>Number:         33151
>Category:       port-xen
>Synopsis:       NetBSD 3 Xen domain0 kernel crashes when attempting to reboot with root mounted on RAID
>Confidential:   no
>Severity:       serious
>Priority:       high
>Responsible:    port-xen-maintainer
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Sun Mar 26 11:50:00 +0000 2006
>Originator:     Teltariat
>Release:        NetBSD 3.0
>Organization:
Err....none right now.
>Environment:
NetBSD bastion 3.0 NetBSD 3.0 (XEN0) #0: Mon Dec 19 01:27:58 UTC 2005  builds@works.netbsd.org:/home/builds/ab/netbsd-3-0-RELEASE/i386/200512182024Z-obj/home/builds/ab/netbsd-3-0-RELEASE/src/sys/arch/i386/compile/XEN0 i386
>Description:
I have managed to get NetBSD's Xen 2.0 dom0 kernel running with root on 
RAID.

The NetBSD Xen kernel is able to successfully boot and load up the NetBSD
system from the autoconfigured (and set as root) RAID array [raid0].

The problem comes when it is time to reboot.

Issuing a simple 'reboot' causing the system to begin unmounting the 
mounted filesystems, however almost immediately after it starts doing
that, the kernel crashes HARD.  I don't understand ddb that well and
I am a complete and utter NetBSD newb (actually I am a newb to
kernel internals for all *nix systems. :-/ )

Again, I don't know much about this, but the crash seems to be occurring
during a 'wdc_exec_command'

The first thing ddb mentions is:
panic: wdc_exec_command: polled command not done
Stopped in pid 105.1 (reboot) at netbsd: cpu-Debugger+0x4

I did not finish copying all of that down.
>How-To-Repeat:
To repeat this problem, you have to set up root on a RAID 1 array, and
run the NetBSD Xen dom0 kernel on it.

I did this as the NetBSD Xen HowTo advised: by using GRUB to boot the
Xen hypervisor (xen.gz) and init the NetBSD XEN0 kernel using the GRUB
module command.

You will run into this crash the instant you try to shutdown or reboot
your machine.

It will crash just as its going through a "wd_shutdown".
>Fix: