rp2405 can't boot

[Trias_1]

dear all

trouble i have very complicated,because of my study case.
in my machine there are 2 disk (c3t15d0-->pri,c1t15d0), both them in one vg (vg00).
then my case was to destroy c1t1d0 configuration by entering command #cp /stand/vmunix /dev/vg00/c1t15d0
then i reboot.
suddenly come error like this
"disk(0/0/2/1.15)/stand/vmunix: cannot open, or not executable"

need help so much.
thanks

[Mridul Shrivastava]

what is the output of hpux ll ?
Have you tried booting from alternate path.. I mean from 0/0/2/3.15

If yes then try booting in lvm maintenance mode, you can boot in that using hpux -lm

What is the role of c1t1d0 here ?? since it is not a primary nor alternate path for booting ??

[Torsten.]

Hi,

what is this: (???)

#cp /stand/vmunix /dev/vg00/c1t15d0

Your kernel is now in the directory

/dev/vg00 named c1t15d0

Or did you use "mv"???

Is the disk mirrored?

[Trias_1]

what is the output of hpux ll ?
Have you tried booting from alternate path.. I mean from 0/0/2/3.15
>> still disk(0/0/2/1.15)/stand/vmunix: cannot open, or not executable"
there's no alt path, since i'm not mirroring.

If yes then try booting in lvm maintenance mode, you can boot in that using hpux -lm
>> show up
REASON FOR ALERT
SOURCE: 1 = processor
SOURCE DETAIL: 1 = processor general SOURCE ID: 0
PROBLEM DETAIL: 0 = no problem detail

LEDs: RUN ATTENTION FAULT REMOTE POWER
ON FLASH FLASH ON ON
LED State: System Running. Unexpected Reboot. Non-critical Error Detected.
Check Chassis and Console Logs for error messages.

0xF8E000301100E000 00000000 0000E000 - type 31 = legacy PA HEX chassis-code
0x58E008301100E000 00006B05 0808173A - type 11 = Timestamp 06/08/2007 08:23:58
A: ack read of this entry - X: Disable all future alert messages

What is the role of c1t1d0 here ?? since it is not a primary nor alternate path for booting ??
>> c1t15d0 just additional disk.

any suggestion?

[Duncan Edmonstone · Accepted solution]

Endras,

I can't beleive you eneterd this command:

cp /stand/vmunix /dev/vg00/c1t15d0

Cos, that would have just created a copy of /stand/vmunix in the directory /dev/vg00 with the name c1t15d0

If however you actually ran *this*

cp /stand/vmunix /dev/vg00/lvol1

Then you're in trouble, cos you overwrote your boot partition on *both* sides of the mirror - time to get out those recovery tapes!

If you actually ran this:

cp /stand/vmunix /dev/dsk/c1t15d0

Then you toasted one of your boot disks, but the other *should* be OK. Of course it won't boot off it unless the alt path was set correctly.

From the BCH prompt enter the command:

pa

to show what boot paths are stored in NVRAM

and then run

sea ipl

and see if you can locate a bootable disk...

HTH

Duncan

[Trias_1]

If you actually ran this:

cp /stand/vmunix /dev/dsk/c1t15d0
>> yes this is what i typed.
i done this to practice vgcfgrestore command.
ut it turn out different.
i typed "sea ipl", it shows 1 path(0/0/2/1.15)
but when i boot to it, again "disk(0/0/2/1.15)/stand/vmunix: cannot open, or not executable" shows up..

should i run recovery console?

[Torsten.]

OK,

cp /stand/vmunix /dev/dsk/c1t15d0

destroyed all the structures on this disk - forget it, you cannot recover.

Looks like the other disk is not proper mirrored.

Do you have an ignite tape?

Now is a good time to use it.

[Duncan Edmonstone]

So c3t15d0 is *not* a mirror of c1t15d0? Its just another disk in vg00 yes?

I hope you have an Ignite backup of some sort!

Doing waht you did will have overwritten the LIF area containing the initial system loader etc. and then gone on to overwrite the start of the /stand filesystem as well!

You can try booting off CD and running a recovery console - you should be able to get your other filesystems up, specifically the root / filesystem so you can get at /etc/lvmconf.

That will let you do a vgcfgrestore, but that will only recreate structures, you'll also need to rebuild your bootlif (I think there's an option to do this in the recovery menus) and then probably try and get your /stand filesystem back...

All in all a fine mess!

HTH

Duncan

[Torsten.]

I hope this is a test system.
Now it is time to do a image restore or new install.

If you want to test disk replacement, mirror your disk next time.

[freddy_21]

please try boot with vmunix.prev

Main Menu: Enter command or menu > bo pri isl
Interact with IPL (Y, N, or Cancel)?> y

Booting...
Boot IO Dependent Code (IODC) revision 1


HARD Booted.

ISL Revision A.00.43 Apr 12, 2000

ISL> hpux /stand/vmunix.prev


thanks
freddy

[Sandy Chen]

Hi,

vmunix.prev not helping in this case. you're destroying the lvm structure. So the only step you can take is restore from ignite backup or reinstall the server.

If this the server I know, then better to reinstall :D

Regards,
Sandy

[Trias_1]

dear all
thanks for all the replies.
i really appreciate it.
i'm fully aware that this is the test machine, so i have the guts to try the misconfiguration which led me to complete chaos.
but i think there's a first time for everyone.
now to solve this once for all.
as u all know i have two disks 1.c3t15d0(pri/boot) and 2.c1t15d0
If i reinstall OS on different disk(c1t15d0) not the primary(c3t15d0)?will i loss all data in c3t15d0? i mean how to keep data in c3t15d0?
that's the primary goal for now.
thanks all

[Torsten.]

I guess you data is already lost.

The disk layout is stored at the beginnig of the disk, there is the LVM related stuff stored. Depending on the size of your kernel, you have overwritten the boot zone, the LVM layout and parts of your first LVOLs.

Now the disk is not a LVM disk (information iabout this is lost), so you cannot mount any LVOL and you cannot read the disk.
Consider the data as lost.