[SOLVED] Grub Load Read Errors; Is the drive salvageable?
Linux - HardwareThis forum is for Hardware issues.
Having trouble installing a piece of hardware? Want to know if that peripheral is compatible with Linux?
Notices
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
If you have any problems with the registration process or your account login, please contact us. If you need to reset your password, click here.
Having a problem logging in? Please visit this page to clear all LQ-related cookies.
Get a virtual cloud desktop with the Linux distro that you want in less than five minutes with Shells! With over 10 pre-installed distros to choose from, the worry-free installation life is here! Whether you are a digital nomad or just looking for flexibility, Shells can put your Linux machine on the device that you want to use.
Exclusive for LQ members, get up to 45% off per month. Click here for more info.
I have a single drive in my system, a SeaGate 76GB SCSI drive.
My machine locked up recently with the following message on
all xterms:
journal commit I/O error.
This machine has a separate /boot partition _however_ that boot
partition is ext2, not ext3 (and therefore has no journalling).
On rebooting (multiple times) the machine would fail to reboot
and I would see one of the two following messages:
Grub Loading Stage2 Read Error
or
Grub Read Error
"smartctl -H /dev/sda" yielded:
SMART Heath Status OK
On booting of a CD in "linux rescue" mode, all disk partitions
mounted except /boot.
I executed "fdisk /dev/sda1" sda1 being the /boot partition
and though I executed no commands to alter the partition,
I exited with "w" to save changes as opposed to "q" which
would exit without saving changes (even though I executed
no "partition altering" commands).
Now, however, a "linux rescue" boot finds and mounts all
partitions including the /boot partition.
---------
I tried
grub
grub> root (hd0,0)
grub> setup (hd0)
I tried:
grub-install --root-directory=/ /dev/sda1
and the machine would still not boot but again get hung
up on one of the following messages:
Grub Loading Stage1.5 Read Error
or
Grub Read Error
I deleted the Stage1.5 files, reran grub again, and sure,
the Stage1.5 error message went away but the machine would
now get hung up on "Stage2 error".
I finally INSTALLED a new OS, CentOS 5.5, the install completed
but on booting, I still get "Grub Stage2" errors.
On this re-install, I left all partitions sizes the same and
so /boot was the same size, same location.
---------
And so, I can't boot off of this disk. I normally would suspect
that the disk has to be discarded _expect_ that in rescue mode
I can mount all partitions and have access to all files.
What more can I do? Is there some way to repair the MBR
that I have not done?
---------
One more piece of information. In /var/log/messages, for the last
month, I was seeing the following information regularly at least
once every two days:
> Dec 21 16:04:33 localhost kernel: >>>>>>>>>>>>>>>>>> Dump Card State Begins <<<<<<<<<<<<<<<<<
> Dec 21 16:04:33 localhost kernel: scsi0: Dumping Card State while idle, at SEQADDR 0x18
> Dec 21 16:04:33 localhost kernel: Card was paused
> Dec 21 16:04:33 localhost kernel: ACCUM = 0x3, SINDEX = 0x48, DINDEX = 0xe4, ARG_2 = 0x0
> Dec 21 16:04:33 localhost kernel: HCNT = 0x0 SCBPTR = 0x2
> Dec 21 16:04:33 localhost kernel: SCSIPHASE[0x0] SCSISIGI[0x4] ERROR[0x0] SCSIBUSL[0x0]
> Dec 21 16:04:33 localhost kernel: LASTPHASE[0x1] SCSISEQ[0x1a] SBLKCTL[0xa] SCSIRATE[0x0]
> Dec 21 16:04:33 localhost kernel: SEQCTL[0x10] SEQ_FLAGS[0xc0] SSTAT0[0x0] SSTAT1[0x0]
> Dec 21 16:04:33 localhost kernel: SSTAT2[0x0] SSTAT3[0x0] SIMODE0[0x8] SIMODE1[0xa4]
> Dec 21 16:04:33 localhost kernel: SXFRCTL0[0x80] DFCNTRL[0x0] DFSTATUS[0x89]
> Dec 21 16:04:33 localhost kernel: STACK: 0xe2 0x164 0x179 0x17
> ...
See the attached file for the complete messages.
---------
Thank you to those who stayed with this post!
Last edited by chicagocoyote; 12-23-2010 at 11:55 AM.
I haven't dealt with SCSI disks since Mac LCIII days, but with IDE and Sata disks I would only trust the manufacturer's test utility to decide on the hdd health. The SMART OK message hints that it's fine. Could it be the SCSI controller and not the drive? The test utilities can be downloaded from the manufacturers, but the Ultimate Boot CD is a free download that has them all, and more.
1. testdisk, and see if it will find sda1 or whatever /boot is. There is a static testdisk binary for download Mebbe you can stick it on a usb disk or something http://www.cgsecurity.org/wiki/TestDisk_Download
2. e2fsck -fcvv /dev/<whatever>
3. chroot to your own system , mount /grub and run (from memory)
Grub install (hd0)
You can get going from your install cd by something like
mount /dev/<root> partition -o ro /
/sbin/init 3
This last will probably complain and spit at you a bit, but _just_might_ stand your system up. It overmounts on the cdrom's mounts, and they become invisible. It will say things at any rate, and might do the trick.
However, I feel like an idiot because the reason the system
would not boot off the SCSI disk is because I had a USB drive
stuck in the machine. I had never had a BIOS that could boot
off of a USB drive before and did not think to check this.
Now that the USB stick is removed, the machine boots up just fine.
There are still those warning signs that the disk may be failing
that I posted earlier though.
Thanks to those who responded as well as those who read through
my post.
We're all allowed one or two of those occasionally. It keeps us humble
I would take the "This disk is dying" type shrieking from a computer very seriously. It's usually accurate. Back up anything you want to have around on your next install, e.g. data, photos, conceivably music of films. That may only be /home/ but you know yourself. I would also back up /etc because you have the old settings you were using, which is very handy in cases of amnesia.
LinuxQuestions.org is looking for people interested in writing
Editorials, Articles, Reviews, and more. If you'd like to contribute
content, let us know.