boot error???

[Jon Mattatall]

Ok....
We had a generator test last week while I was away.
A newly built 11.00 box was powered off, and we started it back up today(turned key).
BOOM....
It will no longer boot. It comes up with

Rights for non-DOD U.S. Government Departments and Agencies are as set forth in FAR 52.227-19(c)(1,2).
/sbin/rc: sshd not found

INIT: Command is respawning too rapidly.
Will try again in 5 minutes.
Check for possible errors.
id:cons "/usr/sbin/getty console console # system console"

INIT: Command is respawning too rapidly.
Will try again in 5 minutes.
Check for possible errors.
id:ems4 "/etc/opt/resmon/lbin/p_client"
Resynchronized volume group /dev/vg00
Resynchronized volume group /dev/vg01
...and hangs.
This has been thrashed about in the forums, I know.
So, I boot to hpux -lm, figuring to check /sbin/rc.* or /sbin/init.d for problems.

I get:
/dev/vg00/lvol1: possible swap device (cannot determine)
fsck SUSPENDED BY USER.
/dev/vg00/lvol1: No such device or address
Unable to mount /stand - please check entries in /etc/fstab
and it boots to level "s".

/etc/fstab:
/dev/vg00/lvol3 / vxfs delaylog 0 1
/dev/vg00/lvol1 /stand hfs defaults 0 1
/dev/vg00/lvol4 /home vxfs delaylog 0 2
/dev/vg00/lvol5 /opt vxfs delaylog 0 2
/dev/vg00/lvol6 /tmp vxfs delaylog 0 2
/dev/vg00/lvol7 /usr vxfs delaylog 0 2
/dev/vg00/lvol8 /var vxfs delaylog 0 2

So I...
# fsck -F hfs /dev/rdsk/c0t6d0
fsck: /dev/rdsk/c0t6d0: possible swap device (cannot determine)

continue (y/n)? y
** /dev/rdsk/c0t6d0
BAD SUPER BLOCK: MAGIC NUMBER WRONG
USE -b OPTION TO FSCK TO SPECIFY LOCATION OF AN ALTERNATE
SUPER-BLOCK TO SUPPLY NEEDED INFORMATION; SEE fsck(1M).

reboot.....

/sbin/ioinitrc:
fsck: /dev/vg00/lvol1: possible swap device (cannot determine)
fsck SUSPENDED BY USER.
mount: file system table may be corrupt
Unable to mount /stand - please check entries in /etc/fstab

The initial problem was fine, I could just init 1,2,3 my way up and all seemed OK till a reboot.
This new thing has me baffled, and I cannot find anyone who will admit to powering the box off.

So...
If someone just twisted the key to off, would this account for the apparent corruption?
I know the first backup Superblock on a Solaris block is at 32, is this the same for HPUX?
Can I... fsck -F hfs -b 32 /dev/rdsk/cntndn with this?
Or is it back to the start? What am I missing?

About to abuse my cold medication,
J

[Sanjay_6]

Hi Jon,

Instead of trying to boot the system, i think you look for an alternative solution, like if you have a root mirror, use that, else get a space disk and install os on that, recover your data on the system first.

If there is nothing on this box, simply reinstall. Seems like a massive corruption of the root disk / filesystem.

Hope this helps.

Regds

[Santosh Nair_1 · Accepted solution]

The first error seems to be resulting from a badly written configuration/init script, i.e. check in /etc/rc.config.d or /sbin/init.d directories for improper syntax etc.

The second error was a result of you're fsck in the raw disk devices, i.e. /dev/rdsk/c0t6d0. The filesystems are on lvols, not the raw disks, i.e. you should have done:

fsck -F hfs /dev/vg00/lvol1.

Try booting into single user, i.e.:

hpux -is -lm

at the ISL prompt and fscking the proper device. Then if that work, try booting into multi-user and checking for errors in the startup scripts (as mentioned above).

If this fails, hopefully you have good backups because I think you're root disk is all messed up.

-Santosh

[Craig Rants]

Jon,
I would not hurt to try an alternate boot block fsck. I have done it on a Sun before like you said, but never on an HP. I don't think that it would hurt, what's the worst thing that could happen?

Good Luck,
C

[Roger Baptiste]

Hi,

I had similar errors during booting one of my systems. The problem turned
to be with errorneous changes
made in the fstab file.
It mounted /home as /opt,
/opt as /usr and so on,
since the lvol<->FS mapping
was screwed up in the fstab file.

Your error seems similar,
so i would suggest you to
have a close look at your
fstab file to see when it
was last changed. Also see
whether you have a old fstab
file (ll /etc/fstab*) around.


<<# fsck -F hfs /dev/rdsk/c0t6d0
fsck: /dev/rdsk/c0t6d0: possible swap device (cannot determine) >>

Nope, this is not solaris ;-) You need to do fsck on
LV's. So, that should be
fsck -F /dev/vgxx/lvolxx


<fsck: /dev/vg00/lvol1: possible swap device (cannot determine) >>

Since it keeps whining
about swap device, check
your boot settings:
#lvlnboot -v
(lvol2 should be swap device)


HTH
raj

[Sridhar Bhaskarla]

Hi Jon,

Your rc was not successful so you are getting the command responding too rapidly.

Your fstab entries tell that your logical volumes are of vxfs type except for /stand which is correctly an hfs.

When you do fsck, instead of specifying /dev/rdsk/c0t6d0 (which is wrong), just do an fsck on each lv . You need to specify the fs type using -F option

#fsck -F hfs /dev/vg00/lvol1
#fsck -F vxfs /dev/vg00/lvol3 and so on..

See if your fsck comes out clean. If so, reboot the system to get it up. Otherwise, try booting from the alternate path and re-establish the mirrors.

-Sri

[Jon Mattatall]

My apologies about the fsck, mass quantities of DayQuil and
SinuTab have rendered me nearly comatose....
Anyway, I've got the box back to the respawn error. I've got a sneaking suspicion that our openssh script is the culprit.
So, I'll start there and if it does'nt come around, well,
it's not production and I've got the install CD....

My stuffy, sneezy thanks,

Jon