harddisk failure, or Xubuntu refuses to load
 

Dear all
I'm in a bind as Xubuntu refuses to boot up and I'm unable to pinpoint
the cause.

Here's what happened:
My Asus laptop was working on the table under Xubuntu Lucid when I
grabbed it as usual with one hand from below. Then I noticed that it
froze as, I assume, I touched (somewhere somehow and too much) the
harddrive. As it had frozen, I then powered it off by holding pressed
for 5sec the 'power on' button. When starting it again, it
automatically launches harddrive checks then throws up a myriad of
error messages on the virtual consoles.

To attempt to diagnose, I booted up on a LiveCD and checked my 3 Linux
partitions using e2fsck and GParted. In both cases the ext2/ext3
partitions turned out as 'clean'. Mounting the drives and accessing
the folder structure and individual files works fine, according to
limited testing. (Booting Windows Vista on the other side of the
computer works fine, too.) However, when I try to boot up Linux _from_
the disk, Ubuntu still automatically launches harddrive checks, throws
up a error messages on the virtual consoles and is unusably slow. I am
attaching a log of 'dmesg' [1]. Essentially the following error
message gets repeated again and again:
[ 1371.730197] ata1.00: exception Emask 0x0 SAct 0x1 SErr 0x0 action 0x0
[ 1371.730205] ata1.00: irq_stat 0x40000008
[ 1371.730211] ata1.00: failed command: READ FPDMA QUEUED
[ 1371.730225] ata1.00: cmd 60/08:00:1d:e6:f5/00:00:0b:00:00/40 tag 0
ncq 4096 in
[ 1371.730227] res 41/40:08:1e:e6:f5/00:00:0b:00:00/00 Emask
0x409 (media error) <F>
[ 1371.730234] ata1.00: status: { DRDY ERR }
[ 1371.730239] ata1.00: error: { UNC }
[ 1371.734703] ata1.00: configured for UDMA/133
[ 1371.734720] sd 0:0:0:0: [sda] Unhandled sense code
[ 1371.734725] sd 0:0:0:0: [sda] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
[ 1371.734733] sd 0:0:0:0: [sda] Sense Key : Medium Error [current] [descriptor]
[ 1371.734743] Descriptor sense data with sense descriptors (in hex):
[ 1371.734748] 72 03 11 04 00 00 00 0c 00 0a 80 00 00 00 00 00
[ 1371.734770] 0b f5 e6 1e
[ 1371.734779] sd 0:0:0:0: [sda] Add. Sense: Unrecovered read error -
auto reallocate failed
[ 1371.734789] sd 0:0:0:0: [sda] CDB: Read(10): 28 00 0b f5 e6 1d 00 00 08 00
[ 1371.734809] end_request: I/O error, dev sda, sector 200664606
[ 1371.734831] ata1: EH complete

So what's going on? My best (and hopeful) guess is that although the
harddisk and file systems are fine along with most of the underlying
data, when grabbing the laptop the OS was accessing/modifying several
files in particular got somehow corrupted and are now impeding a
proper power up of the system. (Although I could get onto the console
and to the GDM login screen, trying to log into X using any user name
takes an eternity and never completes.)

Could anyone suggest how to tackle this? Should I attempt to reinstall
Xubuntu? (I have /home and /boot on separate partitions.) Anything
else?

Regards
Liviu

[1] http://s000.tinyupload.com/index.php?file_id=93465134138264110255


--
Do you know how to read?
http://www.alienetworks.com/srtest.cfm
http://goodies.xfce.org/projects/applications/xfce4-dict#speed-reader
Do you know how to write?
http://garbl.home.comcast.net/~garbl/stylemanual/e.htm#e-mail

--
xubuntu-users mailing list
xubuntu-users@lists.ubuntu.com
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/xubuntu-users

harddisk failure, or Xubuntu refuses to load
 

On 14/07/12 19:48, Liviu Andronic
wrote:



Dear all
I'm in a bind as Xubuntu refuses to boot up and I'm unable to pinpoint
the cause.

Here's what happened:
My Asus laptop was working on the table under Xubuntu Lucid when I
grabbed it as usual with one hand from below. Then I noticed that it
froze as, I assume, I touched (somewhere somehow and too much) the
harddrive. As it had frozen, I then powered it off by holding pressed
for 5sec the 'power on' button. When starting it again, it
automatically launches harddrive checks then throws up a myriad of
error messages on the virtual consoles.

To attempt to diagnose, I booted up on a LiveCD and checked my 3 Linux
partitions using e2fsck and GParted. In both cases the ext2/ext3
partitions turned out as 'clean'. Mounting the drives and accessing
the folder structure and individual files works fine, according to
limited testing. (Booting Windows Vista on the other side of the
computer works fine, too.) However, when I try to boot up Linux _from_
the disk, Ubuntu still automatically launches harddrive checks, throws
up a error messages on the virtual consoles and is unusably slow. I am
attaching a log of 'dmesg' [1]. Essentially the following error
message gets repeated again and again:touch /forcefsck
[ 1371.730197] ata1.00: exception Emask 0x0 SAct 0x1 SErr 0x0 action 0x0
[ 1371.730205] ata1.00: irq_stat 0x40000008
[ 1371.730211] ata1.00: failed command: READ FPDMA QUEUED
[ 1371.730225] ata1.00: cmd 60/08:00:1d:e6:f5/00:00:0b:00:00/40 tag 0
ncq 4096 in
[ 1371.730227] res 41/40:08:1e:e6:f5/00:00:0b:00:00/00 Emask
0x409 (media error) <F>
[ 1371.730234] ata1.00: status: { DRDY ERR }
[ 1371.730239] ata1.00: error: { UNC }
[ 1371.734703] ata1.00: configured for UDMA/133
[ 1371.734720] sd 0:0:0:0: [sda] Unhandled sense code
[ 1371.734725] sd 0:0:0:0: [sda] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
[ 1371.734733] sd 0:0:0:0: [sda] Sense Key : Medium Error [current] [descriptor]
[ 1371.734743] Descriptor sense data with sense descriptors (in hex):
[ 1371.734748] 72 03 11 04 00 00 00 0c 00 0a 80 00 00 00 00 00
[ 1371.734770] 0b f5 e6 1e
[ 1371.734779] sd 0:0:0:0: [sda] Add. Sense: Unrecovered read error -
auto reallocate failed
[ 1371.734789] sd 0:0:0:0: [sda] CDB: Read(10): 28 00 0b f5 e6 1d 00 00 08 00
[ 1371.734809] end_request: I/O error, dev sda, sector 200664606
[ 1371.734831] ata1: EH complete
touch /forcefsck
So what's going on? My best (and hopeful) guess is that although the
harddisk and file systems are fine along with most of the underlying
data, when grabbing the laptop the OS was accessing/modifying several
files in particular got somehow corrupted and are now impeding a
proper power up of the system. (Although I could get onto the console
and to the GDM login screen, trying to log into X using any user name
takes an eternity and never completes.)

Could anyone suggest how to tackle this? Should I attempt to reinstall
Xubuntu? (I have /home and /boot on separate partitions.) Anything
else?

Regards
Liviu

[1] http://s000.tinyupload.com/index.php?file_id=93465134138264110255




Hi,

Although gdm fails to load, can you still login to a tty, and start
xorg the traditional way by issuing the 'startx' command? You could
also force a single fsck check on the next boot, by running touch
/forcefsck. Also, it would be worth seeing if booting into
single user mode is any faster - add the 'single' option at the end
of grub's boot command line, after 'quite splash'. Your log also
seems to mention that you have hard disks in a raid array - "raid6:
sse2x4*** 3968 MB/s" - if so, does the system boot up properly if
you use just one hard disk? Giving the output of smartctl --all
/dev/sda may also be useful, to examine the state of your hard
drive, although it does rely on the smartmontools package being
installed.



Joshua









--
xubuntu-users mailing list
xubuntu-users@lists.ubuntu.com
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/xubuntu-users

harddisk failure, or Xubuntu refuses to load
 

Hello
Thank you for the explanations. Please read below.


On Sat, Jul 14, 2012 at 7:17 PM, Joshua O'Leary
<joshua.oleary@btinternet.com> wrote:
> Although gdm fails to load,
>
It doesn't really fail, though it takes forever. X can start but it is
very, very slow. And after I enter the login credentials, I'm simply
not waiting for it to finish - it takes too long.


> can you still login to a tty, and start xorg the
> traditional way by issuing the 'startx' command? You could also force a
>
Fortunately I can login to a tty and thus obtain all the relevant
logs. But there is no need to 'startx' (see above).


> single fsck check on the next boot, by running touch /forcefsck. Also, it
>
I will try this tomorrow.


> would be worth seeing if booting into single user mode is any faster - add
> the 'single' option at the end of grub's boot command line, after 'quite
> splash'.
>
I checked 'edit' in grub and it seems to me that the 'kernel
(recovery)' option, or whatever it's called, and it contains 'ro
single', so I guess I already tried it. But it doesn't help. So far my
best option is to load the 'server' kernel which is already installed;
then quickly enough I can get to a tty.


> Your log also seems to mention that you have hard disks in a raid
> array - "raid6: sse2x4 3968 MB/s" - if so, does the system boot up
> properly if you use just one hard disk?
>
I am not sure what RAID is, so if I have it I didn't configure it
intentionally. I only have a SATA hard drive, on which I have a
logical /dev/sda3 partition which contains /dev/sda5 for /boot,
/dev/sda6 for / (root) and /dev/sda8 for /home. I always use a single
harddrive to boot.


> Giving the output of smartctl --all
> /dev/sda may also be useful, to examine the state of your hard drive,
> although it does rely on the smartmontools package being installed.
>
Please see the output attached. Should I run a more in depth SMART
test on the disks?

Regards
Liviu
--
xubuntu-users mailing list
xubuntu-users@lists.ubuntu.com
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/xubuntu-users

harddisk failure, or Xubuntu refuses to load
 

On Sat, Jul 14, 2012 at 9:17 PM, Joshua O'Leary
<joshua.oleary@btinternet.com> wrote:
> traditional way by issuing the 'startx' command? You could also force a
> single fsck check on the next boot, by running touch /forcefsck. Also, it
>
This didn't help either. After performing the checks there were no
obvious error messages. Just a (from memory):
mountall: fsck /boot returned status 1.

Then the booting process got bogged down as usual: endless error
messages related to the hard disk.

I took another look at the logs and I noticed two recurring lines:
end_request: I/O error, dev sda, sector 200664606

and:
end_request: I/O error, dev sda, sector 212886687

Does this tell us anything interesting? Regards
Liviu

--
xubuntu-users mailing list
xubuntu-users@lists.ubuntu.com
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/xubuntu-users

harddisk failure, or Xubuntu refuses to load
 

On Sat, Jul 14, 2012 at 9:17 PM, Joshua O'Leary
<joshua.oleary@btinternet.com> wrote:
> properly if you use just one hard disk? Giving the output of smartctl --all
> /dev/sda may also be useful, to examine the state of your hard drive,
>
I also attach another piece of information, namely the output of lshw.

Liviu
--
xubuntu-users mailing list
xubuntu-users@lists.ubuntu.com
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/xubuntu-users

harddisk failure, or Xubuntu refuses to load
 

On 15/07/12 08:27, Liviu Andronic
wrote:



On Sat, Jul 14, 2012 at 9:17 PM, Joshua O'Leary
<joshua.oleary@btinternet.com> wrote:


properly if you use just one hard disk? Giving the output of smartctl --all
/dev/sda may also be useful, to examine the state of your hard drive,



I also attach another piece of information, namely the output of lshw.

Liviu







"end_request: I/O error, dev sda, sector 212886687" - this suggests
that there could be a corruption of the filesystem or damage to the
disk itself. Although windows boots up ok, the damage may have
affected an area of the disk only used by Ubuntu. If you edit your
/etc/fstab and comment out other partitions (home etc.) to see if
booting works any better with just the one partition it would be
useful - then we would know which partitions are actually causing
the problem, or if it is an issue with the disk in general.



Joshua



--
xubuntu-users mailing list
xubuntu-users@lists.ubuntu.com
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/xubuntu-users

harddisk failure, or Xubuntu refuses to load
 

On 15/07/12 07:55, Liviu Andronic wrote:

end_request: I/O error, dev sda, sector 212886687
If you change the hard disk operating mode in the BIOS (e.g. ahci, ide
etc.) does it still have the same errors? Not all hard disks support
advanced features.



--
xubuntu-users mailing list
xubuntu-users@lists.ubuntu.com
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/xubuntu-users

harddisk failure, or Xubuntu refuses to load
 

On 14/07/12 23:25, Liviu Andronic
wrote:



Hello
Thank you for the explanations. Please read below.


On Sat, Jul 14, 2012 at 7:17 PM, Joshua O'Leary
<joshua.oleary@btinternet.com> wrote:


Although gdm fails to load,



It doesn't really fail, though it takes forever. X can start but it is
very, very slow. And after I enter the login credentials, I'm simply
not waiting for it to finish - it takes too long.




can you still login to a tty, and start xorg the
traditional way by issuing the 'startx' command? You could also force a



Fortunately I can login to a tty and thus obtain all the relevant
logs. But there is no need to 'startx' (see above).




single fsck check on the next boot, by running touch /forcefsck. Also, it



I will try this tomorrow.




would be worth seeing if booting into single user mode is any faster - add
the 'single' option at the end of grub's boot command line, after 'quite
splash'.



I checked 'edit' in grub and it seems to me that the 'kernel
(recovery)' option, or whatever it's called, and it contains 'ro
single', so I guess I already tried it. But it doesn't help. So far my
best option is to load the 'server' kernel which is already installed;
then quickly enough I can get to a tty.




Your log also seems to mention that you have hard disks in a raid
array - "raid6: sse2x4 3968 MB/s" - if so, does the system boot up
properly if you use just one hard disk?



I am not sure what RAID is, so if I have it I didn't configure it
intentionally. I only have a SATA hard drive, on which I have a
logical /dev/sda3 partition which contains /dev/sda5 for /boot,
/dev/sda6 for / (root) and /dev/sda8 for /home. I always use a single
harddrive to boot.




Giving the output of smartctl --all
/dev/sda may also be useful, to examine the state of your hard drive,
although it does rely on the smartmontools package being installed.



Please see the output attached. Should I run a more in depth SMART
test on the disks?

Regards
Liviu







"Should I run a more in depth SMART
test on the disks?" This may be helpful in order to work out what
area of the disk is actually causing the problem. Please could you
also give the output of sudo fdisk -l /dev/sda, this would
enable us to work out specifically which area of the disk is damaged
by comparing it to the sector read errors.


--
xubuntu-users mailing list
xubuntu-users@lists.ubuntu.com
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/xubuntu-users

harddisk failure, or Xubuntu refuses to load
 

On Sun, Jul 15, 2012 at 2:02 PM, Joshua O'Leary
<joshua.oleary@btinternet.com> wrote:
> "Should I run a more in depth SMART test on the disks?" This may be helpful
> in order to work out what area of the disk is actually causing the problem.
>
Any suggestion for a tutorial? I'm not very familiar with smartmontools.


> Please could you also give the output of sudo fdisk -l /dev/sda, this would
> enable us to work out specifically which area of the disk is damaged by
> comparing it to the sector read errors.
>
Please see attached. I used
fdisk -lu /dev/sda

Thus I get sizes in sectors instead of cylinders. As far as I can tell
the (obvious) error seems to be located on the / (root) partition.
Does this help?

Regards
Liviu
--
xubuntu-users mailing list
xubuntu-users@lists.ubuntu.com
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/xubuntu-users

harddisk failure, or Xubuntu refuses to load
 

On Sun, Jul 15, 2012 at 1:59 PM, Joshua O'Leary
<joshua.oleary@btinternet.com> wrote:
> If you change the hard disk operating mode in the BIOS (e.g. ahci, ide etc.)
> does it still have the same errors? Not all hard disks support advanced
> features.
>
To my utter surprise, this seems to have done the trick.

I accessed the BIOS and modified the disk operating mode from
'enhanced' to 'compatible', whatever that means. Now Xubuntu boots
quickly, as expected. I checked 'dmesg' and it's clean (there are no
relevant 'error' or 'sector' messages). All in all looks great!
Thanks!

The only issue that I notice is that 'xfce4-notes' refuses to start.
CPU goes to 50% (2 cores). I'm not sure why this happens, especially
since dmesg doesn't complain. But this is part of a bigger worry: As
much as I'm happy that the system boots up gracefully again, I'd like
to perform an audit to see if the disk is damaged and / or needs
changed. Any suggestions on this point?

Regards
Liviu

--
xubuntu-users mailing list
xubuntu-users@lists.ubuntu.com
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/xubuntu-users