Ubuntu 12.04 unstable: not unmounting root hdd cleanly on shutdown and reboot

Asked by Jappreet on 2012-03-05

on 12.04 beta, when i give a shutdown command or reboot command, or i select from menu, both ways the system shuts down. But on the next boot up i get the error msg like this:

/dev/sdb1 not cleanly unmounted, check forced

I think that the system is not getting shutdown properly, which was okay till 11.10.

Can anybody pls help, as i fear it shud not damage my hdd due to excessive unsafe unmounts.

Question information

Have you fsck'd the partition in liveCD?
Is the RAM healthy?

Jappreet (bains-jappi) said : #2

it is fsck'd every time during boot up and reports no errors on the gui.

but i ve not used fsck from live cd as i upgraded from 11.10 directly using update-manager.

RAM also is healthy as no errors have been reported till now/ or rather i ll say i ve not come accross any error for RAM.

Jappreet (bains-jappi) said : #3

Some additional which i m feeling shud show something more, but m not very sure. Pls have a look:

root@jappreet-desktop:/home/jappreet# shutdown -k -v -h now
root@jappreet-desktop:/home/jappreet#
Broadcast message from jappreet@jappreet-desktop
 (/dev/pts/0) at 16:57 ...

The system is going down for halt NOW!

root@jappreet-desktop:/home/jappreet#

Is this response to given command ok or something else shud be there....

Test the RAM via memtest in Grub, even a 10 minute test will show errors up if they exist. You can use custom options in the live CD, there may be some issues which the default check is not fixing...

Jappreet (bains-jappi) said : #5

hi

I did the memtest, RAM is OK no errors reported.

Did fsck from live CD as well, initialy it gave the same msg, ie

/dev/sdb1 not cleanly unmounted, check forced

and then went through all passes successfully, without reporting any errors.

And one more this time wen i booted up back after using fsck on live CD the system booted up without any disk check being forced. So I feel there is something which is not unmounting the sytem properly.

Pls reconsider and suggest any further checks.

Does the system have a make and model?

You could grab the ultimate boot cd and use the manufacturers tool to test the drive at a low level.

Jappreet (bains-jappi) said : #7

Yeah my system is an Intel D102GGC2 board based, Pentium 4 CPU @ 3.06 GHz, 1 GB RAM, 3 HDDS while i use only /dev/sdb1 to be mounted as root partition (ext2).

The subject hdd has two more partitions in addition to the nomal swap partition. One being ext2 and other being FAT32. I dont even mount those partitions still i get this problem.

Also the system is dual boot with Win XP on another Hdd and I get this problem even when i use Ubuntu regularly and not even boot into Win XP at all.

Jappreet (bains-jappi) said : #8

While trying to figure out what is wrong, i have observed that the machine shutdown in a flash as soon as shutdown command is clicked or button is pressed. Which is quite fast than wat used to be previously. So i feel there is some problem in the shutdown scripts/ Upstart jobs, which may pls b looked into.

Thanks

Jappreet (bains-jappi) said : #9

a copy of the kern.log file for a shutdown is pasted below, if it gives any hint:
Mar 6 12:03:34 jappreet-desktop kernel: [ 167.355322] process `skype' is using obsolete setsockopt SO_BSDCOMPAT
Mar 6 14:35:34 jappreet-desktop kernel: [ 9287.176179] unity-2d-shell[1574]: segfault at 151 ip 00000151 sp bff7b66c error 4 in libdconf-qt.so.0.0.0[110000+b000]
Mar 6 14:35:43 jappreet-desktop kernel: [ 9296.700502] init: tty4 main process (1052) killed by TERM signal
Mar 6 14:35:43 jappreet-desktop kernel: [ 9296.703347] init: tty5 main process (1057) killed by TERM signal
Mar 6 14:35:43 jappreet-desktop kernel: [ 9296.704975] init: tty2 main process (1067) killed by TERM signal
Mar 6 14:35:43 jappreet-desktop kernel: [ 9296.706320] init: tty3 main process (1068) killed by TERM signal
Mar 6 14:35:43 jappreet-desktop kernel: [ 9296.707491] init: tty6 main process (1071) killed by TERM signal
Mar 6 14:35:43 jappreet-desktop kernel: [ 9296.708887] init: cron main process (1088) killed by TERM signal
Mar 6 14:35:43 jappreet-desktop kernel: [ 9296.710024] init: tty1 main process (1280) killed by TERM signal
Mar 6 14:35:44 jappreet-desktop kernel: [ 9297.180615] init: plymouth-upstart-bridge main process (4734) terminated with status 1
Mar 6 14:35:44 jappreet-desktop kernel: [ 9297.678411] init: rfkill-store main process (4727) terminated with status 1
Mar 6 14:35:44 jappreet-desktop kernel: [ 9297.679147] init: usplash main process (4744) terminated with status 127
Mar 6 14:35:44 jappreet-desktop kernel: Kernel logging (proc) stopped.

Jappreet (bains-jappi) said : #10

Reverted back to 11.10, will wait for launch of 12.04...... no problems as usual in 11.10....

paolode (paolo-debortoli) said : #11

a bit strange behaviour. I am using ubuntu 12.04 on hp pavilion laptop with cinnamon, nemo, nautilus (but he behaviour is showed in other cases). I have external usb hdd and a crypted partition (luks format) on internal hd. when I unmount external hdds it doesn't show any message like 'it is now safe to remove drive' and sometimes says 'failed to unmount'. when I unmount the crypted partition it says always 'failed to unmount'. however hdds and crypted partition are always unmounted. it seems that the system fails to check and close applications and processes related to hdds and partition. I have another normal xfs backup partition on internal hd, which doesn't show such problems in unmounting.

Mark Lopez (silvenga) said : #12

The last update was a two years ago. I'm still running Ubuntu 12.04 using EXT4 with journaling disabled. When I perform a reboot a fsck check is forced due to incomplete unmounting. The fsck finds incorrect, deleted inode times, fixes them, then reboots.

Has there been a proposed solution?

Can you help with this problem?

Provide an answer of your own, or ask Jappreet for more information if necessary.

To post a message you must log in.