August 14, 200916 yr Hey guys, I hate to say it, but I'm not sure what to do. As the title says, I came home yesterday to see that we had a power outage. Checking the unraid box (it was powered on because the box is set to return to the last state of power on power loss) shows that the array is stopped and all disks are blue. The array is "Stopped. Initial configuration". I haven't done enough experimental "crashes" to troubleshoot this myself in a safe manner, so here I am The syslog: http://pastebin.com/m3dbeff9d I'm presented with the standard Start and Restore options. I haven't done anything yet. I'm thinking this might be what I have to do. Thank you!
August 14, 200916 yr I think you might be right. According to your syslog, your superblock has a bad checksum Aug 14 05:25:38 wegel-nas kernel: md: invalid superblock checksum Aug 14 05:25:38 wegel-nas kernel: md: initializing superblock The post you linked to has the procedure... basically, Press "Restore" to force unRAID to create a new superblock file BUT DO NOT START THE ARRAY, not until you after you perform the next step. Then, at the unRAID command line, type cd mdcmd set invalidslot 99 Then press "Start" to start the array. Let the parity check complete, as you have no idea if the power failure caused a few parity errors. By the time they are reported to you, they will be corrected. Last, purchase an inexpensive UPS. It will get you through most power hits. Joe L.
August 15, 200916 yr Author Oh-oh... root@wegel-nas:~# mdcmd set invalidslot 99 ./mdcmd: line 3: /proc/mdcmd: No such file or directory Here's what happened in the syslog after I ran the command: Aug 14 17:48:36 wegel-nas kernel: md: import disk0: [8,16] (sdb) ST31000340AS 5QJ10BN8 offset: 63 size: 976762552 Aug 14 17:48:36 wegel-nas kernel: md: disk0 new disk Aug 14 17:48:36 wegel-nas kernel: md: import disk1: [8,0] (sda) ST31000340AS 9QJ235V2 offset: 63 size: 976762552 Aug 14 17:48:36 wegel-nas kernel: md: disk1 new disk Aug 14 17:48:36 wegel-nas kernel: md: import disk2: [8,32] (sdc) ST31000340AS 9QJ1LASR offset: 63 size: 976762552 Aug 14 17:48:36 wegel-nas kernel: md: disk2 new disk Aug 14 17:48:43 wegel-nas emhttp: shcmd (11): rmmod md-mod >>/var/log/go 2>&1 Aug 14 17:48:43 wegel-nas kernel: md: unRAID driver removed Aug 14 17:48:43 wegel-nas emhttp: shcmd (12): modprobe md-mod super=/boot/config/super.dat slots =8,16,8,0,8,32 >>/var/log/go 2>&1 Aug 14 17:48:43 wegel-nas kernel: md_mod: Unknown parameter `super' Aug 14 17:48:43 wegel-nas emhttp: shcmd: shcmd (12): exit status: 1 Aug 14 17:48:43 wegel-nas emhttp: mdcmd: open: No such file or directory Aug 14 17:49:37 wegel-nas last message repeated 2 times Aug 14 17:52:56 wegel-nas sshd[3363]: Did not receive identification string from 201.240.94.220 What should I do? Update: Ok, I have an a package that installs modules, which includes stuff like ntfs support, usb cdrom, etc. In that package, I have a md-mod.ko in there; I guess it's somewhat broken. But, the "good" (standard) version must be loaded before I load my version of the modules, so that must be why the problem never popped out before. Could someone be kind enough to tar me the content of /lib/modules for an unraid 4.4.2? I could reboot the box and disable my modules stuff, but I don't feel so safe about rebooting. I did hit once the Restore button, btw. Update2: I created a new key with which I could boot another computer; I then proceeded to tar the content of /lib/modules, and load the "fresh" md-mod.ko on the unraid box. The command now worked, but I wasn't able to start the array (the web page said there was no drive in the array). I rebooted the box, crossing my fingers... ran "mdcmd set invalidslot 99" again, started the array and... Lo and behold, it works It's now doing a parity check. Lesson learned 1: buy a UPS, as Joe L. suggested. Lesson learned 2: do NOT include a non-original md-mod.ko in any module package you may have Thanks.
Archived
This topic is now archived and is closed to further replies.