gmirror panic on RELENG_8

[Steve Price]

Here's a recap of my setup:

1. Express install from 8.2RC1 disc 1 and chose minimal distribution
2. da0 - 8GB swap, rest /
3. da[12] - left untouched
4. set root's password
3. reboot

The box comes up fine. I've changed the prompt from '#' to '[dev]#'
to make it a little easier to follow.

[dev]# uname -a
FreeBSD  8.2-RC1 FreeBSD 8.2-RC1 #0: Thu Dec 23 15:32:35 UTC 2010
root at almeida.cse.buffalo.edu:/usr/obj/usr/src/sys/GENERIC  i386
[dev]# cat /etc/fstab
# Device		Mountpoint	FStype	Options		Dump	Pass#
/dev/da0s1b		none		swap	sw		0	0
/dev/da0s1a		/		ufs	rw		1	1
/dev/acd0		/cdrom		cd9660	ro,noauto	0	0
[dev]# df -k
Filesystem  1024-blocks   Used     Avail Capacity  Mounted on
/dev/da0s1a   275615636 310682 253255704     0%    /
devfs                 1      1         0   100%    /dev
[dev] dmesg | grep '^da.*:'
da0: <FUJITSU MBA3300NC 5E03> Fixed Direct Access SCSI-3 device
da0: 320.000MB/s transfers (160.000MHz, offset 127, 16bit)
da0: Command Queueing enabled
da0: 286102MB (585937500 512 byte sectors: 255H 63S/T 36472C)
da1: <FUJITSU MBA3300NC 5E03> Fixed Direct Access SCSI-3 device
da1: 320.000MB/s transfers (160.000MHz, offset 127, 16bit)
da1: Command Queueing enabled
da1: 286102MB (585937500 512 byte sectors: 255H 63S/T 36472C)
da2: <FUJITSU MBA3300NC 5E03> Fixed Direct Access SCSI-3 device
da2: 320.000MB/s transfers (160.000MHz, offset 127, 16bit)
da2: Command Queueing enabled
da2: 286102MB (585937500 512 byte sectors: 255H 63S/T 36472C)
[dev]# cat last.sh
#!/bin/sh
devs='da0 da1 da2'
for dev in $devs; do
  sectors=`diskinfo $dev | cut -f4`
  size=`diskinfo $dev | cut -f2`
  last=$(($sectors - 1))
  dd if=/dev/$dev of=sector.$dev bs=$size count=1 skip=$last
done
[dev]# sh last.sh
1+0 records in
1+0 records out
512 bytes transferred in 0.000233 secs (2198039 bytes/sec)
1+0 records in
1+0 records out
512 bytes transferred in 0.017391 secs (29441 bytes/sec)
1+0 records in
1+0 records out
512 bytes transferred in 0.012634 secs (40525 bytes/sec)
[dev]# cat sector.da*
[dev]# ls -l sector.da*
-rw-r--r--  1 root  wheel  512 Jan 11 16:34 sector.da0
-rw-r--r--  1 root  wheel  512 Jan 11 16:34 sector.da1
-rw-r--r--  1 root  wheel  512 Jan 11 16:34 sector.da2
[dev]# echo 'geom_mirror_load="YES"' >> /boot/loader.conf
[dev]# cat /boot/loader.conf
geom_mirror_load="YES"
[dev]# kldstat
Id Refs Address    Size     Name
 1    1 0xc0400000 bd9960   kernel
[dev]# sysctl kern.geom.debugflags=17
kern.geom.debugflags: 0 -> 17
[dev]# gmirror label -vb round-robin gm0 /dev/da0
Metadata value stored on /dev/da0.
Done.
[dev]# sed -i.bak -e 's,da0,mirror/gm0,g' /etc/fstab
[dev]# cat /etc/fstab
# Device		Mountpoint	FStype	Options		Dump	Pass#
/dev/mirror/gm0s1b	none		swap	sw		0	0
/dev/mirror/gm0s1a	/		ufs	rw		1	1
/dev/acd0		/cdrom		cd9660	ro,noauto	0	0
[dev]# reboot

Box panics.  If I remove either da1 or da2 the box doesn't
panic.  I have 5 identical drives laying around to use for
da[12].  Pick 2, any 2, and the box panics if all 3 drives
are present on boot.  All 5 drives have been /dev/zero'd
with the following:

dd if=/dev/zero of=/dev/da1 bs=100m

Here's some extra info with only 2 drives present.

dev% gmirror list
Geom name: gm0
State: COMPLETE
Components: 1
Balance: round-robin
Slice: 4096
Flags: NONE
GenID: 0
SyncID: 1
ID: 1042125830
Providers:
1. Name: mirror/gm0
   Mediasize: 299999999488 (279G)
   Sectorsize: 512
   Mode: r2w2e5
Consumers:
1. Name: da0
   Mediasize: 300000000000 (279G)
   Sectorsize: 512
   Mode: r1w1e1
   State: ACTIVE
   Priority: 0
   Flags: NONE
   GenID: 0
   SyncID: 1
   ID: 1109418373

dev% gmirror status
      Name    Status  Components
mirror/gm0  COMPLETE  da0

I've tried every permutation I can think of.  Still
not able to get a coredump which would be immensely
helpful at this point.

Thanks.

-steve