May 7, 201016 yr The UPS didn't kick on for a brownout and so my unRAID box went down... during a preclear of my new 2TB parity drive. Sweet! When it came back up, I suddenly had three causes of panic: 1) Parity errors... 320, but the parity check took less than an hour (possibly a lot less, not sure when power interruption happened) for four 1TB drives (including parity). 2) disk2 is showing as "DISK_DSBL". 3) The new parity drive isn't visible anymore. It's still in the box, but doesn't show under unMENU's array management. ls /dev/sd* shows a /dev/sdf, which smartclt -a /dev/sdf shows about what I'd expect. Zipped syslog is attached. Help will be appreciated. APC UPS will be pitched into pond behind house. EDIT: I tried to see if the movie that was planned for the evening was available on one of the other disks at least and that seemed to result in many hundred lines that resembled the following: May 7 17:01:09 Dingo kernel: REISERFS error (device md3): vs-13070 reiserfs_read_locked_inode: i/o failure occurred trying to find stat data of [244 246 0x0 SD] May 7 17:01:09 Dingo kernel: REISERFS error (device md3): vs-13070 reiserfs_read_locked_inode: i/o failure occurred trying to find stat data of [244 578 0x0 SD] May 7 17:01:09 Dingo kernel: REISERFS error (device md3): vs-13070 reiserfs_read_locked_inode: i/o failure occurred trying to find stat data of [244 568 0x0 SD] syslog-2010-05-07.zip
May 7, 201016 yr APC UPS will be pitched into pond behind house. A battery in a pond? Think of the environment, man! I understand your frustration though, I would be tempted to 'learn' the UPS as well, and learn it good.
May 7, 201016 yr Rather than pitch the UPS into a pond, just adjust the undervoltage/overvoltage trip point settings.
May 7, 201016 yr Author Rather than pitch the UPS into a pond, just adjust the undervoltage/overvoltage trip point settings. That's a reasonable response to the issue, certainly. I'm... outside of reasonable at the moment. Also, I've never done that with a UPS and am busy scrambling to find answers to my other problems... no time for traitorous appliances!
May 8, 201016 yr Author I rebooted and now parity is "DISK_NP_MISSING" and disk3 is "DISK_DISBL_NP". Help, please?
May 11, 201016 yr Author I cannot believe I lost the whole array and there's nothing to be done about it.
May 11, 201016 yr I doubt you lost the whole array. First, now that you rebooted, post a new syslog. It is possible for one disk to fail and make others on the same channel look as if they've failed too. 4 of your disks are "paired" on host controllers. (Host0 and host1) May 7 15:06:03 Dingo emhttp: pci-0000:00:1f.2-scsi-0:0:0:0 host0 (sda) WDC_WD10EVDS-63U8B0_WD-WCAV58033287 May 7 15:06:03 Dingo emhttp: pci-0000:00:1f.2-scsi-0:0:1:0 host0 (sdb) WDC_WD10EADS-00L5B1_WD-WCAU48422471 May 7 15:06:03 Dingo emhttp: pci-0000:00:1f.2-scsi-1:0:0:0 host1 (sdc) ST31000340AS_5QJ11WGN May 7 15:06:03 Dingo emhttp: pci-0000:00:1f.2-scsi-1:0:1:0 host1 (sdd) WDC_WD10EVDS-63U8B0_WD-WCAV58038771 May 7 15:06:03 Dingo emhttp: pci-0000:00:1f.5-scsi-0:0:0:0 host2 (sdf) WDC_WD20EVDS-63T3B0_WD-WCAVY2958523 May 7 15:06:03 Dingo kernel: md: import disk0: [8,16] (sdb) WDC WD10EADS-00L WD-WCAU48422471 offset: 63 size: 976762552 May 7 15:06:03 Dingo kernel: md: import disk1: [8,32] (sdc) ST31000340AS 5QJ11WGN offset: 63 size: 976762552 May 7 15:06:03 Dingo kernel: md: import disk2: [8,48] (sdd) WDC WD10EVDS-63U WD-WCAV58038771 offset: 63 size: 976762552 May 7 15:06:03 Dingo kernel: md: import disk3: [8,0] (sda) WDC WD10EVDS-63U WD-WCAV58033287 offset: 63 size: 976762552 Your parity disk (sdb) is on host0. Disk3 (sda) is also on host0. Either one could affect the other if it has failed. After you post the new syslog, you can go through the process of isolating one drive, then the other. Joe L.
May 25, 201016 yr Author Welp, I'm at a loss... I've done what I know to do to confirm the drives are okay and they appear to be. Is there anything specifically I should be looking at?
May 25, 201016 yr Welp, I'm at a loss... I've done what I know to do to confirm the drives are okay and they appear to be. Is there anything specifically I should be looking at? Perhaps you did not understand what I was referring to when I said to isolate one or the other of the two drives. Power down. Physically disconnect the SATA cable to one of the two drives, either the parity drive ( /dev/sda), or disk3 (/dev/sdb) Power up and see if the array will come on line. If the disconnected drive was causing both on the same channel to fail, it will boot up. If the array still sees two missing drives, Power down. Re-connect the SATA cable to the drive you first disconnected, and disconnect the SATA cable to the other of the two. Power up. With luck, it will be one of the two drives affecting the other. If you are not as lucky, it could still be the disk controller has failed, (or the disk controller card not seated properly in the motherboard), making both disks connected to it look missing. If that is the case, you can, if you have spare ports on the disk controller, use an alternate port. Joe L.
May 28, 201016 yr Author Nope, I understood... unfortunately, nothing brings them up. I tried only having one of them available at a time, switching SATA ports, sacrificing small animals. Parity shows as missing (if cord yanked) or disabled if not and disk2 is always missing, regardless of where it's plugged. In this scenario, what's the next best move? I'd hate to buy a new motherboard and find out that just the drives were bad, but... EDIT: I mixed up disk2 and disk3 in this reply, but have pulled disk3 as well.
Archived
This topic is now archived and is closed to further replies.