Hi Gang-

Running 4.5.6

We had a power outage last night. Upon turning unraid back on parity rebuild began, but no user shares were available, nor listed in the share list, even after 12 hours. Per http://lime-technology.com/forum/index.php?topic=4132.0 this potentially looks like 'expected' behavior, but I don't get it... if you're running anything off user shares this basically renders the system unusable the entire time the parity is doing its thing.

 

Is there any way or setting around this? This seems a poor user experience.

 

 

 

 

 

Hi Gang-

Running 4.5.6

We had a power outage last night. Upon turning unraid back on parity rebuild began, but no user shares were available, nor listed in the share list, even after 12 hours. Per http://lime-technology.com/forum/index.php?topic=4132.0 this potentially looks like 'expected' behavior, but I don't get it... if you're running anything off user shares this basically renders the system unusable the entire time the parity is doing its thing.

 

Is there any way or setting around this? This seems a poor user experience.

It is not normal.

 

Post a system log. (instructions under "Troubleshooting" in the wiki)

There is something else going on affecting your server.

 

Joe L.

No user shares is not normal after a powerdown, but I think it happened to me once or twice when I was testing unRAID.

 

The moral of the story is get a Battery Backup and set it up so the server shuts down gracefully.

 

I'm not positive on the cause of this but if i had to guess it might reside around the replay of transactions of the RFS and when user shares tries to come online.

 

A syslog would be very helpful in this situation.

Thanks for your input. Good to know this is not right.

Log:

http://dl.dropbox.com/u/118181/syslog-2010-11-17.txt

 

 

I killed the parity check and rebooted. Everything came back up fine.

Odd none the less

 

This is the only part in the log that seems to reference the samba shares

Nov 17 07:00:42 Media emhttp: shcmd (61): rm /etc/samba/smb-shares.conf >/dev/null 2>&1
Nov 17 07:00:42 Media emhttp: _shcmd: shcmd (61): exit status: 1

 

Do you see anything else?

 

I killed the parity check and rebooted. Everything came back up fine.

Odd none the less

 

This is the only part in the log that seems to reference the samba shares

Nov 17 07:00:42 Media emhttp: shcmd (61): rm /etc/samba/smb-shares.conf >/dev/null 2>&1
Nov 17 07:00:42 Media emhttp: _shcmd: shcmd (61): exit status: 1

 

Do you see anything else?

 

I see all the lines in your syslog showing it attempting to un-mount /mnt/user

Apparently you were starting a program using /mnt/user, and the unRAID server had not yet set it up after rebooting after your power failure.  This kept it in a loop trying every 5 seconds to un-mount the directory.  It could not because it was busy.

 

You'll need to wait until the array is on-line before starting add-on services attempting to use user-shares.

Nov 17 16:00:11 Media emhttp: shcmd (80): umount /mnt/user >/dev/null 2>&1

Nov 17 16:00:11 Media emhttp: _shcmd: shcmd (80): exit status: 1

Nov 17 16:00:11 Media emhttp: shcmd (81): rmdir /mnt/user >/dev/null 2>&1

Nov 17 16:00:11 Media emhttp: _shcmd: shcmd (81): exit status: 1

Nov 17 16:00:11 Media emhttp: Retry unmounting user share(s)...

Nov 17 16:00:16 Media emhttp: shcmd (82): umount /mnt/user >/dev/null 2>&1

Nov 17 16:00:16 Media emhttp: _shcmd: shcmd (82): exit status: 1

Nov 17 16:00:16 Media emhttp: shcmd (83): rmdir /mnt/user >/dev/null 2>&1

Nov 17 16:00:16 Media emhttp: _shcmd: shcmd (83): exit status: 1

Nov 17 16:00:16 Media emhttp: Retry unmounting user share(s)...

Nov 17 16:00:21 Media emhttp: shcmd (84): umount /mnt/user >/dev/null 2>&1

Nov 17 16:00:21 Media emhttp: _shcmd: shcmd (84): exit status: 1

Nov 17 16:00:21 Media emhttp: shcmd (85): rmdir /mnt/user >/dev/null 2>&1

Nov 17 16:00:21 Media emhttp: _shcmd: shcmd (85): exit status: 1

Nov 17 16:00:21 Media emhttp: Retry unmounting user share(s)...

Nov 17 16:00:26 Media emhttp: shcmd (86): umount /mnt/user >/dev/null 2>&1

Nov 17 16:00:26 Media emhttp: _shcmd: shcmd (86): exit status: 1

Nov 17 16:00:26 Media emhttp: shcmd (87): rmdir /mnt/user >/dev/null 2>&1

Nov 17 16:00:26 Media emhttp: _shcmd: shcmd (87): exit status: 1

Nov 17 16:00:26 Media emhttp: Retry unmounting user share(s)...

Nov 17 16:00:31 Media emhttp: shcmd (88): umount /mnt/user >/dev/null 2>&1

Nov 17 16:00:31 Media emhttp: _shcmd: shcmd (88): exit status: 1

Nov 17 16:00:31 Media emhttp: shcmd (89): rmdir /mnt/user >/dev/null 2>&1

Nov 17 16:00:31 Media emhttp: _shcmd: shcmd (89): exit status: 1

Nov 17 16:00:31 Media emhttp: Retry unmounting user share(s)...

Interesting. Around 16:00 I did try to unmount and reboot remotely, but I forgot Crashplan was still running and had created its own user share thus it was ungraceful. Still, that was 10 hours or so after the initial boot so I'm not sure what took place on bootup.

No big deal really, I'll just make sure my user shares come up before I leave for the day next time (and maybe thorw a UPS on the sucker for a more graceful shutdown)

 

I'm calling no harm no foul on this one

I had the same problem that my user share didnt mount....

 

So inorder to solve this....

 

do i just put a sleep command before i startup sabnzbd, sickbeard & couchpotato?

or is the a more better way to do it like for example by making it check all shares

are online before running the go script?

A delay will typically solve this except on the off chance that a particular drive takes an extra long time to mount, in which case a restart my be required.