May 21, 201610 yr Hi, I'm really quite concerned tbh! 19th May, I came back to a completely crashed Web GUI... Everything seemed to have gone back to default settings... Turns out the USB had overheated because my Asrock board decided it didn't need fans spinning to cool anything!! I was greeted with /dev/sda undetected and a load of warnings. I rebooted and everything was fine though... So I deleted my post on here I was wrong. Disk 3 had 1.48TB of data on it... From after that reboot it only reported 1.28TB!! I was worried, it almost exclusively contains Plex media files, none of which seemed to be missing however. So I'm super confused with unRAID! Moving onto what's happened today though! (2 diag files attached, one for just now and the other for the 19th) I tried to stop my array and got a crap ton of errors along the bottom of the web GUI Error, couldn't stop array, couldn't unmount disk. That went on and I eventually had to hold the power button down Which I hated! I couldn't get to the web GUI, SSH or even ping it. I used the IPMI and grabbed the diag file from the USB If someone could help me figure out what's going on that would be immensely appreciated I've got nothing out of the ordinary going on AFAIK, 2 dockers, couple of users accounts and a handful of SMB shares - Pretty basic in the grand scheme of things raptor-diagnostics-20160521-1535.zip raptor-diagnostics-20160519-2238.zip
May 21, 201610 yr Nothing obvious in the diagnostics beyond what you have already described. And there is no actual evidence about this: ... Disk 3 had 1.48TB of data on it... From after that reboot it only reported 1.28TB... How do you know that? You might try checking the filesystem on disk3. And what does all this have to do with this beta release?
May 21, 201610 yr Author Nothing obvious in the diagnostics beyond what you have already described. And there is no actual evidence about this: ... Disk 3 had 1.48TB of data on it... From after that reboot it only reported 1.28TB... How do you know that? You might try checking the filesystem on disk3. And what does all this have to do with this beta release? Because... That's what it says in the Main tab. What this has to do with the beta is; That either the beta ate a couple hundred gb somehow Or The web GUI is incorrectly reporting disk usage I did a Parity check last night, I was planning on running the btrfs scrub in a moment
May 21, 201610 yr Author Nothing obvious in the diagnostics beyond what you have already described. And there is no actual evidence about this: ... Disk 3 had 1.48TB of data on it... From after that reboot it only reported 1.28TB... How do you know that? You might try checking the filesystem on disk3. And what does all this have to do with this beta release? I've just taken a quick look at the syslog on the diagnostic for the 19th The lines: May 19 18:07:32 Raptor shfs/user: shfs_rmdir: rmdir: /mnt/cache/Appdata/PlexMediaServer/Library/Application Support/Plex Media Server/Cache/Transcode/Sessions/plex-transcode-s5hjmo20s0j0jcjqtbtg4aemi-25431c8c-7410-49a1-8f8e-cb9eb64a6db7 (39) Directory not empty following this are all about the USB dropping out. That's the point at which the web GUI was white instead of black, had no shares, had no dashboard page and eventually crashed And then the syslog from today: May 21 15:32:45 Raptor kernel: mdcmd (52): spinup 0 May 21 15:32:45 Raptor kernel: mdcmd (53): spinup 1 May 21 15:32:45 Raptor kernel: mdcmd (54): spinup 2 May 21 15:32:45 Raptor kernel: mdcmd (55): spinup 3 May 21 15:32:45 Raptor emhttp: shcmd (817): /usr/sbin/hdparm -S0 /dev/sdg &> /dev/null May 21 15:32:53 Raptor emhttp: Sync filesystems... May 21 15:32:53 Raptor emhttp: shcmd (818): sync May 21 15:32:56 Raptor emhttp: shcmd (819): set -o pipefail ; umount /mnt/user |& logger May 21 15:32:56 Raptor shfs/user: fuse_main exit: 0 May 21 15:32:56 Raptor emhttp: shcmd (820): rmdir /mnt/user |& logger May 21 15:32:56 Raptor emhttp: shcmd (821): set -o pipefail ; umount /mnt/user0 |& logger May 21 15:32:56 Raptor shfs/user0: fuse_main exit: 0 May 21 15:32:56 Raptor emhttp: shcmd (822): rmdir /mnt/user0 |& logger May 21 15:32:56 Raptor emhttp: shcmd (823): rm -f /boot/config/plugins/dynamix/mover.cron May 21 15:32:56 Raptor emhttp: shcmd (824): /usr/local/sbin/update_cron &> /dev/null May 21 15:32:56 Raptor emhttp: Unmounting disks... May 21 15:32:56 Raptor emhttp: shcmd (825): umount /mnt/disk1 |& logger May 21 15:32:57 Raptor emhttp: shcmd (826): rmdir /mnt/disk1 |& logger May 21 15:32:57 Raptor emhttp: shcmd (827): umount /mnt/disk2 |& logger May 21 15:32:58 Raptor emhttp: shcmd (828): rmdir /mnt/disk2 |& logger May 21 15:32:58 Raptor emhttp: shcmd (829): umount /mnt/disk3 |& logger May 21 15:32:58 Raptor root: umount: /mnt/disk3: target is busy May 21 15:32:58 Raptor root: (In some cases useful info about processes that May 21 15:32:58 Raptor root: use the device is found by lsof( or fuser(1).) May 21 15:32:58 Raptor emhttp: shcmd (830): umount /mnt/cache |& logger May 21 15:32:58 Raptor emhttp: shcmd (831): rmdir /mnt/cache |& logger May 21 15:32:58 Raptor emhttp: Retry unmounting disk share(s)... Never seen that on any build of unRAID - I stopped the array and that's what happened over and over until the hard power off. Not sure what caused that (but that's why I uploaded the diag bundle )
May 21, 201610 yr Those lines don't say anything about the flash. The flash is mounted at /boot, not at /mnt, and there is nothing about that or about any sd. The flash is usually sda1. Maybe the flash problem is involved in some way, but unRAID doesn't do much with the flash except unpack the OS into RAM when it boots. After that it is only accessed to read and write GUI settings. And as for whether the beta is involved, you already told us you had a severe hardware failure, though perhaps only temporary, due to the cooling issues you mentioned. You really need to make sure that doesn't happen again. Have you checked the disk for filesystem corruption? I am going to split this into its own topic.
May 21, 201610 yr Author Those lines don't say anything about the flash. The flash is mounted at /boot, not at /mnt, and there is nothing about that or about any sd. The flash is usually sda1. Maybe the flash problem is involved in some way, but unRAID doesn't do much with the flash except unpack the OS into RAM when it boots. After that it is only accessed to read and write GUI settings. And as for whether the beta is involved, you already told us you had a severe hardware failure, though perhaps only temporary, due to the cooling issues you mentioned. You really need to make sure that doesn't happen again. Have you checked the disk for filesystem corruption? I am going to split this into its own topic. I believe that you are confused... Which is partially my fault! Ok so. I'm having 2 separate issues 1. 19th May - My CPU fan decided to spin at its lowest RPM while Plex was transcoding 9 seasons of a TV show This led to the Sandisk USB getting super hot, the Xeon hitting 98 C and everything grinding to almost a complete halt... After shutting down, letting everything cool, changing the fan curves! and booting back up. I noticed that I had 1.28TB instead of the 1.48TB of data stored on Disk3 May 19 18:53:41 Raptor kernel: usb 4-2: USB disconnect, device number 2 May 19 18:53:41 Raptor kernel: FAT-fs (sda1): Directory bread(block 32768) failed May 19 18:53:41 Raptor kernel: FAT-fs (sda1): Directory bread(block 32769) failed May 19 18:53:41 Raptor kernel: FAT-fs (sda1): Directory bread(block 32770) failed May 19 18:53:41 Raptor kernel: FAT-fs (sda1): Directory bread(block 32771) failed May 19 18:53:41 Raptor kernel: FAT-fs (sda1): Directory bread(block 32772) failed May 19 18:53:41 Raptor kernel: FAT-fs (sda1): Directory bread(block 32773) failed May 19 18:53:41 Raptor kernel: FAT-fs (sda1): Directory bread(block 32774) failed May 19 18:53:41 Raptor kernel: FAT-fs (sda1): Directory bread(block 32775) failed May 19 18:53:41 Raptor kernel: FAT-fs (sda1): Directory bread(block 32776) failed May 19 18:53:41 Raptor kernel: FAT-fs (sda1): Directory bread(block 32777) failed That's the boot USB disconnecting itself due to heat 2. 21st May - Tried to stop my array. Ticked the box, hit the stop button and the status / errors at the bottom of the web page filled up with May 21 15:32:45 Raptor emhttp: shcmd (817): /usr/sbin/hdparm -S0 /dev/sdg &> /dev/null May 21 15:32:53 Raptor emhttp: Sync filesystems... May 21 15:32:53 Raptor emhttp: shcmd (818): sync May 21 15:32:56 Raptor emhttp: shcmd (819): set -o pipefail ; umount /mnt/user |& logger May 21 15:32:56 Raptor shfs/user: fuse_main exit: 0 May 21 15:32:56 Raptor emhttp: shcmd (820): rmdir /mnt/user |& logger May 21 15:32:56 Raptor emhttp: shcmd (821): set -o pipefail ; umount /mnt/user0 |& logger May 21 15:32:56 Raptor shfs/user0: fuse_main exit: 0 May 21 15:32:56 Raptor emhttp: shcmd (822): rmdir /mnt/user0 |& logger May 21 15:32:56 Raptor emhttp: shcmd (823): rm -f /boot/config/plugins/dynamix/mover.cron May 21 15:32:56 Raptor emhttp: shcmd (824): /usr/local/sbin/update_cron &> /dev/null May 21 15:32:56 Raptor emhttp: Unmounting disks... May 21 15:32:56 Raptor emhttp: shcmd (825): umount /mnt/disk1 |& logger May 21 15:32:57 Raptor emhttp: shcmd (826): rmdir /mnt/disk1 |& logger May 21 15:32:57 Raptor emhttp: shcmd (827): umount /mnt/disk2 |& logger May 21 15:32:58 Raptor emhttp: shcmd (828): rmdir /mnt/disk2 |& logger May 21 15:32:58 Raptor emhttp: shcmd (829): umount /mnt/disk3 |& logger May 21 15:32:58 Raptor root: umount: /mnt/disk3: target is busy May 21 15:32:58 Raptor root: (In some cases useful info about processes that May 21 15:32:58 Raptor root: use the device is found by lsof( or fuser(1).) May 21 15:32:58 Raptor emhttp: shcmd (830): umount /mnt/cache |& logger May 21 15:32:58 Raptor emhttp: shcmd (831): rmdir /mnt/cache |& logger May 21 15:32:58 Raptor emhttp: Retry unmounting disk share(s)...
May 21, 201610 yr I saw the lines about your flash but those weren't the lines you quoted earlier. The inability to unmount disk3 possibly just means something that had files open on it wouldn't stop. Have you tried the Open Files plugin? As for your free space you still might try checking the filesystem. If it was corrupted by any of this I can see how things might be reported differently than expected.
May 21, 201610 yr Author I saw the lines about your flash but those weren't the lines you quoted earlier. The inability to unmount disk3 possibly just means something that had files open on it wouldn't stop. Have you tried the Open Files plugin? As for your free space you still might try checking the filesystem. If it was corrupted by any of this I can see how things might be reported differently than expected. Hey, I've not used that before - I've never ever had an issue with stopping my array, including whilst being in the beta, up until today. I've finished checking the filesystem - There were no errors detected in the BTRFS filesystem
Archived
This topic is now archived and is closed to further replies.