Filesystem Check Failed

zuargo · 2009-09-03 16:44:34

Hi

I installed Archlinux and when the system starts (after grub) sometimes says this:

FILESYSTEM CHECK FAILED

give root password for maintenance or control+D for restart

this happens once of each twice that I start the system

I tried removing and installing again initiscripts (http://bbs.archlinux.org/viewtopic.php?id=56822):

ifconfig eth0 up
dhcpcd eth0
pacman -Syu
pacman -Rs initscripts
pacman -S initscripts
exit

but this does not solved the issue

so, I tried run fsck but the problem still exist

maybe I must do this: http://www.archlinux.org/news/411/

here fidks -l command output:

Disk /dev/sda: 82.3 GB, 82348277760 bytes
255 heads, 63 sectors/track, 10011 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0xa72aa72a

   Device Boot      Start         End      Blocks   Id  System
/dev/sda1               1        2732    21944758+  83  Linux
/dev/sda3            2733        2868     1092420   82  Linux swap / Solaris
/dev/sda4            2869       10011    57376147+  83  Linux

Disk /dev/sdb: 250.1 GB, 250059350016 bytes
255 heads, 63 sectors/track, 30401 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Disk identifier: 0x0005a12e

   Device Boot      Start         End      Blocks   Id  System
/dev/sdb1               1       30401   244196001   83  Linux

the root partition is /dev/sda1

any idea?

sorry for my english but it is not my first language

Last edited by zuargo (2009-09-03 16:54:19)

djszapi · 2009-09-04 21:34:30

Hello zuargo!

Can you show the output of /etc/fstab, blkid, in parted command print ?

zuargo · 2009-09-05 18:35:27

/etc/fstab

#
# /etc/fstab: static file system information
#
# <file system>        <dir>         <type>    <options>          <dump> <pass>
none                   /dev/pts      devpts    defaults            0      0
none                   /dev/shm      tmpfs     defaults            0      0

#/dev/cdrom             /media/cd   auto    ro,user,noauto,unhide   0      0
#/dev/dvd               /media/dvd  auto    ro,user,noauto,unhide   0      0
#/dev/fd0               /media/fl   auto    user,noauto             0      0

/dev/sda1 / ext4 defaults 0 1
/dev/sda3 swap swap defaults 0 0
/dev/sda4 /home ext4 defaults 0 1
/dev/sdb1 /media/sata ext4 defaults 0 1

blkid

[root@zuargo-archlinux ~]# blkid
/dev/sda1: UUID="2727445b-47fa-4ffb-b70d-370a13651a4e" TYPE="ext4"
/dev/sda3: UUID="2222e40b-f06b-496e-bf29-4845a8f4d164" TYPE="swap"
/dev/sda4: UUID="f4129b69-5944-476f-8db7-668cb5e81679" TYPE="ext4"
/dev/sdb1: UUID="3a2a074a-ef50-4faa-94e8-b53b3de51508" TYPE="ext4"
[root@zuargo-archlinux ~]#

in parted commando print? sorry I don't understand what command exactly?

thanks for your interest

zuargo · 2009-09-22 02:55:59

can someone give me some idea?

foutrelis · 2009-09-22 03:22:50

Device nodes (/dev/sdX) are assigned at random to hard disks. You shouldn't use them in configuration files like /etc/fstab. Instead, you can use UUIDs which are guaranteed to be the same on every boot.

For example, the fstab entry for your root partition:

/dev/sda1 / ext4 defaults 0 1

becomes:

UUID=2727445b-47fa-4ffb-b70d-370a13651a4e / ext4 defaults 0 1

Same for the rest of your computer's partitions.

Leonid.I · 2009-09-22 15:15:38

I had the same error yesterday, after the kernel update (to 2.6.30.6). After several reboots every partition was checked and the system is now operational again. My fstab is a standard one:

# 
# /etc/fstab: static file system information
#
# <file system>        <dir>         <type>    <options>          <dump> <pass>
none                   /dev/pts      devpts    defaults            0      0
none                   /dev/shm      tmpfs     defaults            0      0

#/dev/cdrom             /media/cd   auto    ro,user,noauto,unhide   0      0
#/dev/dvd               /media/dvd  auto    ro,user,noauto,unhide   0      0
#/dev/fd0               /media/fl   auto    user,noauto             0      0

UUID=0ee865a6-6f3c-4a9c-a0b9-34f677d6db71 /tmp ext4 defaults,noexec 0 1
UUID=1bb7226e-890e-455c-9df7-199125e7c1a5 /boot ext4 defaults 0 1
UUID=3f79d443-312d-42eb-a59a-72b4e9539072 /home ext4 defaults 0 1
UUID=639adcfa-3920-4416-92c9-ae1db027735b /var ext4 defaults 0 1
UUID=923db624-637f-457c-a90c-d3f8f4a5d982 swap swap defaults 0 0
UUID=f6c29086-f6db-4f62-bd5e-080fcea0050d / ext4 defaults 0 1

and it worked perfectly well before...

Does anyone have an idea as to why this happens?

Leonid.I · 2009-09-22 17:11:12

It seems that the bad superbolock error, like this

The superblock could not be read or does not describe a correct ext4
filesystem. If the device is valid and it really contains an ext4
filesystem (and not swap or ufs or something else), then the superblock
is corrupt, and you might try running e2fsck with an alternate superblock:
    e2fsck -b 8193 <device>

******************* FILESYSTEM CHECK FAILED *************
* Please repair manually and reboot. Note that the root *
* filesystem is currently mounted read-only. To remount *
* it read-write type: mount -n -o remount,rw /          *
* When you exit the mantenance shel the system will     *
* reboot automatically.                                 *
*********************************************************

Give root password for mantenance

is due to an update in udev (which I made yesterday): see http://www.linode.com/forums/viewtopic.php?p=23679.

In this case one needs a reboot to fix the superblock... Can anyone confirm this?

zuargo · 2009-09-23 03:22:49

foutrelis wrote:

Device nodes (/dev/sdX) are assigned at random to hard disks. You shouldn't use them in configuration files like /etc/fstab. Instead, you can use UUIDs which are guaranteed to be the same on every boot.
For example, the fstab entry for your root partition:
/dev/sda1 / ext4 defaults 0 1
becomes:
UUID=2727445b-47fa-4ffb-b70d-370a13651a4e / ext4 defaults 0 1
Same for the rest of your computer's partitions.

Leonid.I wrote:

I had the same error yesterday, after the kernel update (to 2.6.30.6). After several reboots every partition was checked and the system is now operational again. My fstab is a standard one:

Code:
# 
# /etc/fstab: static file system information
#
# <file system>        <dir>         <type>    <options>          <dump> <pass>
none                   /dev/pts      devpts    defaults            0      0
none                   /dev/shm      tmpfs     defaults            0      0

#/dev/cdrom             /media/cd   auto    ro,user,noauto,unhide   0      0
#/dev/dvd               /media/dvd  auto    ro,user,noauto,unhide   0      0
#/dev/fd0               /media/fl   auto    user,noauto             0      0

UUID=0ee865a6-6f3c-4a9c-a0b9-34f677d6db71 /tmp ext4 defaults,noexec 0 1
UUID=1bb7226e-890e-455c-9df7-199125e7c1a5 /boot ext4 defaults 0 1
UUID=3f79d443-312d-42eb-a59a-72b4e9539072 /home ext4 defaults 0 1
UUID=639adcfa-3920-4416-92c9-ae1db027735b /var ext4 defaults 0 1
UUID=923db624-637f-457c-a90c-d3f8f4a5d982 swap swap defaults 0 0
UUID=f6c29086-f6db-4f62-bd5e-080fcea0050d / ext4 defaults 0 1

and it worked perfectly well before...

Does anyone have an idea as to why this happens?

thaaaaaaaaaaaanks apparently that works... I rebooted the computer thrice whit that changes in the fstab and don't have problem

Leonid.I wrote:

It seems that the bad superbolock error, like this

Code:
The superblock could not be read or does not describe a correct ext4
filesystem. If the device is valid and it really contains an ext4
filesystem (and not swap or ufs or something else), then the superblock
is corrupt, and you might try running e2fsck with an alternate superblock:
    e2fsck -b 8193 <device>

******************* FILESYSTEM CHECK FAILED *************
* Please repair manually and reboot. Note that the root *
* filesystem is currently mounted read-only. To remount *
* it read-write type: mount -n -o remount,rw /          *
* When you exit the mantenance shel the system will     *
* reboot automatically.                                 *
*********************************************************

Give root password for mantenance

is due to an update in udev (which I made yesterday): see http://www.linode.com/forums/viewtopic.php?p=23679

In this case one needs a reboot to fix the superblock... Can anyone confirm this?

right, that is the problem...

thanks for your answer guys

fettouhi · 2009-09-23 19:57:05

I have a similar problem with my RAID partition. I also have to boot 2 times in order to log into my machine. In my case it complains about last write time being off and at shutdown lately my partition (root) is always busy. Using dmraid 1.0.0.rc14-4.

Regards

André

Leonid.I · 2009-09-23 21:52:48

fettouhi wrote:

I have a similar problem with my RAID partition. I also have to boot 2 times in order to log into my machine. In my case it complains about last write time being off and at shutdown lately my partition (root) is always busy. Using dmraid 1.0.0.rc14-4.
Regards
André

I'm just curious, did you try to synchronize the Linux kernel clock with NTP servers, then sync the hardware clock with the system time and put HARDWARECLOCK="localtime" into rc.conf?

fettouhi · 2009-09-24 11:59:59

I already have HARDWARECLOCK="localtime" in the rc.conf but how do I sync the hardware clock to the system time and sync kernel clock ntp servers.

Regards

André

Leonid.I · 2009-09-24 19:06:43

fettouhi wrote:

I already have HARDWARECLOCK="localtime" in the rc.conf but how do I sync the hardware clock to the system time and sync kernel clock ntp servers.
Regards
André

openntpd keep the system time right; hwclock -w syncs hardware clock to the linux time; hwclock -s -- vice versa;
hwclock -r shows the current hardware time.

The reason for my question is that I recall that just before encountering that error, I found out that my time was -4 hrs from the true (Indianapolis) time.

I did the same upgrade on a different arch machine and it went smooth, so I really can not blame anything except particular settings...

L.

Last edited by Leonid.I (2009-09-24 19:08:36)

fettouhi · 2009-09-25 07:19:09

I tried syncing up the clocks and still get the error.

Regards

André

kgas · 2009-09-25 08:06:29

Leonid.I, I confirm to your post #7. After a couple of reboot system is up and normal.

Leonid.I · 2009-09-25 18:11:48

kgas wrote:

Leonid.I, I confirm to your post #7. After a couple of reboot system is up and normal.

So, due to the lack of a better explanation, I would blame wrong time settings on one of my machines. It's either that or a particular hardware, which is prety unlikely ...

The problem is that this happens before the syslon-ng is up, and I would really love to have a look into some log files

People from SuSE camp have /var/log/boot.msg, where the boot process is logged.
Is there any way to dump boot logs into a text file before the syslog daemon starts?

Then we could file a bug against some package...

Thanx.

Arch Linux

#1 2009-09-03 16:44:34

Filesystem Check Failed

#2 2009-09-04 21:34:30

Re: Filesystem Check Failed

#3 2009-09-05 18:35:27

Re: Filesystem Check Failed

#4 2009-09-22 02:55:59

Re: Filesystem Check Failed

#5 2009-09-22 03:22:50

Re: Filesystem Check Failed

#6 2009-09-22 15:15:38

Re: Filesystem Check Failed

#7 2009-09-22 17:11:12

Re: Filesystem Check Failed

#8 2009-09-23 03:22:49

Re: Filesystem Check Failed

#9 2009-09-23 19:57:05

Re: Filesystem Check Failed

#10 2009-09-23 21:52:48

Re: Filesystem Check Failed

#11 2009-09-24 11:59:59

Re: Filesystem Check Failed

#12 2009-09-24 19:06:43

Re: Filesystem Check Failed

#13 2009-09-25 07:19:09

Re: Filesystem Check Failed

#14 2009-09-25 08:06:29

Re: Filesystem Check Failed

#15 2009-09-25 18:11:48

Re: Filesystem Check Failed

Board footer