dark_avenger Posted December 23, 2011 Posted December 23, 2011 Shutdown the server today to add in another drive and upon restarting one of the drives has now appeared as unformatted. Was working before the restart and should have 300GB+ of data on it. I have attached a screenshot. Any suggestions?
dark_avenger Posted December 23, 2011 Author Posted December 23, 2011 Syslog attached syslog-2011-12-23.txt
Joe L. Posted December 23, 2011 Posted December 23, 2011 first... DO NOT PRESS THE FORMAT BUTTON. basically, the disk could not be mounted. To unRAID, any disk it cannot mount, for whatever reason, is displayed as unformatted: Dec 23 16:45:28 Storage logger: mount: wrong fs type, bad option, bad superblock on /dev/md3, We've seen a bug in 4.7 unRAID where the MBR on disks are re-written, when disk configurations change, and when re-written no longer point to the correct start of the file-system. When this occurs, the disk will not mount, and the superblock cannot be found. (since the MBR is pointing to the wrong place) Fortunately, when this occurs, the fix is easy... point the MBR to the correct start of the file-system. You can see how to check if this has happened to your disk, and how to fix it here: http://lime-technology.com/forum/index.php?topic=15385.msg144723#msg144723 You'll need to use the correct /dev/[hs]dX device for the initial "dd" command, and then go from there. (your disk3 is /dev/hda ) so... start with fdisk -lu /dev/hda and dd if=/dev/hda count=195 | od -c -A d | sed 30q and then go from there. (post the results) If you see the line with "R e I s E r 2 F s" at address 0097840, then your MBR needs to be set to point to a partition start on sector 63. If you see the line with "R e I s E r 2 F s" at address 0098352 , then your MBR needs to be set to point to a partition start on sector 64.
dark_avenger Posted December 23, 2011 Author Posted December 23, 2011 Thanks for the quick response. I ran both the commands as posted. fdisk -lu /dev/hda returned 1 heads, 63 sectors/track, 12403536 cylinders, total 781422768 sectors Units = sectors of 1 * 512 = 512 bytes Disk identifier: 0x00000000 Device Boot Start End Blocks Id System /dev/hda1 63 781422767 390711352+ 83 Linux Partition 1 does not end on cylinder boundary. dd if=/dev/hda count=195 | od -c -A d | sed 30q returned root@Storage:/# dd if=/dev/hda count=195 | od -c -A d | sed 30q 195+0 records in 195+0 records out 99840 bytes (100 kB) copied, 7.52696 s, 13.3 kB/s 0000000 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 * 0000448 \0 \0 203 \0 \0 \0 ? \0 \0 \0 q 220 223 . \0 \0 0000464 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 * 0000496 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 U 252 0000512 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 * 0097792 001 \0 001 \0 \0 \0 202 322 303 257 y 031 J 373 \r 326 0097808 N 002 377 236 254 030 035 B 265 025 \0 \0 346 025 \0 \0 0097824 001 ` 241 \r \0 \0 \0 020 \0 \0 320 017 0 \0 001 \0 0097840 0 224 034 022 1 224 034 022 2 224 034 022 3 224 034 022 0097856 4 224 034 022 5 224 034 022 6 224 034 022 7 224 034 022 0097872 8 224 034 022 : 224 034 022 ; 224 034 022 < 224 034 022 0097888 = 224 034 022 > 224 034 022 ? 224 034 022 @ 224 034 022 0097904 A 224 034 022 B 224 034 022 C 224 034 022 D 224 034 022 0097920 E 224 034 022 F 224 034 022 G 224 034 022 H 224 034 022 0097936 I 224 034 022 J 224 034 022 K 224 034 022 L 224 034 022 0097952 M 224 034 022 N 224 034 022 O 224 034 022 P 224 034 022 0097968 Q 224 034 022 R 224 034 022 S 224 034 022 T 224 034 022 0097984 U 224 034 022 V 224 034 022 W 224 034 022 X 224 034 022 0098000 Y 224 034 022 Z 224 034 022 [ 224 034 022 \ 224 034 022 0098016 ] 224 034 022 ^ 224 034 022 _ 224 034 022 ` 224 034 022 0098032 a 224 034 022 b 224 034 022 c 224 034 022 d 224 034 022 0098048 e 224 034 022 f 224 034 022 g 224 034 022 h 224 034 022 0098064 i 224 034 022 j 224 034 022 k 224 034 022 l 224 034 022 0098080 m 224 034 022 n 224 034 022 o 224 034 022 p 224 034 022 0098096 q 224 034 022 r 224 034 022 s 224 034 022 t 224 034 022 0098112 u 224 034 022 v 224 034 022 w 224 034 022 x 224 034 022 0098128 y 224 034 022 z 224 034 022 { 224 034 022 | 224 034 022
Joe L. Posted December 23, 2011 Posted December 23, 2011 I don't recognize that as a reiser file system, at least not at that offset. It looks like there is a sequentlal listing of the ASCII character set there. Furthermore, it says the partition starts on sector 63, which is what I would expect to see for a disk that has been in your server for a while. (since prior the ability to specify a format of What else can you tell us about the history of the disk /dev/hda Is it the correct disk as shown on your devices page?
dark_avenger Posted December 23, 2011 Author Posted December 23, 2011 The drive is a 400GB SATA which is hanging off an IDE->SATA converter (ran out of ports) and has been in there since I first built the server over a year ago. The drive I added yesterday is connected to the other port (hdb) The new drive is picked up as I can see it with the pre clear script. I'll shut it down and double check the cables again see if that makes a difference. EDIT: Shutdown and checked the cable, is still showing unformatted and returns the same info as above.
elkay14 Posted December 24, 2011 Posted December 24, 2011 What about plugging the drive into a native SATA port to check it? We need to eliminate the converter as a possible failure.
dark_avenger Posted December 24, 2011 Author Posted December 24, 2011 Unfortunately I don't have any spare ports in the machine. The new drive is 35% through a preclear on the same converter though (1.5TB Seagate) When it finishes preclear I'll disconnect it and see if maybe the converter is having trouble with two drives ?
dark_avenger Posted December 24, 2011 Author Posted December 24, 2011 You can't swap with another drive? I suppose I could, just means the arrary won't work yeah?
Joe L. Posted December 24, 2011 Posted December 24, 2011 Thanks for the quick response. I ran both the commands as posted. fdisk -lu /dev/hda returned 1 heads, 63 sectors/track, 12403536 cylinders, total 781422768 sectors Units = sectors of 1 * 512 = 512 bytes Disk identifier: 0x00000000 Device Boot Start End Blocks Id System /dev/hda1 63 781422767 390711352+ 83 Linux Partition 1 does not end on cylinder boundary. dd if=/dev/hda count=195 | od -c -A d | sed 30q returned root@Storage:/# dd if=/dev/hda count=195 | od -c -A d | sed 30q 195+0 records in 195+0 records out 99840 bytes (100 kB) copied, 7.52696 s, 13.3 kB/s 0000000 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 * 0000448 \0 \0 203 \0 \0 \0 ? \0 \0 \0 q 220 223 . \0 \0 0000464 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 * 0000496 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 U 252 0000512 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 \0 * 0097792 001 \0 001 \0 \0 \0 202 322 303 257 y 031 J 373 \r 326 0097808 N 002 377 236 254 030 035 B 265 025 \0 \0 346 025 \0 \0 0097824 001 ` 241 \r \0 \0 \0 020 \0 \0 320 017 0 \0 001 \0 0097840 0 224 034 022 1 224 034 022 2 224 034 022 3 224 034 022 0097856 4 224 034 022 5 224 034 022 6 224 034 022 7 224 034 022 0097872 8 224 034 022 : 224 034 022 ; 224 034 022 < 224 034 022 0097888 = 224 034 022 > 224 034 022 ? 224 034 022 @ 224 034 022 0097904 A 224 034 022 B 224 034 022 C 224 034 022 D 224 034 022 0097920 E 224 034 022 F 224 034 022 G 224 034 022 H 224 034 022 0097936 I 224 034 022 J 224 034 022 K 224 034 022 L 224 034 022 0097952 M 224 034 022 N 224 034 022 O 224 034 022 P 224 034 022 0097968 Q 224 034 022 R 224 034 022 S 224 034 022 T 224 034 022 0097984 U 224 034 022 V 224 034 022 W 224 034 022 X 224 034 022 0098000 Y 224 034 022 Z 224 034 022 [ 224 034 022 \ 224 034 022 0098016 ] 224 034 022 ^ 224 034 022 _ 224 034 022 ` 224 034 022 0098032 a 224 034 022 b 224 034 022 c 224 034 022 d 224 034 022 0098048 e 224 034 022 f 224 034 022 g 224 034 022 h 224 034 022 0098064 i 224 034 022 j 224 034 022 k 224 034 022 l 224 034 022 0098080 m 224 034 022 n 224 034 022 o 224 034 022 p 224 034 022 0098096 q 224 034 022 r 224 034 022 s 224 034 022 t 224 034 022 0098112 u 224 034 022 v 224 034 022 w 224 034 022 x 224 034 022 0098128 y 224 034 022 z 224 034 022 { 224 034 022 | 224 034 022 Your output quit before sector 64 because of the data there. (we asked it to only print the first 30 lines) I've no idea what that might be, but it could be part of the housekeeping area of the reiserfs file-system. Let's try one more command so we can at least get past that initial data to where the file-system might be if you had it starting on sector 64. Please type: dd if=/dev/hda count=197 | od -c -A d | sed -n '/R...e...I...s...E...r...2...F...s/p' Its output should look a lot like this: 197+0 records in 197+0 records out 100864 bytes (101 kB) copied, 0.00277979 s, 36.3 MB/s 0097840 314 003 002 \0 R e I s E r 2 F s \0 \0 \0 If the address = 0097840 the file system starts on sector 63. If the address = 0098352 the file system starts on sector 64. Joe L.
dark_avenger Posted December 24, 2011 Author Posted December 24, 2011 That command returns root@Storage:~# dd if=/dev/hda count=197 | od -c -A d | sed -n '/R...e...I...s...E...r...2...F...s/p' 197+0 records in 197+0 records out 100864 bytes (101 kB) copied, 7.62787 s, 13.2 kB/s
dark_avenger Posted December 24, 2011 Author Posted December 24, 2011 If I disconnect the 400gb drive and assign the new 1.5tb drive to the slot will it rebuild onto the new drive?
Joe L. Posted December 25, 2011 Posted December 25, 2011 If I disconnect the 400gb drive and assign the new 1.5tb drive to the slot will it rebuild onto the new drive? It will, but it will rebuild exactly what you have now. (and I'm not seeing a reiser file-system on that disk)
dark_avenger Posted December 26, 2011 Author Posted December 26, 2011 If I disconnect the 400gb drive and assign the new 1.5tb drive to the slot will it rebuild onto the new drive? It will, but it will rebuild exactly what you have now. (and I'm not seeing a reiser file-system on that disk) Suggestions? When I get a chance i'll shut it down and remove the new one see if that changes anything if not I'll try plugging it into a different port.
dark_avenger Posted January 3, 2012 Author Posted January 3, 2012 Drive still shows unformatted without the new drive. Is there anyway for me to recover the drive? Ran reiserfsck --check on the drive it returned: root@Storage:/# reiserfsck --check /dev/hda reiserfsck 3.6.21 (2009 www.namesys.com) ************************************************************* ** 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 read-only check consistency of the filesystem on /dev/hda Will put log info to 'stdout' Do you want to run this program?[N/Yes] (note need to type Yes if you do):Yes reiserfs_open: the reiserfs superblock cannot be found on /dev/hda. Failed to open the filesystem. If the partition table has not been changed, and the partition is valid and it really contains a reiserfs partition, then the superblock is corrupted and you need to run this utility with --rebuild-sb. root@Storage:/# Is it likely that the file system is gone or more likely to be a hardware issue? I would profer not to loose what is on the drive but have 80% of it backed up.
Joe L. Posted January 3, 2012 Posted January 3, 2012 Drive still shows unformatted without the new drive. Is there anyway for me to recover the drive? Ran reiserfsck --check on the drive it returned: root@Storage:/# reiserfsck --check /dev/hda reiserfsck 3.6.21 (2009 www.namesys.com) ************************************************************* ** 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 read-only check consistency of the filesystem on /dev/hda Will put log info to 'stdout' Do you want to run this program?[N/Yes] (note need to type Yes if you do):Yes reiserfs_open: the reiserfs superblock cannot be found on /dev/hda. Failed to open the filesystem. If the partition table has not been changed, and the partition is valid and it really contains a reiserfs partition, then the superblock is corrupted and you need to run this utility with --rebuild-sb. root@Storage:/# Is it likely that the file system is gone or more likely to be a hardware issue? I would profer not to loose what is on the drive but have 80% of it backed up. When will everyone learn. The file system is on the FIRST partition, not the base drive. To check it, you must use the name of the first partition, in this case /dev/hda1 If you give it the wrong name, it will not find the superblock, since it is not there. Joe L.
Joe L. Posted January 3, 2012 Posted January 3, 2012 earlier in this thread you proved the superblock is not found where expected (the reiserfsck command also looks for the same string to know it sees a superblock..) Somehow, that area of your disk has unexpected data. We have no way to know how it got there. If you know with reasonable certainty that the disk was partitioned to start on sector 63, (you installed it on an older version of unRAID, prior to 4k alignment being possible) Then, you might be able to rebuild your superblock and get back your data. It will only have a chance of working if you re-create it on the FIRST partition, not the base drive. I would expect the --check on /dev/hda1 to not find a superblock. If you re-create it on /dev/hda1 it might get you back your data. You might need to also use the -S option to scan the entire partition for your files. Anything you do to the physical drive (/dev/hda1) will invalidate parity. If at all possible, first attempt to perform the repair on the /dev/mdX device. at least it will keep the parity in sync) Assuming this is disk3, that device would be /dev/md3 Joe L.
dark_avenger Posted January 3, 2012 Author Posted January 3, 2012 My bad, have run it to check /dev/hda1 and it returns the same error message. It was created at sector 63 as I have no 4K drives and has been left on the default MBR:"unaligned" If I run the rebuild-sb command is it going to attempt to rebuild the file system or just create a huge amount of files in lost+found? reiserfsck --rebuild-sb /dev/md3 ?
dark_avenger Posted January 10, 2012 Author Posted January 10, 2012 Ran reiserfsck --rebuild-sb and reiserfsck --rebuild-tree -S File system is back with a hand full of files in lost+found. Not sure what cause the issue but very glad for it to be all back....
Recommended Posts
Archived
This topic is now archived and is closed to further replies.