missing user shares

[add212]

Two days ago the system crashed, low and behold had bad ram.  Removed it, restarted and finally bio posted.  After the reboot some user shares were missing and decided to wait it out with the parity check thinking it would resolve itself.  Well user shares still missing after reading some other posts of rebooting, restarting several times.  Im out of ideas of what to search for on the forums.

 

 

log attached, also Ive noticed when trying to stop the array It always gets stuck because of mysql, how do I disable and remove it? Wanted to try it out with xbmc but not worth it for me atm.

 

thanks

syslog.zip

[Joe L.]

Two days ago the system crashed, low and behold had bad ram.  Removed it, restarted and finally bio posted.  After the reboot some user shares were missing and decided to wait it out with the parity check thinking it would resolve itself.  Well user shares still missing after reading some other posts of rebooting, restarting several times.  Im out of ideas of what to search for on the forums.

 

 

log attached, also Ive noticed when trying to stop the array It always gets stuck because of mysql, how do I disable and remove it? Wanted to try it out with xbmc but not worth it for me atm.

 

thanks

Your file systems are probably re-playing transactions from before your hard-crash.  That can take 30 minutes or more with large disks and lots of activity. wait a while, the stop and restart your array. the shares should re-appear.

 

You must stop the sql server before attempting to stop the array, otherwise it will wait forever for the disks to be idle, and forever takes a very long time, even with today's faster CPUs.

 

log in via telnet, or on the system  console and type:

killall mysqld

to terminate mysql before you attempt to stop the array.

[add212]

thanks for the quick response.  Gone through many reboots every time it hangs on unmounting, lsof | grep /mnt shows sabnzbd running so that one was killed.  Unmenu shows nothing running and lsof | grep /mnt is blank.  After about 10mins the management gui hangs and Im guessing crashes as a web refreshes to an error page.

 

 

UPDATE

 

ok finally got it to unmount seems my ssd cache drive is now unformatted.  When I try to format I get this "HTTP/1.1 200 OK Server: emhttp Cache-Control: no-cache Content-Type: text/html; charset=utf-8" at the top of the screen and will not continue with formatting

[add212]

ok ssd cache disk has been removed and parity has finished.  I think im screwed the only two plugins I know of running are sickbeard and sabnzbd which have both been stopped in order to see if the plugin was the problem like some other postings.

 

off topic but If i choose to start a fresh usb flash with an updated version will I still lose the files? reason Im asking is because Im thinking of rebuilding but fear of losing 3tb worth of data, or can I just browse the shares since I can still see the files on disk1 and 2 and copy them off?

 

thanks again hope something fixes itself

[lionelhutz]

Wow, those packages like fighting each other. I think your system went back and forth installing 2 different versions of python about 6 times on booting.

 

I wonder if a package is creating a directory in the /mnt/user location and screwing up your filesystem.

 

Rename the packages directory on the flash drive to something else to disable all the packages.

 

Then, make sure the go file (stored in the config directory) has these 2 lines only;

#!/bin/bash

# Start the management Utiity

/usr/local/sbin/emhttp &

 

Actually, you can just put a # in front of any other lines to comment them out.

 

Then see what happens when you boot.