Subject: Re: root on raid0 problems
To: Simon Burge <simonb@NetBSD.ORG>
From: Manuel Bouyer <bouyer@antioche.lip6.fr>
List: current-users
Date: 04/07/2000 12:49:23
On Fri, Apr 07, 2000 at 06:10:10AM +1000, Simon Burge wrote:
> Try
> 
> 	options DEBUG_HALT_BUSY
> 
> and you'll get a dump of busy buffers (and drop onto DDB on very recent
> -current).  I've seen this a lot where an mfs is used on /tmp.  Someone
> (sorry, forgot who) suggested adding
> 
> 	umount -f /tmp
> 	echo 'Unmounting all filesystems...'; umount -a

I don't have any MFS. Here's the debug output:
syncing disks... 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 giving up
Printing vnodes for busy buffers
tag 1 type VBLK, usecount 693, writecount 0, refcount 24,
        tag VT_UFS, ino 389, on dev 16, 0 flags 0x0, effnlink 1, nlink 1
        mode 060640, owner 0, group 5, size 0 not locked
tag 1 type VREG, usecount 0, writecount 0, refcount 1,
        tag VT_UFS, ino 19609, on dev 16, 0 flags 0x0, effnlink 1, nlink 1
        mode 0100644, owner 0, group 0, size 4284 not locked
tag 1 type VREG, usecount 0, writecount 0, refcount 1,
        tag VT_UFS, ino 19647, on dev 16, 0 flags 0x0, effnlink 1, nlink 1
        mode 0100664, owner 0, group 45, size 72 not locked
halted.

I'm not sure how to interpret 'dev 16'.
On '/' I have:
389    0 brw-r-----    1 root     operator  16,    0 Feb  9 00:26 /dev/raid0a
19609 10 -rw-r--r--    1 root     wheel         4284 Apr  7 12:33 /var/log/wtmp
19647  2 -rw-rw-r--    1 root     utmp            72 Apr  7 12:33 /var/run/utmp

As the modes matches I guess it's this one.

--
Manuel Bouyer, LIP6, Universite Paris VI.           Manuel.Bouyer@lip6.fr
--