July 8, 201015 yr I replaced my PS and all looked good. I played back a few short snippets from the drive without a problem (i.e., reads). Thinking the bad times were behind me I then tried to copy a file from my C: drive to the same unRAID drive (i.e., writes)...and it failed. So I used explorer to navigate to the drive and managed to delete a few files...although it took much longer than I expected. So I tried once again to copy a file to the drive and it locked up Explorer. So I went to the webUI and asked unRAID to gracefully shut down the drive. No luck...it hung. So I hit my KVM switch and saw a whole raft of messages that looked like this... [<c0/84f38>]reiserfs_settattr +0x44/0x17f [<c013balc(>]__do_fault +0x2ac/0x2el ... ... etc. So I logged onto root and issued a init 0 command and saw the following messages: INIT 0 Switching to run level :0 INIT Sending processes the TERM signal INIT Sending processes the KILL signal Running shutdown script /etc/rc.d/rc.0 Saving system time to the hardware clock nmdb: no processes killed Unmounting remote files system Then nothing... So...in the end I pulled the power cord and rebooted the box. This time I captured the syslog on reboot right away. Here it is. When unRAID came back online (webUI), all looks normal (i.e., all drives online, all with green status dots). What next??? syslog2.txt
July 8, 201015 yr Author Just noticed this morning that unRAID is about 50% through rebuilding parity. Will update the situation when I get back from work.
July 9, 201015 yr Author It completed the parity rebuild okay. So I brought up Explorer to view a few videos without a problem. But when I tried to copy a video file from my C: drive to my NAS drive, I got a loooong pause (1-2 minutes), followed by an error message that the directory was not accessible because the specified network name was no longer available. So I tried to copy another video file from my D: drive to my NAS. Same loooong pause followed by a new error message... "Cannot copy Mythbusters - "Flu Fiction" (Recorded Jun 9, 2010, DSCHD): Configuration information could not be read from the domain controller, either because the machine is unvailable, or access has been denied." Any thoughts???
July 9, 201015 yr It completed the parity rebuild okay. So I brought up Explorer to view a few videos without a problem. But when I tried to copy a video file from my C: drive to my NAS drive, I got a loooong pause (1-2 minutes), followed by an error message that the directory was not accessible because the specified network name was no longer available. So I tried to copy another video file from my D: drive to my NAS. Same loooong pause followed by a new error message... "Cannot copy Mythbusters - "Flu Fiction" (Recorded Jun 9, 2010, DSCHD): Configuration information could not be read from the domain controller, either because the machine is unvailable, or access has been denied." Any thoughts??? One or more of your file-systems probably needs repair. See the instructions in the wiki here on how to check them: http://lime-technology.com/wiki/index.php?title=Check_Disk_Filesystems Joe L.
July 9, 201015 yr Author Thanks Joe. I tried stopping my unRAID NAS, but of course it hung. I'll try to shut it down gracefully from the OS prompt (init 0), but suspect I won't have a lot of luck there either. I'll give the reiserfsck command a crack tomorrow, or at worst on Saturday after I get it back up & running. BillK
July 9, 201015 yr Thanks Joe. I tried stopping my unRAID NAS, but of course it hung. I'll try to shut it down gracefully from the OS prompt (init 0), but suspect I won't have a lot of luck there either. I'll give the reiserfsck command a crack tomorrow, or at worst on Saturday after I get it back up & running. BillK "init 0" will shut it down, but I doubt if it will be graceful. Good luck. Joe L.
July 11, 201015 yr Author Maybe...but when compared to throwing the power switch, I always thought "init 0" was slightly more graceful. Is "md1" at the linux OS prompt the same as "disk1" in the unRAID UI? I ask because the reiserfsck instructions make it very clear not to run the utility on the parity drive. Is that md0? Thanks.
July 11, 201015 yr Maybe...but when compared to throwing the power switch, I always thought "init 0" was slightly more graceful. Compared to pulling the plug, or turning off the switch, it is more graceful. Is "md1" at the linux OS prompt the same as "disk1" in the unRAID UI?Yes. I ask because the reiserfsck instructions make it very clear not to run the utility on the parity drive. Is that md0?That is because there is no file system on the parity drive, just the parity calculations. No, there is no /dev/md0. Joe L.
July 11, 201015 yr Author When the system came back on line, it had to rebuild parity. After it finished, I decided I'd taken enough risks with drive1 and potential file-system problems, so I slowly but surely copied the contents onto disk7 (successfully). I then tried to delete the files on disk1 when the system hung again. This time when it rebooted, disk1 was shown "Unformatted". So...I asked unRAID to format the disk. That was 8 hours ago (1TB drive). It still says "Started, formatting...", but I can't tell whether it's really doing anything or not (i.e., no status information). Is there a way to tell via the command prompt? Any other advice?
July 12, 201015 yr Author Well....after 12 hours, and a spun down drive, I figured it was pretty safe to move on (i.e., it thought it was formatting the drive, but really wasn't). So I hit the KVM switch and ran the reiserfsck commands against all 7 drives. No reported problems found, although it clearly didn't think md1 was a normal disk. I tried "init 0" again, but as usual it hung. The good news is that when I booted up there were no error messages after the login prompt and the parity drive started building. We'll see.
Archived
This topic is now archived and is closed to further replies.