_files_from_a_NTFS_drive Then mount the image file: mkdir /mnt/image mount -t reiserfs -r image.dd /mnt/image Then copy the contents to an array disk, e.g., /dev/md4. http://lime-technology.com/wiki/index.php/Transferring_Files_Within_the_unRAID_Server
May 3, 201412 yr Author Start unRAID with the 3T disk installed. Mount the disk: http://lime-technology.com/wiki/index.php/Copy_files_from_a_NTFS_drive Then mount the image file: mkdir /mnt/image mount -t reiserfs -r image.dd /mnt/image Then copy the contents to an array disk, e.g., /dev/md4. http://lime-technology.com/wiki/index.php/Transferring_Files_Within_the_unRAID_Server Thanks for the info! I actually ended up formatting the drive as reiserfs instead of NTFS, but I assume the instructions would just be the same, minus the NTFS driver step?
May 3, 201412 yr Author Yes. Mount the disk as needed. Thanks again! I did however just think I'd something though, and that is that I have no open SATA ports left on my server. Is it possible for me to just temporarily use the port I had my troubled 2TB drive connected to before taking it out? If so and I can still write to disk 4, then I should be okay, hopefully! Sent from my SCH-I535 using Tapatalk
Archived
This topic is now archived and is closed to further replies.