syntheticfaith Posted January 27, 2011 Posted January 27, 2011 Due to the snow, we had a power outage, and now my users shares are missing. The server rebooted and ran a parity check which was successful (no errors) but only one share was showing up. That share (new) is used by sabnzbd so maybe it was recreated? It is not displaying the data that was in it though (previous downloads), only new downloads since reboot. All the disk utilization seems normal though, and I can browse individual disk shares successfully. Anyway to get the shares back, I have not tried recreating them, because it seems like "new" share was recreated but didn't display data. 4.4.2 pro. Thanks for any pointers. syslog.txt
prostuff1 Posted January 27, 2011 Posted January 27, 2011 Due to the snow, we had a power outage, and now my users shares are missing. The server rebooted and ran a parity check which was successful (no errors) but only one share was showing up. That share (new) is used by sabnzbd so maybe it was recreated? It is not displaying the data that was in it though (previous downloads), only new downloads since reboot. All the disk utilization seems normal though, and I can browse individual disk shares successfully. Anyway to get the shares back, I have not tried recreating them, because it seems like "new" share was recreated but didn't display data. 4.4.2 pro. Thanks for any pointers. If you can post a syslog (instructions via the troubleshooting link in my signature) that would go a long way towards helping us figure out what is going on. After you have done that we can give more exact advice, but I think a proper stop of the array and restart might bring the shares back.
Joe L. Posted January 27, 2011 Posted January 27, 2011 Due to the snow, we had a power outage, and now my users shares are missing. The server rebooted and ran a parity check which was successful (no errors) but only one share was showing up. That share (new) is used by sabnzbd so maybe it was recreated? It is not displaying the data that was in it though (previous downloads), only new downloads since reboot. All the disk utilization seems normal though, and I can browse individual disk shares successfully. Anyway to get the shares back, I have not tried recreating them, because it seems like "new" share was recreated but didn't display data. 4.4.2 pro. Thanks for any pointers. If you can post a syslog (instructions via the troubleshooting link in my signature) that would go a long way towards helping us figure out what is going on. After you have done that we can give more exact advice, but I think a proper stop of the array and restart might bring the shares back. The transactions in the file-system journals were being replayed prior to them being mounted. These lines in the syslog show that activity: Jan 26 20:54:15 Tower kernel: ReiserFS: md2: replayed 52 transactions in 142 seconds Jan 26 20:54:17 Tower kernel: ReiserFS: md2: Using r5 hash to sort names Jan 26 20:54:28 Tower kernel: can't shrink filesystem on-line Jan 26 20:56:22 Tower ntpd[1276]: synchronized to 149.20.54.20, stratum 2 Jan 26 21:03:32 Tower kernel: ReiserFS: md5: replayed 181 transactions in 699 seconds Jan 26 21:03:33 Tower kernel: ReiserFS: md5: Using r5 hash to sort names Jan 26 21:03:40 Tower kernel: can't shrink filesystem on-line Jan 26 21:04:11 Tower kernel: ReiserFS: md4: replayed 193 transactions in 738 seconds Jan 26 21:04:14 Tower kernel: ReiserFS: md4: Using r5 hash to sort names Jan 26 21:04:19 Tower kernel: can't shrink filesystem on-line Jan 26 21:04:31 Tower kernel: ReiserFS: md6: replayed 220 transactions in 758 seconds Jan 26 21:04:31 Tower kernel: ReiserFS: md6: Using r5 hash to sort names Jan 26 21:04:32 Tower kernel: can't shrink filesystem on-line Jan 26 21:06:13 Tower kernel: ReiserFS: md3: replayed 443 transactions in 860 seconds Jan 26 21:06:13 Tower kernel: ReiserFS: md3: Using r5 hash to sort names Jan 26 21:06:14 Tower kernel: can't shrink filesystem on-line Jan 26 21:06:55 Tower kernel: ReiserFS: md1: replayed 617 transactions in 902 seconds The longest seemed to be just over 900 seconds. (15 minutes) It is easy to panic in those 15 minutes thinking all your files are gone. During this time, the associated files (and shares) would not be visible, as the disks were not yet mounted. If you refresh your browser, odds are they will be visible now. (You might have to "Stop" the array and re-start it to see the user-shares) Joe L.
syntheticfaith Posted January 27, 2011 Author Posted January 27, 2011 Well that's embarrassing. I stopped and started the array, and no dice. But afterwards, I just rebooted the server and they were there. Many thanks to everyone for checking this out.
Stucco Posted June 7, 2011 Posted June 7, 2011 Same thing happened to me. Power outage, rebooted, parity check started and most shares were missing. I waited an hour, still no shares, parity check at ~20%. I stopped the check, and attempted to shut down the array. It read "unmounting" on all but one drive for about 10 minutes, and then the web server stopped responding. Logged in at the console and ran a shutdown now. It went through the routine, logging me out, and then came back to the login prompt, and never shut down. I pressed and held the power button, waited a few minutes, powered it back on and waited for it to boot. On boot completion, immediately all of my shares were available and another parity check had begun. I don't know why all of the above happened, but a reboot, however achieved, solved my problem.
Recommended Posts
Archived
This topic is now archived and is closed to further replies.