I had some problems, which I managed to solve.

Now I am left with a /dev/dsk/c0d0s0 that cant be mounted to / RW by
the system-filesystem-usr service.

What I have to do is

chmod +w /dev/dsk/c0d0s0
svcadm clear svc:/system/filesystem/usr

Then all goes well.  fsck on this device comes back fine.  I have one
big / to fit everything.

Next reboot same problem though.  Any ideas how I can make such a
change permanent?

"Sean" <swnshp@gmail.com> writes:

>I had some problems, which I managed to solve.

>Now I am left with a /dev/dsk/c0d0s0 that cant be mounted to / RW by
>the system-filesystem-usr service.

>What I have to do is

>chmod +w /dev/dsk/c0d0s0
>svcadm clear svc:/system/filesystem/usr

The chmod cannot have any effect.  So this looks like an issue
someplace else; is "/" listed as "mount at boot"? (It shouldn't, for
arcane reasons)

>Next reboot same problem though.  Any ideas how I can make such a
>change permanent?

What's really in the log files under /var/svc/log?

Casper
-- 
Expressed in this posting are my opinions.  They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.

On 2005-12-11 19:32, Sean wrote:
> I had some problems, which I managed to solve.
> 
> Now I am left with a /dev/dsk/c0d0s0 that cant be mounted to / RW by
> the system-filesystem-usr service.
> 
> What I have to do is
> 
> chmod +w /dev/dsk/c0d0s0

echo "And now everyone can whipe your disk" > /dev/dsk/c0d0s0

/birre 

Birre wrote:
> > chmod +w /dev/dsk/c0d0s0
>
> echo "And now everyone can whipe your disk" > /dev/dsk/c0d0s0
>
> /birre

I am setting it up, and am doing that while I investigate the cause.

I am the only user.  If I do whipe my disk, I shall be sure to tell
myself off.

Casper H.S. Dik wrote:
> >chmod +w /dev/dsk/c0d0s0
> >svcadm clear svc:/system/filesystem/usr
>
> The chmod cannot have any effect.  So this looks like an issue
> someplace else; is "/" listed as "mount at boot"? (It shouldn't, for
> arcane reasons)

That's the only way I can progress.  If I do the svcadm before doing
that, it will
fail again.

> What's really in the log files under /var/svc/log?

#cat /etc/svc/volatile/system-filesystem-usr:default.log

[Dec 12 18:13:21 Enabled. ]
[Dec 12 18:13:24 Executing start method ("/lib/svc/method/fs-usr") ]
ERROR: /sbin/mount -m -F ufs -o remount,rw / failed, err=32
mount: /dev/dsk/c0d0s0 write-protected
[Dec 12 18:13:27 Method "start" exited with status 95 ]

I had problems after installing some drivers, and I got around that by
restoring a previous /etc/driver_aliases from failsafe, it seems to be
believing that my IDE HDD was an audio device (it was saying something
like: pci-ide@1f,1 is audioi8101 ). Since i restored the file (from a
backup I had made before installing drivers, I am presented with the
above, left in maintenance mode until I do as it says.

Solaris Express snv24, on Intel X86 architecture.  This is a laptop if
you care to know. HDD is IDE not SATA or anything.  Not in RAID.

"Sean" <swnshp@gmail.com> writes:

>[Dec 12 18:13:21 Enabled. ]
>[Dec 12 18:13:24 Executing start method ("/lib/svc/method/fs-usr") ]
>ERROR: /sbin/mount -m -F ufs -o remount,rw / failed, err=32
>mount: /dev/dsk/c0d0s0 write-protected
>[Dec 12 18:13:27 Method "start" exited with status 95 ]

The error message has nothing to do with permission on the device;
it's complaining about the physical media being supposedly write
protected.

>I had problems after installing some drivers, and I got around that by
>restoring a previous /etc/driver_aliases from failsafe, it seems to be
>believing that my IDE HDD was an audio device (it was saying something
>like: pci-ide@1f,1 is audioi8101 ). Since i restored the file (from a
>backup I had made before installing drivers, I am presented with the
>above, left in maintenance mode until I do as it says.

Hm.  Did you add any driver aliases with "update_drv"?

>Solaris Express snv24, on Intel X86 architecture.  This is a laptop if
>you care to know. HDD is IDE not SATA or anything.  Not in RAID.

What happens when you boot from the failsafe?

What permissions does the device have before the chmod and
which after?

Casper

Casper H.S. Dik wrote:
> "Sean" <swnshp@gmail.com> writes:
>
> >[Dec 12 18:13:21 Enabled. ]
> >[Dec 12 18:13:24 Executing start method ("/lib/svc/method/fs-usr") ]
> >ERROR: /sbin/mount -m -F ufs -o remount,rw / failed, err=32
> >mount: /dev/dsk/c0d0s0 write-protected
> >[Dec 12 18:13:27 Method "start" exited with status 95 ]
>
> The error message has nothing to do with permission on the device;
> it's complaining about the physical media being supposedly write
> protected.
>
> Hm.  Did you add any driver aliases with "update_drv"?

I presume the installation of the drivers did that.  Which is why i
restored it from an old version.  I didnt do anything else after that.

>
> >Solaris Express snv24, on Intel X86 architecture.  This is a laptop if
> >you care to know. HDD is IDE not SATA or anything.  Not in RAID.
>
> What happens when you boot from the failsafe?

It works fine.

> What permissions does the device have before the chmod and
> which after?

/dev/dsk/c0d0s0 -> /devices/pci@0,0/pci-ide@1f,1/ide@0/cmdk@0,0:a

/dev/dsk/c0d0s0 is lrwxrwxrwx before and after
/devices/pci@0,0/pci-ide@1f,1/ide@0/cmdk@0,0:a is brw-rw-rw before and
after.

i.e. the same before and after!

If you want to know another weird thing.  My mouse never worked, unless
I cat the device first!  Don't ask me why.

But if I

#cat /devices/isa/i8042@1,60/mouse@1:l      Then ctrl-c, the mouse
works.

"Sean" <swnshp@gmail.com> writes:

>If you want to know another weird thing.  My mouse never worked, unless
>I cat the device first!  Don't ask me why.

>But if I

>#cat /devices/isa/i8042@1,60/mouse@1:l      Then ctrl-c, the mouse
>works.

Something is seriously screwy.

Does the "fix" you do work without first attempting the chmod?

What does your vfstab look like?

Casper
-- 
Expressed in this posting are my opinions.  They are in no way related
to opinions held by my employer, Sun Microsystems.
Statements on Sun products included here are not gospel and may
be fiction rather than truth.

Casper H.S. Dik wrote:
> Something is seriously screwy.

Yes.  That is true, the mouse thing puzzles me.  The issue is that the
mouse works without doing that cat on the device if ACPI is disabled.
If i pass acpi-enum=off, or acpi-user-options=2 or acpi-user-options=8
to the kernel in GRUB, then my mouse only works after doing the cat on
the device.

> Does the "fix" you do work without first attempting the chmod?

No.  I have to chmod a+w the device to get it to mount RW.

If i just try and do the svcadm clear it will just come out with the
same error I posted previously.  I must chmod it first.

> What does your vfstab look like?

# cat /etc/vfstab
#device         device          mount           FS      fsck    mount
mount
#to mount       to fsck         point           type    pass    at boot
options
#
fd      -       /dev/fd fd      -       no      -
/proc   -       /proc   proc    -       no      -
/dev/dsk/c0d0s1 -       -       swap    -       no      -
/dev/dsk/c0d0s0 /dev/rdsk/c0d0s0        /       ufs     1       no
-
/dev/dsk/c0d0s7 /dev/rdsk/c0d0s7        /export/home    ufs     2
yes     -
/devices        -       /devices        devfs   -       no      -
ctfs    -       /system/contract        ctfs    -       no      -
objfs   -       /system/object  objfs   -       no      -
swap    -       /tmp    tmpfs   -       yes     -