November 5, 20169 yr Hi everyone I have a problem with my Unraid server. It appeared after upgrading to 6.2.3 but it certainly could be unrelated. So, these are the symptoms/timeline: 1. 2 days ago I show a redball on disk6 of my 10 disk + parity array. After checking it (there was a write error) and checking the drive, I found it OK and rebuilt it. Everything went fine and all files were accessible. 2. I was planning for a parity check but hadn't started it yet, when I saw these messages: Nov 4 12:02:13 towerP shfs/user: err: shfs_readdir: fstatat: Dirk.Gentlys.Holistic.Detective.Agency.S01E02.720p.HDTV.x264-KILLERS.mp4 (2) No such file or directory Nov 4 12:02:13 towerP shfs/user: err: shfs_readdir: readdir_r: /mnt/disk6/-TV/-- NOT SEEN (2) No such file or directory Nov 4 12:02:39 towerP shfs/user: err: shfs_readdir: fstatat: Dirk.Gentlys.Holistic.Detective.Agency.S01E02.720p.HDTV.x264-KILLERS.mp4 (2) No such file or directory Nov 4 12:02:39 towerP shfs/user: err: shfs_readdir: readdir_r: /mnt/disk6/-TV/-- NOT SEEN (2) No such file or directory Nov 5 03:00:01 towerP shfs/user: err: shfs_readdir: fstatat: Dirk.Gentlys.Holistic.Detective.Agency.S01E02.720p.HDTV.x264-KILLERS.mp4 (2) No such file or directory Nov 5 03:00:01 towerP shfs/user: err: shfs_readdir: readdir_r: /mnt/disk6/-TV/-- NOT SEEN (2) No such file or directory Nov 5 10:30:02 towerP kernel: md: sync done. time=100688sec Nov 5 10:30:02 towerP kernel: md: recovery thread: completion status: 0 Nov 5 10:35:33 towerP kernel: REISERFS warning: reiserfs-5089 is_internal: free space seems wrong: level=3, nr_items=159, free_space=376 rdkey Nov 5 10:35:33 towerP kernel: REISERFS error (device md6): vs-5150 search_by_key: invalid format found in block 973198707. Fsck? Nov 5 10:35:33 towerP kernel: REISERFS (device md6): Remounting filesystem read-only Nov 5 10:35:33 towerP kernel: REISERFS error (device md6): vs-13070 reiserfs_read_locked_inode: i/o failure occurred trying to find stat data of [4379 26687 0x0 SD] Nov 5 10:35:33 towerP kernel: REISERFS warning: reiserfs-5089 is_internal: free space seems wrong: level=3, nr_items=159, free_space=376 rdkey Nov 5 10:35:33 towerP kernel: REISERFS error (device md6): vs-5150 search_by_key: invalid format found in block 973198707. Fsck? Nov 5 10:35:33 towerP kernel: REISERFS error (device md6): vs-13070 reiserfs_read_locked_inode: i/o failure occurred trying to find stat data of [4379 26640 0x0 SD] Then, this message keeps repeating: Nov 5 13:59:55 towerP kernel: REISERFS error (device md6): vs-5150 search_by_key: invalid format found in block 973198707. Fsck? Nov 5 13:59:55 towerP kernel: REISERFS error (device md6): vs-13070 reiserfs_read_locked_inode: i/o failure occurred trying to find stat data of [4379 26663 0x0 SD] 3. I could access the array for a while (even files on disk 6) but now it is not accessible through the web interface but is accessible by command line. UPDATE: accessible through the web interface again 4. There is a process "reiserfsck" running so I dont want to reboot. Any input as to how i should proceed? (syslog - pruned because of size - attached) syslog.bak.zip
November 6, 20169 yr You need to use Check Disk File systems to check and fix Disk 6. When you say reiserfsck is running, that's not something that runs unless the user runs it. Do you recall how you started it? You can check top or htop at the command line, to see if it's actually using CPU. Or check a ps report in the diagnostics or from the Tools menu. Once you know it isn't running, you can restart the system and then check Disk 6. By the way, we always prefer the diagnostics (Tools -> Diagnostics) over the syslog. It includes the syslog, system and config files, and SMART reports, zipped. The syslog is truncated if large, since we only need the first megabyte. For more help, see Need help? Read me first!
November 6, 20169 yr Author You need to use Check Disk File systems to check and fix Disk 6. Thanks, doing that now, keeping my fingers crossed. There were errors according to the test, so i am running using the --rebuild-tree option When you say reiserfsck is running, that's not something that runs unless the user runs it. Do you recall how you started it? You can check top or htop at the command line, to see if it's actually using CPU. Or check a ps report in the diagnostics or from the Tools menu. Once you know it isn't running, you can restart the system and then check Disk 6. I did not manually start it. But, come to think of it, I suspect that it was the result of using unmenu->Disk Management->Filesystem check By the way, we always prefer the diagnostics (Tools -> Diagnostics) over the syslog. It includes the syslog, system and config files, and SMART reports, zipped. The syslog is truncated if large, since we only need the first megabyte. For more help, see Need help? Read me first! I know, unfortunately, at that moment, unraid web interface was not working, so i got the syslog using the command line
November 6, 20169 yr Community Expert By the way, we always prefer the diagnostics (Tools -> Diagnostics) over the syslog. It includes the syslog, system and config files, and SMART reports, zipped. The syslog is truncated if large, since we only need the first megabyte. For more help, see Need help? Read me first! I know, unfortunately, at that moment, unraid web interface was not working, so i got the syslog using the command line Read the link. It tells you how to get the diagnostics from the command line.
November 7, 20169 yr Author By the way, we always prefer the diagnostics (Tools -> Diagnostics) over the syslog. It includes the syslog, system and config files, and SMART reports, zipped. The syslog is truncated if large, since we only need the first megabyte. For more help, see Need help? Read me first! I know, unfortunately, at that moment, unraid web interface was not working, so i got the syslog using the command line Read the link. It tells you how to get the diagnostics from the command line. Next time (hopefully not soon) I will know . btw, after fixing everything and rechecking the drive (reiserfs check shows no problems anymore) i got the attached SMART report which seems OK. But clicking on the disk from main I get this (screencap attached): 5 Reallocated sectors count 187 Reported uncorrectable errors 188 Command time-out 197 Current pending sector count 198 Uncorrectable sector count It is difficult to understand which one is correct, when I have to decide whether I should RMA my WD red 6TB. why this difference? WDC_WD60EFRX-68MYMN1_WD-WX31D743YHKP-20161107-1419.txt
November 7, 20169 yr Community Expert 5 Reallocated sectors count 187 Reported uncorrectable errors 188 Command time-out 197 Current pending sector count 198 Uncorrectable sector count Those are not a problem, they are the attributes unRAID is monitoring and will warn if they change.
November 7, 20169 yr Author 5 Reallocated sectors count 187 Reported uncorrectable errors 188 Command time-out 197 Current pending sector count 198 Uncorrectable sector count Those are not a problem, they are the attributes unRAID is monitoring and will warn if they change. Thanks for the info, good to know
November 7, 20169 yr I personally think 199 (the CRC count) should be added to that list, just type it into the Custom box. I think ALL users should be monitoring it. I've requested it in the past, but it hasn't made it high enough on the To Do list.
November 7, 20169 yr Community Expert I personally think 199 (the CRC count) should be added to that list, just type it into the Custom box. I think ALL users should be monitoring it. I've requested it in the past, but it hasn't made it high enough on the To Do list. + 1 I manually added it in all my servers.
November 11, 20169 yr Author I personally think 199 (the CRC count) should be added to that list, just type it into the Custom box. I think ALL users should be monitoring it. I've requested it in the past, but it hasn't made it high enough on the To Do list. Good to know, will do
Archived
This topic is now archived and is closed to further replies.