Xen devel - Jul 2006 - xenconsole: Could not read tty from store: No such file or directory

Hi,

I built xen  3.0-unstable with VT-x support on  SLES 10.1. The machine
boots ok and I start  xend manually. I encounter  the following
problem is with guest creation:
$xm create -c /etc/xen/vm/SLES9x32_16G

Using config file "/etc/xen/vm/SLES9x32_16G".
Started domain SLES9x32_16G
xenconsole: Could not read tty from store: No such file or directory

The guest is created but remains blocked.

Using xm console <ID> does not resolve the problem.
I saw previous postings describing the same error. In one instance the was
due
to absence  of root filesystem for guest,  which appears to be  not the case
here  (details below). In  another  instance the  consoled was not running.
I checked consoled using ps -ef and found it running with ''?''
in the tty
field.

Below is infomation about  the changeset, guest configuration, and xen
grub configuration. Please  let me know if more  information is needed to
determine the problem. It''s most likely I''m missing something
in the setup.

Thanks,
  --Mahmoud

-----------------------------------------------

Latest ChangeSet: Thu Jul 13 15:33:08 2006 +0100 10680:f692a0a476c5

----------------------------------------------

/etc/xen/vm/SLES9x32_16G:

disk = [ ''file:/sdd/sles9x32_16G,ioemu:hda,w'' ]
memory = 1024
vcpus = 1
builder = ''hvm''
device_model = ''/usr/lib64/xen/bin/qemu-dm''
kernel = ''/usr/lib/xen/boot/hvmloader''
name = ''SLES9x32_16G''
vif = [ ''type=ioemu,bridge=xenbr0'' ]
stdvga = 1
sdl = 1
vnc = 0
vncviewer = 1
ne2000 = 0
localtime = 0
on_poweroff = ''destroy''
on_reboot = ''restart''
on_crash = ''restart''
boot = ''c''
cdrom = ''/mnt/sles9rc5x32_dvd.iso''
----------------------------------------

Xen''s entry  in /boot/grub/menu.lst:

title  Xen --  SUSE Linux  Enterprise Server  10
root (hd0,1)
kernel  /boot/xen.gz dom0_mem=2000m
module /boot/vmlinuz-2.6-xen  root=/dev/sda2   vga=0x314  resume=/dev/sda1
splash=silent showopts module /boot/initrd-2.6-xen


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

Mahmoud,

Often times this message occurs with hvm domains because qemu-dm
does not start properly.  The file /var/log/qemu-dm.*.log is created 
when you try to start an hvm domain.  It should provide at least 
a hint to what the problem is.

Also, a lot of qemu changes have been checked in recently.  You may 
want to update to a more recent changeset.

Natasha


On Wed, Jul 26, 2006 at 02:20:36PM -0700, Mahmoud Elhaddad
wrote:
> Hi,
> 
> I built xen  3.0-unstable with VT-x support on  SLES 10.1. The machine
> boots ok and I start  xend manually. I encounter  the following
> problem is with guest creation:
> $xm create -c /etc/xen/vm/SLES9x32_16G
> 
> Using config file "/etc/xen/vm/SLES9x32_16G".
> Started domain SLES9x32_16G
> xenconsole: Could not read tty from store: No such file or directory
> 
> The guest is created but remains blocked.
> 
> Using xm console <ID> does not resolve the problem.
> I saw previous postings describing the same error. In one instance the was
> due
> to absence  of root filesystem for guest,  which appears to be  not the
case
> here  (details below). In  another  instance the  consoled was not running.
> I checked consoled using ps -ef and found it running with
''?'' in the tty
> field.
> 
> Below is infomation about  the changeset, guest configuration, and xen
> grub configuration. Please  let me know if more  information is needed to
> determine the problem. It''s most likely I''m missing
something in the setup.
> 
> Thanks,
>  --Mahmoud
> 
> -----------------------------------------------
> 
> Latest ChangeSet: Thu Jul 13 15:33:08 2006 +0100 10680:f692a0a476c5
> 
> ----------------------------------------------
> 
> /etc/xen/vm/SLES9x32_16G:
> 
> disk = [ ''file:/sdd/sles9x32_16G,ioemu:hda,w'' ]
> memory = 1024
> vcpus = 1
> builder = ''hvm''
> device_model = ''/usr/lib64/xen/bin/qemu-dm''
> kernel = ''/usr/lib/xen/boot/hvmloader''
> name = ''SLES9x32_16G''
> vif = [ ''type=ioemu,bridge=xenbr0'' ]
> stdvga = 1
> sdl = 1
> vnc = 0
> vncviewer = 1
> ne2000 = 0
> localtime = 0
> on_poweroff = ''destroy''
> on_reboot = ''restart''
> on_crash = ''restart''
> boot = ''c''
> cdrom = ''/mnt/sles9rc5x32_dvd.iso''
> ----------------------------------------
> 
> Xen''s entry  in /boot/grub/menu.lst:
> 
> title  Xen --  SUSE Linux  Enterprise Server  10
> root (hd0,1)
> kernel  /boot/xen.gz dom0_mem=2000m
> module /boot/vmlinuz-2.6-xen  root=/dev/sda2   vga=0x314  resume=/dev/sda1
> splash=silent showopts module /boot/initrd-2.6-xen
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xensource.com
> http://lists.xensource.com/xen-devel

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

natasha@us.ltcfwd.linux.ibm.com wrote on 07/27/2006 09:48:34 AM:
> Mahmoud,
>
> Often times this message occurs with hvm domains because qemu-dm
> does not start properly.  The file /var/log/qemu-dm.*.log is created
> when you try to start an hvm domain.  It should provide at least
> a hint to what the problem is.
>
> Also, a lot of qemu changes have been checked in recently.  You may
> want to update to a more recent changeset.
>
> Natasha
Mahmoud,

I saw the same message when my qemu-dm died.  I debugged it down to find it
was caused by an unitialized variable.  I posted a patch.  (
http://lists.xensource.com/archives/html/xen-devel/2006-07/msg00850.html)
You might try it to see if you are running into the same problem.  My patch
is a little over zealous in initializing variables, but that shouldn''t
hurt.  Christian Limpach said he will apply a subset of the patch (
http://lists.xensource.com/archives/html/xen-devel/2006-07/msg00899.html).
It hasn''t hit the xen-unstable tree yet, but it should be coming soon.

Steve D.


_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel

Thank you Natasha and Steve. I rebuilt using the latest changeset and aplied
the patch. The patch did work...the SLES guest now boots but freezes while
ReiserFS is "checking the transaction log".
It''s not clear to me if it''s another  problem with qemu.
The qemu.*.log now looks like this:
domid: 3
qemu: the number of cpus is 1
shared page at pfn:3ffff, mfn: 1d7e19
char device redirected to /dev/pts/2
Issued domain 3 poweroff

Thanks,
 -Mahmoud

On 7/27/06, Steve Dobbelstein <steved@us.ibm.com>
wrote:
>
> natasha@us.ltcfwd.linux.ibm.com wrote on 07/27/2006 09:48:34 AM:
>
> > Mahmoud,
> >
> > Often times this message occurs with hvm domains because qemu-dm
> > does not start properly.  The file /var/log/qemu-dm.*.log is created
> > when you try to start an hvm domain.  It should provide at least
> > a hint to what the problem is.
> >
> > Also, a lot of qemu changes have been checked in recently.  You may
> > want to update to a more recent changeset.
> >
> > Natasha
>
> Mahmoud,
>
> I saw the same message when my qemu-dm died.  I debugged it down to find
> it
> was caused by an unitialized variable.  I posted a patch.  (
> http://lists.xensource.com/archives/html/xen-devel/2006-07/msg00850.html)
> You might try it to see if you are running into the same problem.  My
> patch
> is a little over zealous in initializing variables, but that
shouldn''t
> hurt.  Christian Limpach said he will apply a subset of the patch (
> http://lists.xensource.com/archives/html/xen-devel/2006-07/msg00899.html
> ).
> It hasn''t hit the xen-unstable tree yet, but it should be coming
soon.
>
> Steve D.
>
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@lists.xensource.com
> http://lists.xensource.com/xen-devel
>

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel