From: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
To: micah anderson <micah@riseup.net>
Cc: xen-devel@lists.xensource.com
Subject: Re: domU crashing, trying to determine why
Date: Fri, 14 Jan 2011 12:45:30 -0500 [thread overview]
Message-ID: <20110114174530.GA26732@dumpdata.com> (raw)
In-Reply-To: <87pqrzpf2t.fsf@algae.riseup.net>
On Fri, Jan 14, 2011 at 12:31:22PM -0500, micah anderson wrote:
> On Mon, 3 Jan 2011 12:14:25 -0500, Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> wrote:
> > On Mon, Jan 03, 2011 at 12:03:34PM -0500, micah anderson wrote:
> > > On Mon, 3 Jan 2011 11:40:31 -0500, Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> wrote:
> > > > > > I've got this setup:
> > > > > >
> > > > > > vcpus=2
> > > > > > cpus=[1,3]
> > > >
> > > > Is there anything in the Xen hypervisor ring buffer? You might need to enable the
> > > > full logging of the guest and dom0 to get anything there.
> > >
> > > How do I get at the ring buffer, and what specifically should I enable
> >
> > 'xm dmesg'
>
> (XEN) mm.c:645:d2 Non-privileged (2) attempt to map I/O space 000000f0
Hmm, that should not happen anymore, as there are is a bunch of
code guarding domains from trying to touch 0->0x100. Unless your guest
has the old stuff.
> (XEN) traps.c:2544: GPF (0068): ff1886c2 -> ff1887c9
> (XEN) domain_crash_sync called from entry.S (ff188600)
> (XEN) Domain 2 (vcpu#1) crashed on cpu#3:
> (XEN) ----[ Xen-3.2-1 x86_32p debug=n Tainted: C ]----
Ugh, ancient hypervisor.
> (XEN) CPU: 3
> (XEN) EIP: 0061:[<c01013a7>]
Now use gdb to lookup that EIP value. Or use your guest System.map
to see what function this could map too.
> (XEN) EFLAGS: 00000246 CONTEXT: guest
> (XEN) eax: 00000000 ebx: 00000001 ecx: 00000000 edx: ed447f90
> (XEN) esi: 00000001 edi: 00000001 ebp: 00000000 esp: ed447f84
> (XEN) cr0: 8005003b cr4: 000006f0 cr3: 00bdac80 cr2: 095346dc
> (XEN) ds: 007b es: 007b fs: 00d8 gs: 0000 ss: 0069 cs: 0061
> (XEN) Guest stack trace from esp=ed447f84:
> (XEN) c0105f52 ffffffff 00000001 673bfa5d 0002dba1 00000000 00000000 ffffffff
> (XEN) c01028ab c0102810 00000000 00000000 00000000 00000000 00000000 00000000
And potentially c01028ab c0105f52 ..
> (XEN) 00000000 00000000 00000000 00000000 00000000 00000000 000000d8 00000000
> (XEN) 00000000 00000000 00000000 00000000 00000000 00000000 00000000
>
next prev parent reply other threads:[~2011-01-14 17:45 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-01-01 19:16 domU crashing, trying to determine why Micah Anderson
2011-01-01 21:41 ` Pasi Kärkkäinen
2011-01-02 0:43 ` micah anderson
2011-01-02 12:59 ` Pasi Kärkkäinen
2011-01-02 18:15 ` micah anderson
2011-01-02 18:39 ` Pasi Kärkkäinen
2011-01-03 3:41 ` micah anderson
2011-01-03 5:34 ` Pasi Kärkkäinen
2011-01-03 6:32 ` micah anderson
2011-01-03 16:40 ` Konrad Rzeszutek Wilk
2011-01-03 17:03 ` micah anderson
2011-01-03 17:14 ` Konrad Rzeszutek Wilk
2011-01-14 17:31 ` micah anderson
2011-01-14 17:45 ` Konrad Rzeszutek Wilk [this message]
2011-01-04 10:54 ` Ian Campbell
2011-01-04 19:49 ` micah anderson
2011-01-05 13:27 ` Ian Campbell
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20110114174530.GA26732@dumpdata.com \
--to=konrad.wilk@oracle.com \
--cc=micah@riseup.net \
--cc=xen-devel@lists.xensource.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).