August 16, 201114 yr I have a feeling if I don't start the raid, with all the problems that may create, I will not be able to properly do the smartcl check or the reaschk. I keep getting device not found, no matter which I use now. They all report good on boot, for smartcl purposes, however Without the array started, there will not be /dev/mdX devices and you will get the "not found" messages. You can run reiserfsck --check /dev/sdX1 commands on each of the data drives. (note the "1" at the end of the name. The file system is on the first partition. Normally this first partition is connected internally/logically to the /dev/mdX partition. The "--check" option will tell you if the file system needs repair. It will not invalidate parity. Do not run reiserfsck on the parity drive. It has no file system. Running any of the repair options in reiserfsck (--fix-fixable, --rebuild-tree, --rebuild-sb, etc ) on the /dev/sdX1 device will invalidate parity. That will then require a parity calc be re-run to correct parity once all the data disks are back to having good file systems. If a check option indicates repair is needed, post the output of the --check output before doing anything. In particular, rebuilding the superblock needs the prompts to be answered correctly for unRAID, and the defaults are not correct. (you can cause damage and make subsequent repair harder) If you have a current syslog, zip it up and attach it to your next post. Joe L.
August 16, 201114 yr I have a feeling if I don't start the raid, with all the problems that may create, I will not be able to properly do the smartcl check or the reaschk. I keep getting device not found, no matter which I use now. They all report good on boot, for smartcl purposes, however Without the array started, there will not be /dev/mdX devices and you will get the "not found" messages. You can run reiserfsck --check /dev/sdX1 commands on each of the data drives. (note the "1" at the end of the name. The file system is on the first partition. Normally this first partition is connected internally/logically to the /dev/mdX partition. The "--check" option will tell you if the file system needs repair. It will not invalidate parity. Do not run reiserfsck on the parity drive. It has no file system. Running any of the repair options in reiserfsck (--fix-fixable, --rebuild-tree, --rebuild-sb, etc ) on the /dev/sdX1 device will invalidate parity. That will then require a parity calc be re-run to correct parity once all the data disks are back to having good file systems. If a check option indicates repair is needed, post the output of the --check output before doing anything. In particular, rebuilding the superblock needs the prompts to be answered correctly for unRAID, and the defaults are not correct. (you can cause damage and make subsequent repair harder) If you have a current syslog, zip it up and attach it to your next post. Joe L. You will probably find some file systems that are not corrupt, since you just "formatted" them. You'll then need to force reiserfsck to scan through the apparently empty file-system to look for files that used to exist. To do that, assuming the file-system is sane, you'll need to run on each of the data drives: reiserfsck --rebuild-tree -S /dev/sdX1 The above command will take many hours for any given drive. The session must stay connected. It is recommended you do it from the system console. Your accidentally deleted files will be in a lost+found folder created on the drive by the reiserfsck command.
August 16, 201114 yr Author How do I make sure I am not running the reischk on the parity drive? My first drive in the unit is the parity drive, when I run the check, that's not disk1? or is it? Roger
August 16, 201114 yr Author So do I need to start the raid in the GUI menu in order to start reiserfck? when I try it now, I typed exactly what you posted, reiserfck --check /dev/sdX1, I get -bash: resiserfck: command not found
August 16, 201114 yr Author Sorry, forgot to run the samba start first, but; after I did and the reiserfsck --check /dev/sdX1 loaded, and I typed the "Yes", I got Failed to open the device 'dev/sdX1': No such file or directory. Is that again, because I need to start the raid? I remember you telling me not do do anything that causes anything to be written, and I know if I start the raid it will try and write the new Disk 8 it found (which of course is the same Disk 8 that was always there, and now shows blue).
August 16, 201114 yr Look at the devices page. The port identifier is listed as part of each device. It will appear as sda, sdb, sdc etc or possibly hda, hdb etc for some drives. Use that indentifier in place of "sdX" and MAKE SURE YOU ADD THE 1 on the end. You posted a new disk12 which makes it sound like you had 11 good data disks with data before you hit the format. So, you have to run the reiserfsck command on each disk or 11 different times. Peter
August 16, 201114 yr Let's try it again, here is the syslog Looking in your syslog is the device inventory as known by unRAID. The three letter device name in the parens is the device name, ie. the parity disk is /dev/sda The inventory also gives the model and serial number for each disk. Since the three letter device names can change from one boot to another, use the pci-device names (4.X series uses this) or model/serial numbers (that is what current beta unRAID does) Aug 16 14:30:05 Grad_Tower emhttp: Device inventory: Aug 16 14:30:05 Grad_Tower emhttp: pci-0000:00:1f.2-scsi-0:0:0:0 host1 (sda) ST31500341AS_9VS1RT8W Aug 16 14:30:05 Grad_Tower emhttp: pci-0000:00:1f.2-scsi-1:0:0:0 host2 (sdb) WDC_WD10EACS-00ZJB0_WD-WCASJ0353159 Aug 16 14:30:05 Grad_Tower emhttp: pci-0000:00:1f.2-scsi-2:0:0:0 host3 (sdc) WDC_WD10EACS-00ZJB0_WD-WCASJ0235583 Aug 16 14:30:05 Grad_Tower emhttp: pci-0000:00:1f.2-scsi-3:0:0:0 host4 (sdd) WDC_WD10EACS-00ZJB0_WD-WCASJ0418884 Aug 16 14:30:05 Grad_Tower emhttp: pci-0000:00:1f.2-scsi-4:0:0:0 host5 (sde) WDC_WD10EACS-00ZJB0_WD-WCASJ0418898 Aug 16 14:30:05 Grad_Tower emhttp: pci-0000:00:1f.2-scsi-5:0:0:0 host6 (sdg) WDC_WD10EACS-00ZJB0_WD-WCASJ0235537 Aug 16 14:30:05 Grad_Tower emhttp: pci-0000:02:00.0-scsi-1:0:0:0 host8 (sdh) ST31000333AS_9TE08HKZ Aug 16 14:30:05 Grad_Tower emhttp: pci-0000:04:03.0-scsi-0:0:0:0 host9 (sdi) WDC_WD10EACS-00ZJB0_WD-WCASJ0418746 Aug 16 14:30:05 Grad_Tower emhttp: pci-0000:04:03.0-scsi-1:0:0:0 host10 (sdj) ST31500341AS_9VS078XL Aug 16 14:30:05 Grad_Tower emhttp: pci-0000:04:03.0-scsi-2:0:0:0 host11 (sdk) ST31500341AS_9VS02711 Aug 16 14:30:05 Grad_Tower emhttp: pci-0000:04:03.0-scsi-3:0:0:0 host12 (sdl) ST31500541AS_5XW00W1M Aug 16 14:30:05 Grad_Tower emhttp: pci-0000:04:04.0-scsi-0:0:0:0 host13 (sdm) WDC_WD10EACS-00ZJB0_WD-WCASJ0343021 Aug 16 14:30:05 Grad_Tower emhttp: pci-0000:04:04.0-scsi-1:0:0:0 host14 (sdn) WDC_WD10EACS-00ZJB0_WD-WCASJ0418602 Aug 16 14:30:05 Grad_Tower emhttp: pci-0000:04:04.0-scsi-2:0:0:0 host15 (sdo) WDC_WD10EACS-00ZJB0_WD-WCASJ0957015 Aug 16 14:30:05 Grad_Tower emhttp: pci-0000:04:04.0-scsi-3:0:0:0 host16 (sdp) WDC_WD10EACS-00ZJB0_WD-WCASJ1437511 When imported, the parity disk is shown as disk0. Your parity disk is /dev/sda Aug 16 14:30:05 Grad_Tower kernel: md: import disk0: [8,0] (sda) ST31500341AS 9VS1RT8W size: 1465138552 Aug 16 14:30:05 Grad_Tower kernel: md: import disk1: [8,16] (sdb) WDC WD10EACS-00Z WD-WCASJ0353159 size: 976762552 Aug 16 14:30:05 Grad_Tower kernel: md: import disk2: [8,64] (sde) WDC WD10EACS-00Z WD-WCASJ0418898 size: 976762552 Aug 16 14:30:05 Grad_Tower kernel: md: import disk3: [8,96] (sdg) WDC WD10EACS-00Z WD-WCASJ0235537 size: 976762552 Aug 16 14:30:05 Grad_Tower kernel: md: import disk4: [8,32] (sdc) WDC WD10EACS-00Z WD-WCASJ0235583 size: 976762552 Aug 16 14:30:05 Grad_Tower kernel: md: import disk5: [8,48] (sdd) WDC WD10EACS-00Z WD-WCASJ0418884 size: 976762552 Aug 16 14:30:05 Grad_Tower kernel: md: import disk6: [8,192] (sdm) WDC WD10EACS-00Z WD-WCASJ0343021 size: 976762552 Aug 16 14:30:05 Grad_Tower kernel: md: import disk7: [8,208] (sdn) WDC WD10EACS-00Z WD-WCASJ0418602 size: 976762552 Aug 16 14:30:05 Grad_Tower kernel: md: import disk8: [8,224] (sdo) WDC WD10EACS-00Z WD-WCASJ0957015 size: 976762552 Aug 16 14:30:05 Grad_Tower kernel: md: disk8 replaced Aug 16 14:30:05 Grad_Tower kernel: md: import disk9: [8,240] (sdp) WDC WD10EACS-00Z WD-WCASJ1437511 size: 976762552 Aug 16 14:30:05 Grad_Tower kernel: md: import disk10: [8,128] (sdi) WDC WD10EACS-00Z WD-WCASJ0418746 size: 976762552 Aug 16 14:30:05 Grad_Tower kernel: md: import disk11: [8,144] (sdj) ST31500341AS 9VS078XL size: 1465138552 Aug 16 14:30:05 Grad_Tower kernel: md: import disk12: [8,160] (sdk) ST31500341AS 9VS02711 size: 1465138552 Aug 16 14:30:05 Grad_Tower kernel: md: import disk13: [8,176] (sdl) ST31500541AS 5XW00W1M size: 1465138552 Aug 16 14:30:05 Grad_Tower kernel: md: import disk14: [8,112] (sdh) ST31000333AS 9TE08HKZ size: 976762552 unRAID thinks Disk8 has been changed. Has it?
August 16, 201114 yr Look at the devices page. The port identifier is listed as part of each device. It will appear as sda, sdb, sdc etc or possibly hda, hdb etc for some drives. Use that indentifier in place of "sdX" and MAKE SURE YOU ADD THE 1 on the end. You posted a new disk12 which makes it sound like you had 11 good data disks with data before you hit the format. So, you have to run the reiserfsck command on each disk or 11 different times. Peter correct... but there are apparently 14 data disks. reiserfsck --check /dev/sdb1 reiserfsck --check /dev/sdc1 reiserfsck --check /dev/sdd1 reiserfsck --check /dev/sde1 reiserfsck --check /dev/sdg1 reiserfsck --check /dev/sdh1 reiserfsck --check /dev/sdi1 reiserfsck --check /dev/sdj1 reiserfsck --check /dev/sdk1 reiserfsck --check /dev/sdl1 reiserfsck --check /dev/sdm1 reiserfsck --check /dev/sdn1 reiserfsck --check /dev/sdo1 reiserfsck --check /dev/sdp1
August 16, 201114 yr Author In answer to Joe L, and thank you so much for your time and help. Trying to be real careful here. Disk 8, it's the same as it's always been, haven't replaced it. Just when I rebooted it, now comes up as new. I did earlier make the mistake of hitting the start and checking the box and it tried to rebuild and failed, but the rest of the array seemed to be online and working, except as I noted I couldn't see the files on Disk 3, 5, and 8, all the rest appeared there, even ran a movie just to see. So, now based on your log and my syslog. After I log in, and type samba stop. I type the following: reiserfsck --check /dev/sdb1 then reiserfsck --check /dev/sdc1 and so on. Yes, there are in fact 14 disks, plus one parity disk, I think I should have 14 good disks with data. But I get that the format has likely messed with it all. Hoping it's stil there
August 16, 201114 yr Author I was thinking about leaving work and going home to try to work on this. If I can get confirmation of the next step or someone willing to actually talk me through I will do that. Is the next step, after I log into the server, cd root, type samba stop, then: reiserfsck --check /dev/sdb1 reiserfsck --check /dev/sdc1 and down until I've done 14 drives?
August 16, 201114 yr I was thinking about leaving work and going home to try to work on this. If I can get confirmation of the next step or someone willing to actually talk me through I will do that. Is the next step, after I log into the server, cd root, type samba stop, then: reiserfsck --check /dev/sdb1 reiserfsck --check /dev/sdc1 and down until I've done 14 drives? exactly. If you do it in a telnet session you should be able to capture the output for pasting. If not, you can modify the command slightly by adding a "tee" command at the end. reiserfsck --check /dev/sdb1 | tee >/boot/sdb_check.txt Since your array is not started, no need to stop samba.
August 16, 201114 yr Author For everyones information who is trying to help (any many thanks) Started reiserfsck --check /dev/sdb1, Got the following so far. Replaying journal: done. Reiserfs journal /dev/sdb1 in blocks [18..8211]: 0 transactions replayed. Checking internal tree .. \/ 2 (of 12// 120 ... And it's jsut checking away. Am I doing this correct?
August 16, 201114 yr Author Joe L, started this before getting your post, now checking tree 8, should I re check with tee command sdb1, or if nothing comes up, just move on to the next device?
August 16, 201114 yr Joe L, started this before getting your post, now checking tree 8, should I re check with tee command sdb1, or if nothing comes up, just move on to the next device? just let it run. All you really care about is if it tells you you need to run fix-fixable, or rebuild-tree, or of all is OK.
August 16, 201114 yr Author Thanks Joe, Right now it's nome Checking Semantic tree: Looks like it's looking at Mp3's I have right now, which makes sense, I have a lot of my music on disk 1. When I get all done, I would expect since now files are appearing on Disk 3, 5, and 8, that I will need to run the fix command, before doing anything else. I expect this is obvious, but; before I do that I should load up the check.txt files, which I will get on the other 13 disks, and load up only if there is an issue, and than wait for you before entering the next step? Does that make sense? Roger
August 16, 201114 yr Thanks Joe, Right now it's nome Checking Semantic tree: Looks like it's looking at Mp3's I have right now, which makes sense, I have a lot of my music on disk 1. When I get all done, I would expect since now files are appearing on Disk 3, 5, and 8, that I will need to run the fix command, before doing anything else. I expect this is obvious, but; before I do that I should load up the check.txt files, which I will get on the other 13 disks, and load up only if there is an issue, and than wait for you before entering the next step? Does that make sense? Roger run the --check on all of the disks before you proceed. Then you can evaluate what you need to do next. If any say to run fix-fixable, that would be next, one at a time, each in turn.
August 16, 201114 yr Author Finished the first check on the sdb1, report, says No corruptions found. There are on the filesystem, Leaves 256453, Internal nodes 1741, Directories 14937, Other files 211062, Data block pointers 228142194 (0 of them are zero), Safe links 0, ON to the next disk. Roger
August 16, 201114 yr Author Through disk 2, no errors so far, I expect I should see them on Disk 3, Disk 5, and of course Disk 8. Stay tuned. Does this make sense so far? Disk 8 is the one that got disabled after it said it was trying to format. Disk 3, and Disk 5 showed up good in the array and in the gui, but; when I tried to access files, no files showed. Roger
August 16, 201114 yr Author O.K., as susptected, first problem found. ON Disk 3, Says journal parameters from the superblock does not match to the journal headers ones. It looks like that you created your fs with old reiserfsprogs. Journal header is fixed. Replaying journal: Done Reiserfs journal /dev/sdd1 in blocks [18..8211]: 0 transactions replayed, Checking internal tree. Finished Comparing bitmaps..Fatal corruptions were found, Semantic pass skipped, 1 found corruptions can be fixed only when running with -- rebuild-tree. I did run with the tee command to build the txt file, but; can't get access to it in the gui, Should I get to the end and reboot to get to it? SHould I pull the USB out and get to it on a regular computer? MOving to disk 4
August 16, 201114 yr Author Disk 4 is O.K., so I think the theory that Disk 3, 5, and 8 have the issue may hold up. No corruptions found on it. Got to the answer pretty quick. Does this mean it's O.K., or could there still be something wrong with the tree?
August 16, 201114 yr Through disk 2, no errors so far, I expect I should see them on Disk 3, Disk 5, and of course Disk 8. Stay tuned. Does this make sense so far? Disk 8 is the one that got disabled after it said it was trying to format. Disk 3, and Disk 5 showed up good in the array and in the gui, but; when I tried to access files, no files showed. Roger Yes, it makes sense so far. As I said, just do the check's now. Then we can evaluate. I have no interest in how many nodes, leaves, etc. Just if the check passes. Don't fix anything just yet. You'll get to that soon enough.
Archived
This topic is now archived and is closed to further replies.