ioinitrc error when booting

[Michal Golata]

Hi,

When I'm booting the server, I see following errors displayed:

/sbin/ioinitrc:
lvlnboot: Warning: couldn't query physical volume "/dev/dsk/c4t0d3":
The specified path does not correspond to physical volume attached to
this volume group
lvlnboot: Warning: couldn't query physical volume "/dev/dsk/c10t0d3":
The specified path does not correspond to physical volume attached to
this volume group
lvlnboot: Warning: couldn't query all of the physical volumes.

The devices above (c4t0d3 and c10t0d3) is in fact the same device which is LUN accessed via separate FC links. This LUN is used for a volume group which has nothing to do with boot process. The OS is started from internal disk where vg00 is created. Further, up to my knowlege no swap, dump, root nor boot volume is defined on affected volume group.

When the system is up, I cannot see any errors related with those devices. I run ioscan, diskinfo, vgdisplay and pvdisplay. All the command lists the devices and the volume group defined on this LUN is accessible.

Can anybody tell what is the reason for those errors and how to get rid of it?

Regards,

Michal

[Eric Antunes]

Hi Michal,

Fisrt of all, did you change something recently in the LVM configuration?

Please also post the output of:

ioscan -fnCdisk

Best Regards,

Eric Antunes

[Michal Golata]

Hi,

I reviewed old logs I have and I could see that the error started to appear when the system was unsuccessfully restored from the Ignite backup. I can't remember the details but somehow the system was restored back to work after this unsuccessful Ignite restore. From that time everything seems to work fine apart from this ioinitrc error. Recently some swap configuration was modified on the system but prior to and after the change none of swaps was defined on the affected volume group. But again, the error did not start to occur after the swap changes, it was there earlier. The output from ioscan is following:

Class I H/W Path Driver S/W State H/W Type Description
===========================================================================
disk 0 0/0/0/3/0.6.0 sdisk CLAIMED DEVICE HP 36.4GST336753LC
/dev/dsk/c0t6d0 /dev/rdsk/c0t6d0
disk 1 0/0/10/0/0.8.0.110.0.0.0 sdisk CLAIMED DEVICE HP A6189B
/dev/dsk/c4t0d0 /dev/rdsk/c4t0d0
disk 5 0/0/10/0/0.8.0.110.0.0.1 sdisk CLAIMED DEVICE HP A6189B
/dev/dsk/c4t0d1 /dev/rdsk/c4t0d1
disk 9 0/0/10/0/0.8.0.110.0.0.3 sdisk CLAIMED DEVICE HP A6189B
/dev/dsk/c4t0d3 /dev/rdsk/c4t0d3
disk 2 1/0/0/3/0.6.0 sdisk CLAIMED DEVICE HP 36.4GST336753LC
/dev/dsk/c6t6d0 /dev/rdsk/c6t6d0
disk 3 1/0/0/3/1.2.0 sdisk CLAIMED DEVICE HP DVD-ROM 305
/dev/dsk/c7t2d0 /dev/rdsk/c7t2d0
disk 4 1/0/2/0/0.8.0.108.0.0.0 sdisk CLAIMED DEVICE HP A6189B
/dev/dsk/c10t0d0 /dev/rdsk/c10t0d0
disk 6 1/0/2/0/0.8.0.108.0.0.1 sdisk CLAIMED DEVICE HP A6189B
/dev/dsk/c10t0d1 /dev/rdsk/c10t0d1
disk 10 1/0/2/0/0.8.0.108.0.0.3 sdisk CLAIMED DEVICE HP A6189B
/dev/dsk/c10t0d3 /dev/rdsk/c10t0d3


Please keep in mind the volume group on affected devices seems to work fine and none of vgdisplay, pvdisplay, diskinfo nor ioconfig return any errors. The problem occurs only during the boot.

Regards,

Michal

[Steven E. Protter · Accepted solution]

Shalom Michal,

A disk has died or is in the lvmtab configuration and doesn't belong there.

cd /etc
mv lvmtab lvmtab.sav
vgscan -a

This should rebuild the lvmtab file and eliminate the problem. You may however get errors and be forced to take other steps such as vgreduce -f on the disks that are reporting problems at boot time.

It is good you noticed this and the problem should be addressed.

SEP

[Michal Golata]

Hi,

Thanks for the suggestions. Before I do it, I'd like to ask one more thing (as the machine is a production box and I really don't want to make a mistake there...). Here is the output of /etc/lvmtab:

nfm2:/rhome/root# strings /etc/lvmtab
/dev/nfmdg
/dev/dsk/c4t0d3
/dev/dsk/c10t0d3
/dev/vg00
/dev/dsk/c0t6d0
/dev/vg01
/dev/dsk/c4t0d1
/dev/dsk/c10t0d1
/dev/vg02
/dev/dsk/c4t0d0
/dev/dsk/c10t0d0

And the part of vgdisplay for affected /dev/nfmdg volume group:

nfm2:/rhome/root# vgdisplay -v /dev/nfmdg

--- Physical volumes ---
PV Name /dev/dsk/c4t0d3
PV Name /dev/dsk/c10t0d3 Alternate Link
PV Status available
Total PE 26747
Free PE 4425
Autoswitch On


So one can see that the disks ioinitrc complains about are really used by the /dev/nfmdg volume group. But isn't it a problem that this /dev/nfmdg is the first entry in /etc/lvmtab? Do you think that this could be the root cause?

And one more question. Is it better to remove /etc/lvmtab and run vgscan -a from single user or I can easly do it from init 3? I really hope I don't break anything...

Regards,

Michal

[Eric Antunes]

Hi,

Do you have volume group nfmdg (vgdisplay -v nfmdg)?

Eric

[Eric Antunes]

Hi,

Of course you have a nfmdg VG.

This is happening because this nfmdg is the first of lvmtab. The first should be vg00...

Do as SEP says and check lvmtab again (it is safe ;-) )!

Cheers,

Eric

[Michal Golata]

Guys,

Thanks for suggestions. Renaming /etc/lvmtab and running vgscan -a helped. After the reboot no errors.

Regards,

Michal