Dom0 OOPS - causes restart

Dom0 OOPS - causes restart

[Niraj Tolia]

** Forwarding to the list to provide context to a later email
** now that the listserv seems to be back up     


---------- Forwarded message ----------
From: Niraj Tolia <ntolia@gmail.com>
Date: Mon, 12 Jul 2004 14:05:46 +0100
Subject: Dom0 OOPS - causes restart
To: "xen-devel@lists.sourceforge.net" <xen-devel@lists.sourceforge.net>

Got an OOPS in dom0 (included below) that caused the machine to
reboot. Did a bk pull this morning (-r 1.1069). This is what I did to
get the oops.

a) loaded iSCSI drivers in dom0 (for the Intel 1000 Pro T iSCSI initiator)
b) Created a domain that used one of the partitions from the iSCSI
drive that was mounted (giving the 'phy:sda1,hda1,rw' option in the
domain config file). The domain uses the xen0 kernel image
c) Tried to run Postmark within the newly created domain (not dom0).

FWIW, I can run Postmark and IOzone just fine in dom0.

I also ran into another problem and I don't know if its related. When
trying to create a new domain (same config as specified above), I got
an OOPS in dom1 (not 0) just after the point when the kernel says its
trying to DHCP for addresses. However, I wasn't quick enough to
capture it. Could not reproduce the problem after a reboot.

Niraj

ksymoops 2.4.9 on i686 2.4.22-1.2194.nptl.  Options used
     -v ./vmlinux-syms-2.4.26-xen0 (specified)
     -K (specified)
     -L (specified)
     -O (specified)
     -m ./System.map (specified)

Unable to handle kernel paging request at virtual address c4986ffe
c01c34e8
Oops: 0003
CPU:    0
EIP:    0819:[<c01c34e8>]    Not tainted
Using defaults from ksymoops -t elf32-i386 -a i386
EFLAGS: 00213297
eax: 0c0005df   ebx: c63ab920   ecx: 00000010   edx: c4986ffe
esi: c08eb8f0   edi: c5af9008   ebp: 00000000   esp: c0273b04
ds: 0821   es: 0821   ss: 0821
Process swapper (pid: 0, stackpage=c0273000)<1>
Stack: 00000004 c5af9008 c01c30ae c4986ffe c08eb8f0 00000010 00000000 c017af2b
       c63ab920 c0273b64 00000004 00000000 c5af9008 c0273b44 c01c3070 80000000
       c02daf40 c5af9008 c498700e c08eb900 c01c3070 c01c786e 00000002 00000004
Call Trace: [<c01c30ae>] [<c017af2b>] [<c01c3070>] [<c01c3070>] [<c01c786e>]
   [<c01c3070>] [<c017aba6>] [<c01c3070>] [<c017aeea>] [<c01c3070>]
[<c01c31b9>]
   [<c01c3070>] [<c017af2b>] [<c01c3160>] [<c01c72d6>] [<c01c3160>]
[<c017af2b>]
   [<c01c7240>] [<c01c3160>] [<c01c7452>] [<c01c7240>] [<c017aba6>]
[<c01c3160>]
   [<c017aeea>] [<c01c3160>] [<c01c3291>] [<c01c3160>] [<c01c3f2f>]
[<c017af2b>]
   [<c01c3e70>] [<c01c3e70>] [<c01c6d65>] [<c01c3e70>] [<c01c6c50>]
[<c017aba6>]
   [<c01c6c50>] [<c017af2b>] [<c01c6c50>] [<c01c6c50>] [<c01c3e70>]
[<c01c7147>]
   [<c01c6c50>] [<c017aba6>] [<c01c3e70>] [<c017aeea>] [<c01c3e70>]
[<c01c3e70>]
   [<c01c40f9>] [<c01c3e70>] [<c0172f69>] [<c0173156>] [<c01732b5>]
[<c000dd47>]
   [<c00ae29a>] [<c00b2a2f>] [<c00aca8b>] [<c00aa992>]
 <0>Kernel panic: Aiee, killing interrupt handler!
Warning (Oops_read): Code line not seen, dumping what data is available

>>EIP; c01c34e8 <__constant_memcpy+88/110>   <=====

>>eax; 0c0005df <__start___xen_guest+bffa145/bfff9b66>
>>esp; c0273b04 <init_task_union+1b04/2000>

Trace; c01c30ae <br_dev_queue_push_xmit+3e/f0>
Trace; c017af2b <nf_hook_slow+bb/190>
Trace; c01c3070 <br_dev_queue_push_xmit+0/f0>
Trace; c01c3070 <br_dev_queue_push_xmit+0/f0>
Trace; c01c786e <br_nf_post_routing+12e/150>
Trace; c01c3070 <br_dev_queue_push_xmit+0/f0>
Trace; c017aba6 <nf_iterate+76/b0>
Trace; c01c3070 <br_dev_queue_push_xmit+0/f0>
Trace; c017aeea <nf_hook_slow+7a/190>
Trace; c01c3070 <br_dev_queue_push_xmit+0/f0>
Trace; c01c31b9 <br_forward_finish+59/60>
Trace; c01c3070 <br_dev_queue_push_xmit+0/f0>
Trace; c017af2b <nf_hook_slow+bb/190>
Trace; c01c3160 <br_forward_finish+0/60>
Trace; c01c72d6 <br_nf_forward_finish+96/110>
Trace; c01c3160 <br_forward_finish+0/60>
Trace; c017af2b <nf_hook_slow+bb/190>
Trace; c01c7240 <br_nf_forward_finish+0/110>
Trace; c01c3160 <br_forward_finish+0/60>
Trace; c01c7452 <br_nf_forward_ip+102/120>
Trace; c01c7240 <br_nf_forward_finish+0/110>
Trace; c017aba6 <nf_iterate+76/b0>
Trace; c01c3160 <br_forward_finish+0/60>
Trace; c017aeea <nf_hook_slow+7a/190>
Trace; c01c3160 <br_forward_finish+0/60>
Trace; c01c3291 <__br_forward+61/70>
Trace; c01c3160 <br_forward_finish+0/60>
Trace; c01c3f2f <br_handle_frame_finish+bf/150>
Trace; c017af2b <nf_hook_slow+bb/190>
Trace; c01c3e70 <br_handle_frame_finish+0/150>
Trace; c01c3e70 <br_handle_frame_finish+0/150>
Trace; c01c6d65 <br_nf_pre_routing_finish+115/270>
Trace; c01c3e70 <br_handle_frame_finish+0/150>
Trace; c01c6c50 <br_nf_pre_routing_finish+0/270>
Trace; c017aba6 <nf_iterate+76/b0>
Trace; c01c6c50 <br_nf_pre_routing_finish+0/270>
Trace; c017af2b <nf_hook_slow+bb/190>
Trace; c01c6c50 <br_nf_pre_routing_finish+0/270>
Trace; c01c6c50 <br_nf_pre_routing_finish+0/270>
Trace; c01c3e70 <br_handle_frame_finish+0/150>
Trace; c01c7147 <br_nf_pre_routing+287/350>
Trace; c01c6c50 <br_nf_pre_routing_finish+0/270>
Trace; c017aba6 <nf_iterate+76/b0>
Trace; c01c3e70 <br_handle_frame_finish+0/150>
Trace; c017aeea <nf_hook_slow+7a/190>
Trace; c01c3e70 <br_handle_frame_finish+0/150>
Trace; c01c3e70 <br_handle_frame_finish+0/150>
Trace; c01c40f9 <br_handle_frame+139/1f0>
Trace; c01c3e70 <br_handle_frame_finish+0/150>
Trace; c0172f69 <netif_receive_skb+b9/210>
Trace; c0173156 <process_backlog+96/170>
Trace; c01732b5 <net_rx_action+85/160>
Trace; c000dd47 <do_softirq+d7/f0>
Trace; c00ae29a <do_IRQ+9a/a0>
Trace; c00b2a2f <evtchn_do_upcall+af/110>
Trace; c00aca8b <hypervisor_callback+33/49>
Trace; c00aa992 <cpu_idle+72/d0>

1 warning issued.  Results may not be reliable.


-------------------------------------------------------
This SF.Net email sponsored by Black Hat Briefings & Training.
Attend Black Hat Briefings & Training, Las Vegas July 24-29 - 
digital self defense, top technical experts, no vendor pitches, 
unmatched networking opportunities. Visit www.blackhat.com

Re: Dom0 OOPS - causes restart

[ron minnich]

anyone doing LILO and xen? I don't see any example .conf files any more. 
I'm going to bite the bullet here and cut my desktop over to Xen if 
possible. Should be amusing.

ron



-------------------------------------------------------
This SF.Net email sponsored by Black Hat Briefings & Training.
Attend Black Hat Briefings & Training, Las Vegas July 24-29 - 
digital self defense, top technical experts, no vendor pitches, 
unmatched networking opportunities. Visit www.blackhat.com

Re: Dom0 OOPS - causes restart

[Kip Macy]

You can't use lilo with Xen. As for using it for your desktop, I don't
anticipate any problems - even a laptop should work now with NGIO.



				-Kip

On Mon, 12 Jul 2004, ron minnich wrote:

> anyone doing LILO and xen? I don't see any example .conf files any more.
> I'm going to bite the bullet here and cut my desktop over to Xen if
> possible. Should be amusing.
>
> ron
>
>
>
> -------------------------------------------------------
> This SF.Net email sponsored by Black Hat Briefings & Training.
> Attend Black Hat Briefings & Training, Las Vegas July 24-29 -
> digital self defense, top technical experts, no vendor pitches,
> unmatched networking opportunities. Visit www.blackhat.com
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/xen-devel
>


-------------------------------------------------------
This SF.Net email sponsored by Black Hat Briefings & Training.
Attend Black Hat Briefings & Training, Las Vegas July 24-29 - 
digital self defense, top technical experts, no vendor pitches, 
unmatched networking opportunities. Visit www.blackhat.com

Re: Dom0 OOPS - causes restart

[Ian Pratt]

> anyone doing LILO and xen? I don't see any example .conf files any more. 

Xen's boot image conforms to the 'multiboot' standard, which
turns out not to be such a standard after all... In fact, Grub is
about the only implementation I know of (other than our patched
version of kexec)

I think multiboot is generally a good thing, and I can't think of
any downside to using Grub over Lilo (other than lilo's -R
option, for which standard Grub doesn't really have an
equivalent.)

> I'm going to bite the bullet here and cut my desktop over to Xen if 
> possible. Should be amusing.

Have fun ;-)  

I ran it on my laptop for a while, but the lack of APM/ACPI
support got to me eventually. In the new IO world it mightn't be
too hard to make ACPI work: all the nasty interpreter stuff would
live in Linux, and we'd just need the bare minimum in
Xen. Haven't looked too closely though -- there may be a gotcha.

Ian


-------------------------------------------------------
This SF.Net email sponsored by Black Hat Briefings & Training.
Attend Black Hat Briefings & Training, Las Vegas July 24-29 - 
digital self defense, top technical experts, no vendor pitches, 
unmatched networking opportunities. Visit www.blackhat.com

Re: Dom0 OOPS - causes restart

[Keir Fraser]

> > anyone doing LILO and xen? I don't see any example .conf files any more. 
> 
> Xen's boot image conforms to the 'multiboot' standard, which
> turns out not to be such a standard after all... In fact, Grub is
> about the only implementation I know of (other than our patched
> version of kexec)
> 
> I think multiboot is generally a good thing, and I can't think of
> any downside to using Grub over Lilo (other than lilo's -R
> option, for which standard Grub doesn't really have an
> equivalent.)

Every major distro includes my patch that provides 'lilo -R'
functionality for GRUB. At least a few also provide a handy wrapper
script to make it easier to use.

I think most people consider LILO obsolete these days.

 -- Keir


-------------------------------------------------------
This SF.Net email sponsored by Black Hat Briefings & Training.
Attend Black Hat Briefings & Training, Las Vegas July 24-29 - 
digital self defense, top technical experts, no vendor pitches, 
unmatched networking opportunities. Visit www.blackhat.com