domU crash

domU crash

[Patrick O'Rourke]

Apologies if this is a known issue (I scanned the archive / bugzilla and 
did not see it).

Anyway, I upated from the today's src tarball and did a make world / 
install.  Upon booting a domU I got the following crash:

> Bringing up loopback interface:  network-functions: line 137: cd: /var/run/netreport: No such file or directory
> [  OK  ]
> Bringing up interface eth0:  
> Determining IP information for eth0...Unable to handle kernel paging request at virtual address cc3de022
>  printing eip:
> *pde = ma 00000000 pa 55555000
> Oops: 0000 [#1]
> PREEMPT SMP 

I compared the domU .config with my previous working version from 8/11 
and noticed that CONFIG_XEN_NETDEV_GRANT_{TX,RX} options are now turned 
on (the only domU config changes).  So I rebuilt the domU and it booted 
fine, network came up, etc.

I also noticed that running an install from with the domU directory 
places the domU kernel in /boot/boot:

> [root@laurel linux-2.6.12-xenU]# make ARCH=xen install
>   CHK     include/linux/version.h
> make[1]: `arch/i386/kernel/asm-offsets.s' is up to date.
>   CHK     include/asm-xen/asm_offsets.h
>   CHK     include/linux/compile.h
>   CHK     usr/initramfs_list
> mkdir -p /boot/boot
> ln -f -s vmlinuz-2.6.12-xenU /boot/boot/vmlinuz-2.6.12-xenU
> rm -f /boot/boot/vmlinuz-2.6.12-xenU
> install -m0644 vmlinuz /boot/boot/vmlinuz-2.6.12-xenU
> install -m0644 vmlinux /boot/boot/vmlinux-syms-2.6.12-xenU
> install -m0664 .config /boot/boot/config-2.6.12-xenU
> install -m0664 System.map /boot/boot/System.map-2.6.12-xenU

Pat

-- 
Patrick O'Rourke
porourke@egenera.com

Re: domU crash

[Patrick O'Rourke]

Patrick O'Rourke wrote:

> I compared the domU .config with my previous working version from 8/11 
> and noticed that CONFIG_XEN_NETDEV_GRANT_{TX,RX} options are now turned 
> on (the only domU config changes).  So I rebuilt the domU and it booted 
> fine, network came up, etc.

Just to follow up, the reason is for the crash was that I had built my 
dom0 kernel with an older config (one I've been using for this machine) 
and as a result the dom0 kernel was still being built with these options 
off.  Once I enabled the options in both dom0 and domU, the issue went away.

Pat

-- 
Patrick O'Rourke
porourke@egenera.com

domU crash

[Stephan Böni]

[-- Attachment #1.1: Type: text/plain, Size: 2238 bytes --]

Well, i'm starting a domU but it crashes. What i've done wrong?
 
> cd /etc/xen
> xm create inhouse
Using config file "inhouse".
Started domain inhouse
>
> cat /var/log/xend.log
....
[2005-10-15 16:19:43 xend] DEBUG (image:156) initDomain: cpu=-1 mem_kb=2097152 ssidref=-1 dom=7
[2005-10-15 16:19:43 xend] DEBUG (XendDomainInfo:843) init_domain> Created domain=7 name=inhouse memory=2048
[2005-10-15 16:19:43 xend] INFO (XendRoot:141) EVENT> xend.domain.create ['inhouse', 7]
[2005-10-15 16:19:43 xend] INFO (image:217) buildDomain os=linux dom=7 vcpus=1
[2005-10-15 16:19:43 xend] DEBUG (image:255) dom            = 7
[2005-10-15 16:19:43 xend] DEBUG (image:256) image          = /boot/vmlinuz-xen
[2005-10-15 16:19:43 xend] DEBUG (image:257) store_evtchn   = 1
[2005-10-15 16:19:43 xend] DEBUG (image:258) console_evtchn = 2
[2005-10-15 16:19:43 xend] DEBUG (image:259) cmdline        =  ip=:1.2.3.4:::inhouse:eth0:dhcp 3
[2005-10-15 16:19:43 xend] DEBUG (image:260) ramdisk        = /boot/initrd-xen
[2005-10-15 16:19:43 xend] DEBUG (image:261) flags          = 0
[2005-10-15 16:19:43 xend] DEBUG (image:262) vcpus          = 1
[2005-10-15 16:19:45 xend] INFO (xsobj:326) DBMap>introduceDomain> 7 833153 <EventChannel dom1:0:22 dom2:7:1> /domain/7a1ede72-57f0-4367-b438-151252f83d82
[2005-10-15 16:19:45 xend] DEBUG (XendDomainInfo:698) writing key availability=online to path /cpu/0 in store
[2005-10-15 16:19:45 xend] WARNING (XendDomainInfo:1100) Unknown config field boot
[2005-10-15 16:19:45 xend] DEBUG (blkif:24) exception looking up device number for hda1: [Errno 2] No such file or directory: '/dev/hda1'
[2005-10-15 16:19:45 xend] ERROR (XendDomainInfo:443) ['vif', ['mac', 'aa:00:00:20:a0:fa']]
[2005-10-15 16:19:45 xend] INFO (XendRoot:141) EVENT> xend.domain.unpause ['inhouse', 7]
[2005-10-15 16:19:46 xend] DEBUG (XendDomain:233) XendDomain>reap> domain died name=inhouse id=7
[2005-10-15 16:19:46 xend] INFO (XendRoot:141) EVENT> xend.domain.exit ['inhouse', 7, 'crash']
>
> cat inhouse | grep -v ^#
kernel = "/boot/vmlinuz-xen"
ramdisk = "/boot/initrd-xen"
memory = 2048
name = "inhouse"
disk = [ 'phy:xenvg/inhouse-root,hda1,w' ]
dhcp="dhcp"
hostname = "inhouse"
extra = "3"
>

[-- Attachment #1.2: Type: text/html, Size: 3896 bytes --]

[-- Attachment #2: Type: text/plain, Size: 137 bytes --]

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

Re: domU crash

[Keir Fraser]

On 15 Oct 2005, at 15:28, Stephan Böni wrote:

> Well, i'm starting a domU but it crashes. What i've done wrong?
>  
> > cd /etc/xen
> > xm create inhouse
> Using config file "inhouse".
> Started domain inhouse
> >

xend is trying to look up the device number for /dev/hda1. Since your 
system doesn;t have a /dev/hda1, and you are presumably using udev, 
this lookup fails. If you created that node manually in /dev/ then all 
should be fine.

Really the device name should be passed to domU as a string, and 
resolved to a number there. Or xend can just have a table of simple 
name lookups for thing like hda, hdb, ...

  -- Keir

Re: domU crash

[Stephan Böni]

> xend is trying to look up the device number for /dev/hda1. Since your 
> system doesn;t have a /dev/hda1, and you are presumably using udev, 
> this lookup fails. If you created that node manually in /dev/ 
> then all 
> should be fine.
> 
> Really the device name should be passed to domU as a string, and 
> resolved to a number there. Or xend can just have a table of simple 
> name lookups for thing like hda, hdb, ...

Hi Keir

Now i've created it:

# cd /dev
# mknod hda b 3 0
# mknod hda1 b 3 1
# chown root.disk hda*
# chmod 664 hda*
# ls -l /dev/hd*
brw-rw-r--  1 root disk 3, 0 2005-10-15 16:52 /dev/hda
brw-rw-r--  1 root disk 3, 1 2005-10-15 16:53 /dev/hda1
#
# /etc/init.d/xend restart
Restarting xend (old pid 6559)                                done
#
# xm create /etc/xen/inhouse
Using config file "/etc/xen/inhouse".
Started domain inhouse
#
# xm list
Name              Id  Mem(MB)  CPU VCPU(s)  State  Time(s)
Domain-0           0      123    0      1   r----     22.8
#
# cat /var/log/xend.log
.....
[2005-10-15 16:55:19 xend] DEBUG (image:156) initDomain: cpu=-1 mem_kb=2097152 ssidref=-1 dom=2
[2005-10-15 16:55:19 xend] DEBUG (XendDomainInfo:843) init_domain> Created domain=2 name=inhouse memory=2048
[2005-10-15 16:55:19 xend] INFO (XendRoot:141) EVENT> xend.domain.create ['inhouse', 2]
[2005-10-15 16:55:19 xend] INFO (image:217) buildDomain os=linux dom=2 vcpus=1
[2005-10-15 16:55:19 xend] DEBUG (image:255) dom            = 2
[2005-10-15 16:55:19 xend] DEBUG (image:256) image          = /boot/vmlinuz-xen
[2005-10-15 16:55:19 xend] DEBUG (image:257) store_evtchn   = 1
[2005-10-15 16:55:19 xend] DEBUG (image:258) console_evtchn = 2
[2005-10-15 16:55:19 xend] DEBUG (image:259) cmdline        =  ip=:1.2.3.4:::inhouse:eth0:dhcp root=/dev/hda1 rw 3
[2005-10-15 16:55:19 xend] DEBUG (image:260) ramdisk        = /boot/initrd-xen
[2005-10-15 16:55:19 xend] DEBUG (image:261) flags          = 0
[2005-10-15 16:55:19 xend] DEBUG (image:262) vcpus          = 1
[2005-10-15 16:55:21 xend] INFO (xsobj:326) DBMap>introduceDomain> 2 833158 <EventChannel dom1:0:18 dom2:2:1> /domain/d3ec0c25-f952-4384-b620-7a53f2bc5bb7
[2005-10-15 16:55:21 xend] DEBUG (XendDomainInfo:698) writing key availability=online to path /cpu/0 in store
[2005-10-15 16:55:21 xend] WARNING (XendDomainInfo:1100) Unknown config field boot
[2005-10-15 16:55:21 xend] ERROR (XendDomainInfo:443) ['vif', ['mac', 'aa:00:00:39:63:6e']]
[2005-10-15 16:55:21 xend] INFO (XendRoot:141) EVENT> xend.domain.unpause ['inhouse', 2]
[2005-10-15 16:55:21 xend] DEBUG (XendDomain:233) XendDomain>reap> domain died name=inhouse id=2
[2005-10-15 16:55:21 xend] INFO (XendRoot:141) EVENT> xend.domain.exit ['inhouse', 2, 'crash']
.....
#
# cat /etc/xen/inhouse | grep -v ^#
kernel = "/boot/vmlinuz-xen"
ramdisk = "/boot/initrd-xen"
memory = 2048
name = "inhouse"
disk = [ 'phy:xenvg/inhouse-root,hda1,w' ]
dhcp="dhcp"
hostname = "inhouse"
root = "/dev/hda1 rw"
extra = "3"
#

It does still not work. :-(

Stephan

Re: domU crash

[Ted Kaczmarek]

On Sat, 2005-10-15 at 15:51 +0100, Keir Fraser wrote:
> On 15 Oct 2005, at 15:28, Stephan Böni wrote:
> 
> > Well, i'm starting a domU but it crashes. What i've done wrong?
> >  
> > > cd /etc/xen
> > > xm create inhouse
> > Using config file "inhouse".
> > Started domain inhouse
> > >
> 
> xend is trying to look up the device number for /dev/hda1. Since your 
> system doesn;t have a /dev/hda1, and you are presumably using udev, 
> this lookup fails. If you created that node manually in /dev/ then all 
> should be fine.
> 
> Really the device name should be passed to domU as a string, and 
> resolved to a number there. Or xend can just have a table of simple 
> name lookups for thing like hda, hdb, ...
> 
>   -- Keir
I am really confused on this one, as I can get it to work no matter what
I use sometimes, other times I have seen it where it won't work unless I
only use /dev/sdXX where XX does not exist at all on the dom0.

With changeset 7353 on my UP machine I got 100% success rate of domU
creation like this, all the domU's in this case are Centos 4.1, which
uses udev. The dom0 has a /dev/sda and /dev/sdb, if I used /dev/sdaX or
sdbX on this machine I always got domU creation failures, I switched all
the domU's to use /dev/sddX and 100% creation every time.



My SMP dom0 has sda, sdb and sdc, I use lvm's for domU's all located on
sdb.

disk = [ 'phy:VG01/idom1,sda1,w' ,'phy:VG01/domain1-swap,sda2,w']
root = "/dev/sda1"


What should I be using? 

The domU's will all come up like this,with roughly 30% failure rate on
domU creation and 20% failure rate on vif creation( still digging on
this so not a big deal) and reattempting create will resolve both
issues.


It would probably save everyone some grief if this would be clarified.
Based upon what you are saying the domU should be using a device
designation that really exists, yet I regularly see where that indeed is
not the case, and sometimes has even hindered successful domU creation.

If someone would be so kind as to explain exactly how this really should
be setup it would probably increase everyones productivity involved. I
would gladly create the doc if I knew for sure what is the proper setup.


Regards,
Ted

Re: domU crash

[Ewan Mellor]

On Mon, Oct 17, 2005 at 02:34:04PM +0200, Stephan Böni wrote:

> > Looks like the domain is crashing early during boot. Have you done a 
> > completely clean rebuild and reinstall of your domU kernel 
> > image? This 
> > is most often caused by an inconsistent kernel image...
> 
> I've updated now, but still i can't boot my domU. :-(

The log you sent shows the domain being successfully created and unpaused;
it should then start to boot.  What happens next?

Ewan.

Re: domU crash

[Stephan Böni]

> is most often caused by an inconsistent kernel image...
> > 
> > I've updated now, but still i can't boot my domU. :-(
> 
> The log you sent shows the domain being successfully created 
> and unpaused;
> it should then start to boot.  What happens next?

Nothing.

# xm list
Name              ID  Mem(MiB)  CPU  VCPUs  State   Time(s)
Domain-0           0       251    0      1  r-----     36.1
inhouse            8      1024    0      1  ------    252.2

Stephan

Re: domU crash

[Ewan Mellor]

On Mon, Oct 17, 2005 at 03:01:55PM +0200, Stephan Böni wrote:

> > is most often caused by an inconsistent kernel image...
> > > 
> > > I've updated now, but still i can't boot my domU. :-(
> > 
> > The log you sent shows the domain being successfully created 
> > and unpaused;
> > it should then start to boot.  What happens next?
> 
> Nothing.
> 
> # xm list
> Name              ID  Mem(MiB)  CPU  VCPUs  State   Time(s)
> Domain-0           0       251    0      1  r-----     36.1
> inhouse            8      1024    0      1  ------    252.2

You have a domain here, and it's already used 252 seconds of CPU time, so it's
doing _something_.  Is that time going up rapidly?  Try xm list a few times to
find out.  What's on the domain's console (xm console inhouse)?  And it's
Ctrl-] to exit the console again.

Ewan.

Re: domU crash

[Stephan Böni]

> > > The log you sent shows the domain being successfully created 
> > > and unpaused;
> > > it should then start to boot.  What happens next?
> > 
> > Nothing.
> > 
> > # xm list
> > Name              ID  Mem(MiB)  CPU  VCPUs  State   Time(s)
> > Domain-0           0       251    0      1  r-----     36.1
> > inhouse            8      1024    0      1  ------    252.2
> 
> You have a domain here, and it's already used 252 seconds of 
> CPU time, so it's
> doing _something_.  Is that time going up rapidly?  Try xm 
> list a few times to
> find out.  What's on the domain's console (xm console 
> inhouse)?  And it's
> Ctrl-] to exit the console again.

Oops. I've rebooted the hole system and now it works. :-)

Thanks.

Stephan

Re: domU crash

[Stephan Böni]

> > > > The log you sent shows the domain being successfully created 
> > > > and unpaused;
> > > > it should then start to boot.  What happens next?
> > > 
> > > Nothing.
> > > 
> > > # xm list
> > > Name              ID  Mem(MiB)  CPU  VCPUs  State   Time(s)
> > > Domain-0           0       251    0      1  r-----     36.1
> > > inhouse            8      1024    0      1  ------    252.2
> > 
> > You have a domain here, and it's already used 252 seconds of 
> > CPU time, so it's
> > doing _something_.  Is that time going up rapidly?  Try xm 
> > list a few times to
> > find out.  What's on the domain's console (xm console 
> > inhouse)?  And it's
> > Ctrl-] to exit the console again.
> 
> Oops. I've rebooted the hole system and now it works. :-)

Bullshit! It works only one time and then i've got the same problem:

# xm list
Name              ID  Mem(MiB)  CPU  VCPUs  State   Time(s)
Domain-0           0       251    0      1  r-----     26.4
inhouse            3      1024    0      1  ------     28.9
# xm list
Name              ID  Mem(MiB)  CPU  VCPUs  State   Time(s)
Domain-0           0       251    0      1  r-----     26.5
inhouse            3      1024    0      1  ------     36.1
# xm list
Name              ID  Mem(MiB)  CPU  VCPUs  State   Time(s)
Domain-0           0       251    0      1  r-----     26.8
inhouse            3      1024    0      1  ------    165.8
# xm list
Name              ID  Mem(MiB)  CPU  VCPUs  State   Time(s)
Domain-0           0       251    0      1  r-----     27.0
inhouse            3      1024    0      1  ------    178.8

# xm console inhouse

(nothing displayed!)

Stephan

Re: domU crash

[Ted Kaczmarek]

On Mon, 2005-10-17 at 15:34 +0200, Stephan Böni wrote:
> 
> > > > > The log you sent shows the domain being successfully created 
> > > > > and unpaused;
> > > > > it should then start to boot.  What happens next?
> > > > 
> > > > Nothing.
> > > > 
> > > > # xm list
> > > > Name              ID  Mem(MiB)  CPU  VCPUs  State   Time(s)
> > > > Domain-0           0       251    0      1  r-----     36.1
> > > > inhouse            8      1024    0      1  ------    252.2
> > > 
> > > You have a domain here, and it's already used 252 seconds of 
> > > CPU time, so it's
> > > doing _something_.  Is that time going up rapidly?  Try xm 
> > > list a few times to
> > > find out.  What's on the domain's console (xm console 
> > > inhouse)?  And it's
> > > Ctrl-] to exit the console again.
> > 
> > Oops. I've rebooted the hole system and now it works. :-)
> 
> Bullshit! It works only one time and then i've got the same problem:
> 
> # xm list
> Name              ID  Mem(MiB)  CPU  VCPUs  State   Time(s)
> Domain-0           0       251    0      1  r-----     26.4
> inhouse            3      1024    0      1  ------     28.9
> # xm list
> Name              ID  Mem(MiB)  CPU  VCPUs  State   Time(s)
> Domain-0           0       251    0      1  r-----     26.5
> inhouse            3      1024    0      1  ------     36.1
> # xm list
> Name              ID  Mem(MiB)  CPU  VCPUs  State   Time(s)
> Domain-0           0       251    0      1  r-----     26.8
> inhouse            3      1024    0      1  ------    165.8
> # xm list
> Name              ID  Mem(MiB)  CPU  VCPUs  State   Time(s)
> Domain-0           0       251    0      1  r-----     27.0
> inhouse            3      1024    0      1  ------    178.8
> 
> # xm console inhouse
> 
> (nothing displayed!)
> 
> Stephan
> 
Make sure you start the domU in console and have a serial console
connected to xend. You will likely have a better chance of seeing what
is happening.

Regards,
Ted

Re: domU crash

[Stephan Böni]

[-- Attachment #1: Type: text/plain, Size: 515 bytes --]

> Make sure you start the domU in console and have a serial console
> connected to xend. You will likely have a better chance of seeing what
> is happening.

How must i do that?

In normal case i do:

# cd /etc/xen
# xm create inhouse
# xm list
----> repeat until a "b" is in the state.
# xm console inhouse
----> the login prompt will be displayed.

In Xen 2.0 i've got displayed all boot messages with xm console, but now i see nothing like that. How can i view this boot informations?

Stephan

[-- Attachment #2: Type: text/plain, Size: 138 bytes --]

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

Re: domU crash

[Stephan Böni]

[-- Attachment #1: Type: text/plain, Size: 2842 bytes --]

Very strange things occure, when starting my domU called "inhouse". It has a lot problems with mounting my drives. Here is the configuration file:

#  -*- mode: python; -*-
 
# configuration name:
name    = "inhouse"
 
# usable ram:
memory  = 2048

# usable cpus:
cpu     = 0
maxcpus = 1
 
# kernel und initrd:
kernel  = "/boot/vmlinuz-xen"
ramdisk = "/boot/initrd-xen"
 
# boot device:
root    = "/dev/hda1"
 
# boot to run level:
extra   = "3"
 
# network interface:
vif  = [ 'mac=aa:cc:00:00:00:01, bridge=xenbr0' ]
hostname = name
 
# storage devices:
disk = [ 'phy:xenvg/inhouse-root,hda1,w',
         'phy:xenvg/inhouse-cdb,hda2,w',
         'phy:xenvg/inhouse-comserver,hda3,w',
         'phy:xenvg/inhouse-docarchive,hda4,w',
         'phy:xenvg/inhouse-home,hda5,w',
         'phy:xenvg/inhouse-journal,hda6,w',
         'phy:xenvg/inhouse-mailserver,hda7,w',
         'phy:xenvg/inhouse-pos,hda8,w',
         'phy:xenvg/inhouse-pub,hda9,w',
         'phy:xenvg/inhouse-srv,hda10,w',
         'phy:xenvg/inhouse-tmp,hda11,w',
         'phy:xenvg/inhouse-transfer,hda12,w',
         'phy:xenvg/inhouse-usr,hda13,w',
         'phy:xenvg/inhouse-var,hda14,w',
         'phy:xenvg/inhouse-swap,hdb1,w',
         'phy:xenvg/susedvd,hdc1,r' ]

And here you see, what the mount displays after successful boot of this domU, which is rarely:

# mount
/dev/sda8 on / type reiserfs (rw,acl,user_xattr)
proc on /proc type proc (rw)
sysfs on /sys type sysfs (rw)
tmpfs on /dev/shm type tmpfs (rw)
devpts on /dev/pts type devpts (rw,mode=0620,gid=5)
usbfs on /proc/bus/usb type usbfs (rw)
/dev/mapper/xenvg-inhouse--root on / type reiserfs (rw)

Well, /dev/sda8 is another bootable disk. Horrible but true: it mounts realy strange things. The /etc/fstab file looks like that:

/dev/hda1   /                 reiserfs  defaults  1 1
/dev/hda2   /cdb              reiserfs  defaults  1 2
/dev/hda3   /data/comserver   reiserfs  defaults  1 2
/dev/hda4   /data/docarchive  reiserfs  defaults  1 2
/dev/hda5   /home             reiserfs  defaults  1 2
/dev/hda6   /journal          reiserfs  defaults  1 2
/dev/hda7   /data/mailserver  reiserfs  defaults  1 2
/dev/hda8   /data/pos         reiserfs  defaults  1 2
/dev/hda9   /data/pub         reiserfs  defaults  1 2
/dev/hda10  /srv              reiserfs  defaults  1 2
/dev/hda11  /tmp              reiserfs  defaults  1 2
/dev/hda12  /data/transfer    reiserfs  defaults  1 2
/dev/hda13  /usr              reiserfs  defaults  1 2
/dev/hda14  /var              reiserfs  defaults  1 2
/dev/hda15  /data/webgate     reiserfs  defaults  1 2
/dev/hdb1   swap              swap      pri=42    0 0
/dev/hdc1   /setup/susedvd    reiserfs  ro,noauto

Do i something wrong or exists there a realy stange bug in Xen?

Stephan


[-- Attachment #2: Type: text/plain, Size: 138 bytes --]

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

Re: domU crash

[Stephan Böni]

[-- Attachment #1: Type: text/plain, Size: 539 bytes --]


> > > Make sure you start the domU in console and have a serial console
> > > connected to xend. You will likely have a better chance 
> of seeing what
> > > is happening.
> > 
> > How must i do that?
> > 
> > In normal case i do:
> > 
> > # cd /etc/xen
> > # xm create inhouse
> 
> xm create -c inhouse

That's not really better:


# xm create -c inhouse
Using config file "inhouse".
Started domain inhouse
 
 
Welcome to SUSE LINUX 10.0 (X86-64) - Kernel 2.6.13-15c-xen (tty1).
 
 
(none) login:


Stephan

[-- Attachment #2: Type: text/plain, Size: 138 bytes --]

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

Re: domU crash

[Stephan Böni]

[-- Attachment #1: Type: text/plain, Size: 3291 bytes --]

Deleting the /etc/mtab solved this.

Stephan

> -----Ursprüngliche Nachricht-----
> Von: Stephan Böni 
> Gesendet: Montag, 17. Oktober 2005 18:11
> An: xen-devel@lists.xensource.com; Ewan Mellor; Ted Kaczmarek
> Betreff: Re: [Xen-devel] domU crash
> 
> 
> Very strange things occure, when starting my domU called 
> "inhouse". It has a lot problems with mounting my drives. 
> Here is the configuration file:
> 
> #  -*- mode: python; -*-
>  
> # configuration name:
> name    = "inhouse"
>  
> # usable ram:
> memory  = 2048
> 
> # usable cpus:
> cpu     = 0
> maxcpus = 1
>  
> # kernel und initrd:
> kernel  = "/boot/vmlinuz-xen"
> ramdisk = "/boot/initrd-xen"
>  
> # boot device:
> root    = "/dev/hda1"
>  
> # boot to run level:
> extra   = "3"
>  
> # network interface:
> vif  = [ 'mac=aa:cc:00:00:00:01, bridge=xenbr0' ]
> hostname = name
>  
> # storage devices:
> disk = [ 'phy:xenvg/inhouse-root,hda1,w',
>          'phy:xenvg/inhouse-cdb,hda2,w',
>          'phy:xenvg/inhouse-comserver,hda3,w',
>          'phy:xenvg/inhouse-docarchive,hda4,w',
>          'phy:xenvg/inhouse-home,hda5,w',
>          'phy:xenvg/inhouse-journal,hda6,w',
>          'phy:xenvg/inhouse-mailserver,hda7,w',
>          'phy:xenvg/inhouse-pos,hda8,w',
>          'phy:xenvg/inhouse-pub,hda9,w',
>          'phy:xenvg/inhouse-srv,hda10,w',
>          'phy:xenvg/inhouse-tmp,hda11,w',
>          'phy:xenvg/inhouse-transfer,hda12,w',
>          'phy:xenvg/inhouse-usr,hda13,w',
>          'phy:xenvg/inhouse-var,hda14,w',
>          'phy:xenvg/inhouse-swap,hdb1,w',
>          'phy:xenvg/susedvd,hdc1,r' ]
> 
> And here you see, what the mount displays after successful 
> boot of this domU, which is rarely:
> 
> # mount
> /dev/sda8 on / type reiserfs (rw,acl,user_xattr)
> proc on /proc type proc (rw)
> sysfs on /sys type sysfs (rw)
> tmpfs on /dev/shm type tmpfs (rw)
> devpts on /dev/pts type devpts (rw,mode=0620,gid=5)
> usbfs on /proc/bus/usb type usbfs (rw)
> /dev/mapper/xenvg-inhouse--root on / type reiserfs (rw)
> 
> Well, /dev/sda8 is another bootable disk. Horrible but true: 
> it mounts realy strange things. The /etc/fstab file looks like that:
> 
> /dev/hda1   /                 reiserfs  defaults  1 1
> /dev/hda2   /cdb              reiserfs  defaults  1 2
> /dev/hda3   /data/comserver   reiserfs  defaults  1 2
> /dev/hda4   /data/docarchive  reiserfs  defaults  1 2
> /dev/hda5   /home             reiserfs  defaults  1 2
> /dev/hda6   /journal          reiserfs  defaults  1 2
> /dev/hda7   /data/mailserver  reiserfs  defaults  1 2
> /dev/hda8   /data/pos         reiserfs  defaults  1 2
> /dev/hda9   /data/pub         reiserfs  defaults  1 2
> /dev/hda10  /srv              reiserfs  defaults  1 2
> /dev/hda11  /tmp              reiserfs  defaults  1 2
> /dev/hda12  /data/transfer    reiserfs  defaults  1 2
> /dev/hda13  /usr              reiserfs  defaults  1 2
> /dev/hda14  /var              reiserfs  defaults  1 2
> /dev/hda15  /data/webgate     reiserfs  defaults  1 2
> /dev/hdb1   swap              swap      pri=42    0 0
> /dev/hdc1   /setup/susedvd    reiserfs  ro,noauto
> 
> Do i something wrong or exists there a realy stange bug in Xen?
> 
> Stephan
> 
> 

[-- Attachment #2: Type: text/plain, Size: 138 bytes --]

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