November 14, 201015 yr i followed this article http://lime-technology.com/wiki/index.php?title=Make_unRAID_Trust_the_Parity_Drive%2C_Avoid_Rebuilding_Parity_Unnecessarily to move around drives except i did not do the last two steps by mistake and i have the parity building now for about 6 hours and i have discovered that i have massive data loss, i was using 1.2TB of the unraid box and it only show me that i am using 760GB.the array was valid and the parity sync was done before i swopped the disks around and added 2 new ones.so too sum up i had a 3 disk array all fine and green and the i moved 1 disk around and added 2 more and then initconfig without the other 2 steps at the bottom of the article, have i lost the 500GB and my wife's only copy of a photo collection :'( or do i have a chance to recover, the rebuild is still busy but i do not see any bytes being added back to the array (i was hoping for this)..... PLEASE HELP ME ..... if you can thanks all sorry forgot to add running v4.5.6
November 14, 201015 yr i followed this article http://lime-technology.com/wiki/index.php?title=Make_unRAID_Trust_the_Parity_Drive%2C_Avoid_Rebuilding_Parity_Unnecessarily to move around drives except i did not do the last two steps by mistake and i have the parity building now for about 6 hours and i have discovered that i have massive data loss, i was using 1.2TB of the unraid box and it only show me that i am using 760GB.the array was valid and the parity sync was done before i swopped the disks around and added 2 new ones.so too sum up i had a 3 disk array all fine and green and the i moved 1 disk around and added 2 more and then initconfig without the other 2 steps at the bottom of the article, have i lost the 500GB and my wife's only copy of a photo collection :'( or do i have a chance to recover, the rebuild is still busy but i do not see any bytes being added back to the array (i was hoping for this)..... PLEASE HELP ME ..... if you can thanks all Step 1. Explain in detail what you did. (I know you tried, but I'm having a tough time figuring it out.) What exactly did you intend to do with the trust-parity procedure. Describe what you mean by "swap disks around" What do you mean by the "rebuild" ? Is the parity drive being written to? Step 2. Attach a syslog to your next post. Capture it BEFORE you next boot or power cycle the server. It sounds as if you are just now in the middle of re-calculating parity based on your currently assigned disks. There should be no file loss from that. If you accidentally assigned a previous data disk to the parity slot, then you are overwriting it with parity calculations. Stop the parity calculation immediately, although it might be WAY too late as you said you are 6 hours into it. That's almost enough time to completely overwrite any disk. Describe in detail your original 3 disk assignments, and where you added/assigned the 2 new ones. Also, in your honor I just edited the "trust-parity" procedure to add the following: It should NEVER be used if you are adding or removing disks in the array. It is ONLY to be used if exactly the same the drives last used to perform a full parity calculation are present and working AND only those exact same drives are currently assigned to the array. If any are missing, or any disks have been added, or deleted, or have failed since the last full parity calculation, this procedure will falsely lead you to think parity protection is in effect when it is not. Joe L.
November 14, 201015 yr Author i have posted a screenshot which i hope helps explain what i am about to say. i had a working system with three drives in it 1 parity (which is definitely still the same as it's my only 2TB drive),my 2 data drives where /dev/md1 and dev/md3 and /dev/md2 was empty, so i read the article and thought i could make the change and ended up with /dev/md1 in the same place and moving /dev/md3 to /dev/md2 and /dev/md3 and /dev/md4 added as new drives and formatted and now the parity build is still going on.... i have attached the syslog file.. please help Joe L. you are my only hope... (not too cheesy i hope) just for info sake i have not rebooted at all since the change. i hope i have answered sensibly sorry had to zip the syslog no space for attachement
November 14, 201015 yr I can see in the syslog the actions you took. It does not look like you did anything really that unusual. The actions you took were identical to those if you had just started up the array with existing drives. From what I can see disk3 and disk4 were just added and formatted. Those are the new disks? You started with this set of disks: Nov 13 09:16:59 Terok-Nor emhttp: pci-0000:00:1f.2-scsi-0:0:0:0 host3 (sdg) ST32000542AS_9XW0CB9E Nov 13 09:16:59 Terok-Nor emhttp: pci-0000:02:00.0-scsi-0:0:0:0 host0 (sdb) WDC_WD1001FALS-00J7B1_WD-WMATV1934324 Nov 13 09:16:59 Terok-Nor emhttp: pci-0000:02:00.0-scsi-0:1:0:0 host0 (sdc) ST3250318AS_9VYALQJH Nov 13 09:16:59 Terok-Nor emhttp: pci-0000:02:00.0-scsi-0:2:0:0 host0 (sdd) WDC_WD1001FALS-55J7B0_WD-WMATV2905264 Nov 13 09:16:59 Terok-Nor emhttp: pci-0000:02:00.0-scsi-0:3:0:0 host0 (sde) WDC_WD1001FALS-00J7B1_WD-WMATV2369948 Nov 13 09:16:59 Terok-Nor emhttp: pci-0000:02:00.0-scsi-0:4:0:0 host0 (sdf) SAMSUNG_HD103SI_S1VSJ9BZ300526 At that point, the array was assigned like this: Nov 13 09:16:59 Terok-Nor kernel: md: import disk0: [8,96] (sdg) ST32000542AS 9XW0CB9E offset: 63 size: 1953514552 Nov 13 09:16:59 Terok-Nor kernel: md: import disk1: [8,48] (sdd) WDC WD1001FALS-5 WD-WMATV2905264 offset: 63 size: 976762552 Nov 13 09:16:59 Terok-Nor kernel: md: disk1 wrong Nov 13 09:16:59 Terok-Nor kernel: md: disk2 removed Nov 13 09:16:59 Terok-Nor kernel: md: import disk3: [8,32] (sdc) ST3250318AS 9VYALQJH offset: 63 size: 244198552 Nov 13 09:16:59 Terok-Nor kernel: md: disk3 wrong Then you started re-assigning disks: Nov 13 09:18:56 Terok-Nor kernel: md: import disk0: [8,96] (sdg) ST32000542AS 9XW0CB9E offset: 63 size: 1953514552 Nov 13 09:18:56 Terok-Nor kernel: md: import disk1: [8,80] (sdf) SAMSUNG HD103SI S1VSJ9BZ300526 offset: 63 size: 976761496 Nov 13 09:18:56 Terok-Nor kernel: md: import disk2: [8,64] (sde) WDC WD1001FALS-0 WD-WMATV2369948 offset: 63 size: 976762552 Nov 13 09:18:56 Terok-Nor kernel: md: disk2 replaced Nov 13 09:18:56 Terok-Nor kernel: md: disk3 missing next you moved /dev/sde to disk3 from disk2: Nov 13 09:19:19 Terok-Nor kernel: md: import disk0: [8,96] (sdg) ST32000542AS 9XW0CB9E offset: 63 size: 1953514552 Nov 13 09:19:19 Terok-Nor kernel: md: import disk1: [8,80] (sdf) SAMSUNG HD103SI S1VSJ9BZ300526 offset: 63 size: 976761496 Nov 13 09:19:19 Terok-Nor kernel: md: disk2 removed Nov 13 09:19:19 Terok-Nor kernel: md: import disk3: [8,64] (sde) WDC WD1001FALS-0 WD-WMATV2369948 offset: 63 size: 976762552 Then you pre-cleared /dev/sdd (WDC_WD1001FALS-55J7B0_WD-WMATV2905264) and /dev/sdc (ST3250318AS_9VYALQJH) Eventually, at the time you typed "initconfig" you had the disks assigned like this: Nov 14 09:10:20 Terok-Nor kernel: md: initializing superblock Nov 14 09:10:20 Terok-Nor kernel: md: import disk0: [8,96] (sdg) ST32000542AS 9XW0CB9E offset: 63 size: 1953514552 Nov 14 09:10:20 Terok-Nor kernel: md: disk0 new disk Nov 14 09:10:20 Terok-Nor kernel: md: import disk1: [8,80] (sdf) SAMSUNG HD103SI S1VSJ9BZ300526 offset: 63 size: 976761496 Nov 14 09:10:20 Terok-Nor kernel: md: disk1 new disk Nov 14 09:10:20 Terok-Nor kernel: md: import disk2: [8,64] (sde) WDC WD1001FALS-0 WD-WMATV2369948 offset: 63 size: 976762552 Nov 14 09:10:20 Terok-Nor kernel: md: disk2 new disk Nov 14 09:10:20 Terok-Nor kernel: md: import disk3: [8,16] (sdb) WDC WD1001FALS-0 WD-WMATV1934324 offset: 63 size: 976762552 Nov 14 09:10:20 Terok-Nor kernel: md: disk3 new disk Nov 14 09:10:20 Terok-Nor kernel: md: import disk4: [8,48] (sdd) WDC WD1001FALS-5 WD-WMATV2905264 offset: 63 size: 976762552 Nov 14 09:10:20 Terok-Nor kernel: md: disk4 new disk Now... of these... which two are your original two data drives? (I'm guessing disk4 was originally one of your two data drives. You need to verify by model/serial number, not by md1, md2 or disk1, disk2.) /dev/sdd is currently assigned as disk4. (/dev/md4) disk3 and disk4 just had new file-systems created on them, which make me suspect that those were the disks you pre-cleared. So... which model/serial number disks were your original two data disks? Joe L.
November 14, 201015 yr Author I can see in the syslog the actions you took. It does not look like you did anything really that unusual. So... which model/serial number disks were your original two data disks?[/b] Joe L. Nov 14 09:10:20 Terok-Nor kernel: md: import disk1: [8,80] (sdf) SAMSUNG HD103SI S1VSJ9BZ300526 offset: 63 size: 976761496 Nov 14 09:10:20 Terok-Nor kernel: md: import disk2: [8,64] (sde) WDC WD1001FALS-0 WD-WMATV2369948 offset: 63 size: 976762552 these are the original 2 disks i had before the whole trouble and agree with you i do not understand where my data went i did not delete anything or have any strange problems except for the idiot things i did...
November 14, 201015 yr I can see in the syslog the actions you took. It does not look like you did anything really that unusual. So... which model/serial number disks were your original two data disks?[/b] Joe L. Nov 14 09:10:20 Terok-Nor kernel: md: import disk1: [8,80] (sdf) SAMSUNG HD103SI S1VSJ9BZ300526 offset: 63 size: 976761496 Nov 14 09:10:20 Terok-Nor kernel: md: import disk2: [8,64] (sde) WDC WD1001FALS-0 WD-WMATV2369948 offset: 63 size: 976762552 these are the original 2 disks i had before the whole trouble and agree with you i do not understand where my data went i did not delete anything or have any strange problems except for the idiot things i did... Then your data is still there... (unless you deleted it) If you think you did delete it, then you can attempt to recover what you deleted, but only if you do not overwrite it with other files first. Have you actually looked at the disk yet to see if your data is there? it is very possible it is just there as expected.
November 14, 201015 yr Author i have looked via the samba shares and i have telnet-ed to the cli and browsed /mnt/disk...etc and looked for it and it's not there. can i run a reiserfs chk or am i thinking wrong, i really have no idea what happened to the data and i did not delete anything it's just weird... the data recovery is for reiser ? can you point me maybe to a starter for data recovery please.
November 14, 201015 yr i have looked via the samba shares and i have telnet-ed to the cli and browsed /mnt/disk...etc and looked for it and it's not there. can i run a reiserfs chk or am i thinking wrong, i really have no idea what happened to the data and i did not delete anything it's just weird... the data recovery is for reiser ? can you point me maybe to a starter for data recovery please. Has the parity calculation completed? (should be close by now if not finished) The basic procedure is described in the here: http://lime-technology.com/forum/index.php?topic=5087.msg47070#msg47070
November 14, 201015 yr Author Array StatusSTARTED; 5 disks in array. Rebuilding Parity Total Size 1,953,514,552 KB Current 608,127,044 (31.1%) Speed 15,132 KB/sec Finish 1481.6 minutes Sync Errors 0 (corrected) as of 1 minute ago still a while before finished so can i MAYBE have some hope that the parity build COULD fix it ? should i let the parity build finish??
November 14, 201015 yr Array StatusSTARTED; 5 disks in array. Rebuilding Parity Total Size 1,953,514,552 KB Current 608,127,044 (31.1%) Speed 15,132 KB/sec Finish 1481.6 minutes Sync Errors 0 (corrected) as of 1 minute ago still a while before finished so can i MAYBE have some hope that the parity build COULD fix it ? Parity is NOT going to fix it. You are currently reading the data disks and WRITING to the parity disk. Therefore, parity, once completed will reflect the current contents of the disks. Forthermore, it was valid ONLY for the prior disk configuration... with three disks. By adding the two additional disks it was no longer valid. It is why it now needs to be re-calculated. By typing "initconfig" you specifically asked unRAID to completely forget its old parity information and immediately invalidate it. Have you checked the OTHER data disk. I know you think the photos were on one disk, but perhaps they were actually on the other.
November 14, 201015 yr Author Have you checked the OTHER data disk. I know you think the photos were on one disk, but perhaps they were actually on the other. yes i have checked my 2 original data disks and the others have nothing on their filesystems so can i stop the rebuild and run this (can i run this in a screen session) I have managed to accidently delete a whole bunch of files. The disk hasn't been written to since. Is there a way/program to help me recover them? Were they on your unRAId server? If yes, then see this entry in the wiki: http://lime-technology.com/wiki/index.php?title=FAQ#How_can_I_undelete_files_from_an_unRAID_disk.3F In essence stop samba [all your shares will disappear from network] killall smbd nmbd un-mount the disk (md1=disk1, md2=disk2, md10-disk10, etc. Use the correct disk for your needs. this example is for disk1) umount /dev/md1 note the above command is umount, not unmount. run reiserfsck on the disk to rebuild the file tree. Any files that can be recovered will. This process could take many hours for a large disk as the entire disk will need to be scanned. reiserfsck --rebuild-tree -S /dev/md1 It will ask you to confirm the rebuild of the file tree. Make certain you answer Yes (three letters, Capital "Y", lower case "es") Anything else will be considered as "no" re-mount the disk (again, use the correct "md" and "disk" for your situation) mount /dev/md1 /mnt/disk1 re-start samba /usr/sbin/smbd -D /usr/sbin/nmbd -D your deleted files (any that could be recovered) will be in a lost+found folder on the disk. See here for a more general description of the process http://antrix.net/journal/techtalk/reiserfs_data_recovery_howto.comments
November 14, 201015 yr DO NOT STOP THE PARITY CALCULATION.. DO NOT STOP THE ARRAY EITHER. It has nothing to do with what you will be doing, and it will not interfere with your actions trying to recover your files. Joe L.
November 15, 201015 yr Author DO NOT STOP THE PARITY CALCULATION.. DO NOT STOP THE ARRAY EITHER. It has nothing to do with what you will be doing, and it will not interfere with your actions trying to recover your files. Joe L. well i think it's too late for that i ran the rebuild last night and at 5 this morning this is the current output and i have no web interface or way to see the current array status.. so i think the array is stopped or else i would have seen the webpage, so i am gonna go to work and leave it running, hope it's the right move. oh i did reboot last night as /dev/md1 was stuck unmounting for about 1 hour when i treid to run the reiser rebuild so i did a reboot command at the cli. root@Terok-Nor:/boot/unmenu# screen -r ** 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. ** ** If you are using the latest reiserfsprogs and it fails ** ** please email bug reports to [email protected], ** ** providing as much information as possible -- your ** ** hardware, kernel, patches, settings, all reiserfsck ** ** messages (including version), the reiserfsck logfile, ** ** check the syslog file for any related information. ** ** If you would like advice on using this program, support ** ** is available for $25 at www.namesys.com/support.html. ** ************************************************************* Will rebuild the filesystem (/dev/md1) 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 '/dev/md1' in blocks [18..8211]: 0 transactions replayed ########### reiserfsck --rebuild-tree started at Sun Nov 14 22:30:16 2010 ########### Pass 0: ####### Pass 0 ####### The whole partition (244190374 blocks) is to be scanned Skipping 15663 blocks (super block, journal, bitmaps) 244174711 blocks will be read 0%.... left 130875138, 4607 /seccc
November 15, 201015 yr A disk cannot un-mount if it is "busy" A disk is busy if a file on it is open for reading or writing. A disk is busy if it is the current directory for a program. If you change directory to /mnt/disk1, it therefore makes disk1 busy. It cannot be un-mounted. Good luck with your search for your files.
November 15, 201015 yr Author First let me say thank you VERY VERY much for all your help Joe. But all is lost i'm afraid it managed to find the movies but not 1 photo... weird... So be it... my stupidity not unRAID's so i take this as a lesson learned and onto the next
November 15, 201015 yr First let me say thank you VERY VERY much for all your help Joe. But all is lost i'm afraid it managed to find the movies but not 1 photo... weird... So be it... my stupidity not unRAID's so i take this as a lesson learned and onto the next you might try this command and see if they are somewhere you did not look: find /mnt -type f -iname *.jpg -print
Archived
This topic is now archived and is closed to further replies.