Hi ,

Can anyone shed some light on why the following is happening and a possible
solution

My two main questions are why I am getting invalid wwn numbers ?
and how do I resolve

Jul 20 16:54:34 svc.startd[7]: svc:/network/rpc/bind:default: Method
"/lib/svc/method/rpc-bind start" failed with exit status 1.


The history behind this is I have just restore a ufsdump of root onto this
server from another machine and have got it to boot- well partly.


=========================
{6} ok boot -r
Resetting...
Software Power ON
Clock board TOD does not match TOD on any IO board.
screen not found.
Can't open input device.
Keyboard not present. Using ttya for input and output.
Clock board TOD does not match TOD on any IO board.

5-slot Sun Enterprise E3500, No Keyboard
OpenBoot 3.2.29, 512 MB memory installed, Serial #13221782.
Copyright 2001 Sun Microsystems, Inc. All rights reserved
Ethernet address 8:0:20:c9:bf:96, Host ID: 80c9bf96.


Rebooting with command: boot -r
Invalid wwn number 21000004 cf98b281
Invalid wwn number 21000020 37a53139
Invalid wwn number 21000020 37a52f9a
Boot device: /sbus@2,0/SUNW,socal@d,10000/sf@0,0/ssd@0,0 File and args: -r
SunOS Release 5.10 Version Generic_118822-25 64-bit
Copyright 1983-2005 Sun Microsystems, Inc. All rights reserved.
Use is subject to license terms.
Failed to plumb IPv4 interface(s): ge0
Configuring devices.
Hostname: sol10
checking ufs filesystems
/dev/rdsk/c0t0d0s7: is logging.
Jul 20 16:54:34 svc.startd[7]: svc:/network/rpc/bind:default: Method
"/lib/svc/method/rpc-bind start" failed with exit status 1.
Jul 20 16:54:35 svc.startd[7]: svc:/network/rpc/bind:default: Method
"/lib/svc/method/rpc-bind start" failed with exit status 1.
Jul 20 16:54:35 svc.startd[7]: svc:/network/rpc/bind:default: Method
"/lib/svc/method/rpc-bind start" failed with exit status 1.
[ network/rpc/bind:default failed (see 'svcs -x' for details) ]
Jul 20 16:54:36 svc.startd[7]: svc:/platform/sun4u/sf880drd:default: Method
"/lib/svc/method/sf880dr" failed with exit status 96.
[ platform/sun4u/sf880drd:default misconfigured (see 'svcs -x' for
details) ]
====================================

Thanks

Paul

p@p.com wrote:
> My two main questions are why I am getting invalid wwn numbers ?
> and how do I resolve
[snip]
> The history behind this is I have just restore a ufsdump of root onto this
> server from another machine and have got it to boot- well partly.
[snip]
> Rebooting with command: boot -r
> Invalid wwn number 21000004 cf98b281
> Invalid wwn number 21000020 37a53139
> Invalid wwn number 21000020 37a52f9a

No bueno. The wwn numbers are globally unique. You've overwritten the
entries for these with the ufsdump it seems. So you're using the wrong
wwn's now and won't be able to find disks. There are necessary entires
for them in /dev, /devices and etc/path_to_inst.

Ideally you should restore these from backups. If you have none, to the
best of my knowledge the following can straighten this out:

1. boot from cdrom (boot cdrom -s)
2. mount your root files system to /a (mount /dev/dsk/<root partition>
/a)
3. remove /dev and /devices entries (rm -r /a/devices/*; rm -r
/a/dev/*)
4. copy over current /devices entries (cd /devices; find . | cpio -pmd
/a/devices)
5. copy over current /dev entries (cd /dev; find . | cpio -pmd /a/dev)
6. remove path_to_inst (rm /a/etc/path_to_inst)
7. reboot with -a to rebuild path_to_inst (reboot -- -a)
(when asked if you want to rebuild path_to_inst enter 'y')

~ Shay

Thanks for your reply,

I used the following, I can't see much difference and I still get the WWN
errors?

# mount -F ufs /dev/dsk/<root disk device name> /mnt
# rm -rf /mnt /dev
# rm -rf /mnt /devices
# devfsadm -r /mnt -p /mnt/etc/path_to_inst
# cp -R /devices /mnt/devices <Solaris 10 only>

# cp -R /dev /mnt/dev <Solaris 10 only>
# init 0
ok boot -r

Paul


"Shay" <shayylynn@aol.com> wrote in message
news:1153426412.404078.96920@m79g2000cwm.googlegro ups.com...
>
> p@p.com wrote:
>> My two main questions are why I am getting invalid wwn numbers ?
>> and how do I resolve
> [snip]
>> The history behind this is I have just restore a ufsdump of root onto
>> this
>> server from another machine and have got it to boot- well partly.
> [snip]
>> Rebooting with command: boot -r
>> Invalid wwn number 21000004 cf98b281
>> Invalid wwn number 21000020 37a53139
>> Invalid wwn number 21000020 37a52f9a
>
> No bueno. The wwn numbers are globally unique. You've overwritten the
> entries for these with the ufsdump it seems. So you're using the wrong
> wwn's now and won't be able to find disks. There are necessary entires
> for them in /dev, /devices and etc/path_to_inst.
>
> Ideally you should restore these from backups. If you have none, to the
> best of my knowledge the following can straighten this out:
>
> 1. boot from cdrom (boot cdrom -s)
> 2. mount your root files system to /a (mount /dev/dsk/<root partition>
> /a)
> 3. remove /dev and /devices entries (rm -r /a/devices/*; rm -r
> /a/dev/*)
> 4. copy over current /devices entries (cd /devices; find . | cpio -pmd
> /a/devices)
> 5. copy over current /dev entries (cd /dev; find . | cpio -pmd /a/dev)
> 6. remove path_to_inst (rm /a/etc/path_to_inst)
> 7. reboot with -a to rebuild path_to_inst (reboot -- -a)
> (when asked if you want to rebuild path_to_inst enter 'y')
>
> ~ Shay
>

p@p.com wrote:

> I used the following, I can't see much difference and I still get the WWN
> errors?
>
> # mount -F ufs /dev/dsk/<root disk device name> /mnt
> # rm -rf /mnt /dev
> # rm -rf /mnt /devices
> # devfsadm -r /mnt -p /mnt/etc/path_to_inst
> # cp -R /devices /mnt/devices <Solaris 10 only>
>
> # cp -R /dev /mnt/dev <Solaris 10 only>
> # init 0
> ok boot -r

As far as I can tell, this might not works for drive replacement if the
drive uses a WWN. You might have to rebuild /etc/path_to_inst from
scratch. I'm not sure if devfsadm will be sufficient (people have both
sworn by this method and disavowed it as insufficient - apparently
mileage varies with this approach from release to release).

Try the same procedure, but remove /mnt/etc/path_to_inst entirely (in
lieue of the devfsadm run) and reboot with the "reboot -- -a". This
should give you the option to rebuild path_to_inst. Alternately you can
boot from cdrom and compare the files to see if there are any
discernable difference between the active file and the one on /mnt.

If that doesn't correct the issue, I'm not sure what is causing the
errors. At that point, if it didn't correct the issue, I'd probably
build the drive fresh and copy everything over But path_to_inst and the
/dev and /devices dirs.

~ Shay

Shay wrote:
> p@p.com wrote:
>
> > I used the following, I can't see much difference and I still get the WWN
> > errors?
> >
> > # mount -F ufs /dev/dsk/<root disk device name> /mnt
> > # rm -rf /mnt /dev
> > # rm -rf /mnt /devices
> > # devfsadm -r /mnt -p /mnt/etc/path_to_inst
> > # cp -R /devices /mnt/devices <Solaris 10 only>
> >
> > # cp -R /dev /mnt/dev <Solaris 10 only>
> > # init 0
> > ok boot -r
>
> As far as I can tell, this might not works for drive replacement if the
> drive uses a WWN. You might have to rebuild /etc/path_to_inst from
> scratch. I'm not sure if devfsadm will be sufficient (people have both
> sworn by this method and disavowed it as insufficient - apparently
> mileage varies with this approach from release to release).
>
> Try the same procedure, but remove /mnt/etc/path_to_inst entirely (in
> lieue of the devfsadm run) and reboot with the "reboot -- -a". This
> should give you the option to rebuild path_to_inst. Alternately you can
> boot from cdrom and compare the files to see if there are any
> discernable difference between the active file and the one on /mnt.
>
> If that doesn't correct the issue, I'm not sure what is causing the
> errors. At that point, if it didn't correct the issue, I'd probably
> build the drive fresh and copy everything over But path_to_inst and the
> /dev and /devices dirs.
>
> ~ Shay

Another possible solution would be to copy the old path_to_inst, blow
away the device entries
in path_to_inst, blow away /devices and /dev and boot -r