Hard Disk Failure?
 

Hi all,

I am using SecureCRT. When I go inside the directory of the external HD, I do an ls, the following was displayed.

?--------- ? ? ? ? ? archive
?--------- ? ? ? ? ? backup

The archive and backup directory are blinking. When I cd into archive, it says permission is denied. I was using root.

Can someone tell me what is happening? Please help.

Regards,
Allan

Please add all info about the device and any history of problems, failed commands and whatever you think could be interesting. And is this by any chance a Reiser or XFS formatted partition?

I am not sure what type of partition is it. It's an external hard disk that I used to do a daily backup of important files. Can you guide me on how to check the partition type?

fdisk -l (as root)

What I recommend is that if it supports SMART, run:

Where /dev/sda is the device.

It also has the ability to run diagnostics test on it, see 'man smartctl'.

I rebooted the linux box. Now the drive is not showing when I do a df command. However, I can still access to the folder. Previously, it was mounted to /media. Now, when I go into /media, all the files are gone. That means the blinking folder were not showing at all. Can I still retrieve the data? It is very important. Please help...

The result of fdisk -l:

Disk /dev/sda: 73.2 GB, 73294413824 bytes
255 heads, 63 sectors/track, 8910 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

Device Boot Start End Blocks Id System
/dev/sda1 * 1 13 104391 83 Linux
/dev/sda2 14 8910 71465152+ 8e Linux LVM

Disk /dev/sdb: 219.8 GB, 219880095744 bytes
255 heads, 63 sectors/track, 26732 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

Device Boot Start End Blocks Id System
/dev/sdb1 * 1 26732 214724758+ 8e Linux LVM

The result of df:

Filesystem 1K-blocks Used Available Use% Mounted on
/dev/mapper/vol0-root
1032088 761836 217824 78% /
/dev/sda1 101086 31477 64390 33% /boot
/dev/mapper/vol1-data
144497320 80807696 56350744 59% /data
none 2073768 0 2073768 0% /dev/shm
/dev/mapper/vol0-home
5160576 43760 4854672 1% /home
/dev/mapper/vol0-mkcdrec
2064208 970312 989040 50% /mkcdrec
/dev/mapper/vol0-pronto
5160576 802812 4095620 17% /pro
/dev/mapper/vol1-test
61927420 10330964 48451240 18% /test
/dev/mapper/vol0-tmp 5160576 79684 4818748 2% /tmp
/dev/mapper/vol0-usr 5160576 3113044 1785388 64% /usr
/dev/mapper/vol0-var 5160576 437680 4460752 9% /var

The result of cat /etc/fstab:

/dev/vol0/root / ext3 defaults 1 1
LABEL=/boot /boot ext3 defaults 1 2
/dev/vol1/data /data ext3 defaults 1 2
none /dev/pts devpts gid=5,mode=620 0 0
none /dev/shm tmpfs defaults 0 0
/dev/vol0/home /home ext3 defaults 1 2
/dev/vol0/mkcdrec /mkcdrec ext3 defaults 1 2
/dev/vol0/pronto /pro ext3 defaults 1 2
none /proc proc defaults 0 0
none /sys sysfs defaults 0 0
/dev/vol1/test /test ext3 defaults 1 2
/dev/vol0/tmp /tmp ext3 defaults 1 2
/dev/vol0/usr /usr ext3 defaults 1 2
/dev/vol0/var /var ext3 defaults 1 2
/dev/vol0/swap swap swap defaults 0 0
/dev/sdc1 /media/archive ext2 suid,dev,exec 0 0

Rebooting usually doesn't make that kind of hardware problems disappear. Could check your syslog and dmesg now and at the time of the directory access problem for any related error messages.

I do not know what I am looking at when I vi the dmesg log file. Can someone guide me?

And the syslog file, there are more than 10 syslog files. Which one should I be looking at?

If you can't find anything odd maybe just attach one dmesg (if your distro saves copies, that is) and one syslog from the time you started experiencing problems?

Hi,

I got dmesg from /var/log and syslog from /etc/sysconfig.

There are several other syslog. They are located at

/etc/webmin/syslog
/etc/logrotate.d/syslog
/etc/sysconfig/syslog
/etc/rc.d/init.d/syslog
/var/lock/subsys/syslog
/var/opt/mkcdrec/busybox/include/config/config/feature/ipc/syslog
/usr/libexec/webmin/mscstyle3/syslog
/usr/libexec/webmin/caldera/syslog
/usr/libexec/webmin/syslog
/usr/libexec/webmin/blue-theme/syslog

I don't know which one is the correct one. None of the files looks like log.

Syslog should be /var/log/messages (see /etc/syslog.conf). Dmesg shows nothing overly interesting. If you're not into log interpreting you could run logwatch on syslog. With "--archives --range All" it will search logrotated ones as well and pick out kernel errors real easy.

Hi,

Attached is the message log file retrieve from /var/log.

Just now, I unplugged the external drive and plug it in again. Normally, it should auto mount. But nothing happened when I try just now.

Is the external hard disk gone? Is there anyway I can still retrieve the data? There are important data inside. Please help...

Allantch,

Can you see power lights active? I would try the drive on another machine to double check activity and that you do not have a USB bus failure. If you determine the drive will not spin-up, choices are limited. The 'electronics' in the enclosure may be defective and/or the drive is 'shot'. As a last resort, you can always try to remove the drive from the enclosure and hook it directly to an internal connection in your computer. I've read of some people putting a drive sealed in a baggie into the freezer for an hour and then trying to boot. Long shot but if all else fails....


Best regards,

marvinudy

Hi,

Thanks for the suggestion. I did unplug the hard drive from the server and plug it into another machine before. However, the other machine is running on WindowsXP instead of Linux. The drive did not appear in the WindowsXP machine.

Then I plug it into the Linux box again. No auto-mount. Nothing happened. The light is on though and there are sounds coming out of the hard disks. So I assume the drive is still spinning.

Just for your information, the hard disk is not powered by the USB itself. It is powered by another AC adapter. So when I turned on the power, the lights are active.

I will try to plug it to a different Linux box. Thanks.

Regards,
Allan