LVLNBOOT

[Josee Bourget-Thuma]

System is HP-UX 11.00 on K580.
While booting I get the following error:

lvlnboot: Couldn't query all the physical volumes.

Funny enough, the system comes up just fine anyway.

Some please point me in the right direction.
I want to be sure I have the correct volume setup in case of recovery.

Thanks a lot!
Josee...

[Mark Mitchell]

Sounds like there may be either A, some disk failure and the mirror is being used or B, the lvm file references some groups that no longer exist on the server. Start by looking in the syslog.log file under /var/adm/syslog. Also take a look at your disks for any service lights.

[Josee Bourget-Thuma]

I'd checked the drive lights already. Lights are indicating everything is fine.

Only message possibly related that I see in syslog is:

ps2_readbyte_timeout: no byte after 500 uSec

I read this message to an HP engineer but he had nothing to say about this.

Anymore clues?
Thanks for your help!

Josee...

[Patrick Wallek]

Here are a couple of links from the ITRC TKB:

http://us-support.external.hp.com/cki/bin/doc.pl/sid=a87b15f41ab12d1edc/screen=ckiDisplayDocument?docId=200000039545423

http://us-support.external.hp.com/cki/bin/doc.pl/sid=a87b15f41ab12d1edc/screen=ckiDisplayDocument?docId=200000024614078

[Josee Bourget-Thuma]

ok, reviewed the suggested docs.
Which method is the prefered one?

vgexport?
or
modifying /sbin/ioinitrc?

I understand what the process does while doing the vgexport but don't understand what the change to ioinitrc implies... please ellaborate on that.

For the record, the ps2 error message previously mentioned is just a ps2 driver not being used on the system. Simply ignore or remove.

Thanks!
Josee...

[Joseph A Benaiah_1 · Accepted solution]

Josee,

I had a problem with a server at work, lvlnboot said that it could not query all the physical devices. The disks in question were part of vg00 which is mirrored. All disks in the other vgs are on an EMC disk array. The checks I did were as follows:

1. dd if=/dev/rdsk/c?t?d? of=/dev/null bs=64k on both internal disks. Both came back without any errors

2. I suspected that someone might have replaced one of the disks in the Hass without syncing them so I ran:

lifls -l /dev/rdsk/c?t?d?

This was successful.

3. strings /etc/lvmtab revealed that one of the disks was missing from vg00 so I did the following:

cp -p /etc/lvmtab /etc/lvmtab.sav
vgscan - to produce a new file which now had the correct information in it
lvlnboot -R - to establish any links.
lvlnboot -v - to check that the contents of the BDRA are correct.

Cheers,

Joseph.