Fedora xen - Apr 2006 - pygrub failed

Hello, FC5 Xen folks.

When I tried to install Xen3.0 on FC5, it has failed in making a guest OS image.
It seems to be because that pygrub.py can not recognize my dom0 menu.lst
correctly.
Or, maybe it is wrong that I used LVM, but this is default of current FC5
(Why?).

BTW, I am testing FC5 Xen dom0 on VMware 5.5.1, and it runs, although
many guys have reported that it has a lot of trouble.
For me, the problem is not VMware or dom0, but installer of domU.

Any help?

                 --- Okajima, Jun. Tokyo, Japan.



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

[root@localhost ~]# xm list
Name                              ID Mem(MiB) VCPUs State  Time(s)
Domain-0                           0      164     1 r-----   427.8
[root@localhost ~]# xenguest-install.py
What is the name of your virtual machine? FC5_1
 How much RAM should be allocated (in megabytes)? 256
 What would you like to use as the disk (path)? /XEN/FC5_1.img
 What is the install location?
http://ring.sakura.ad.jp/archives/linux/fedora/linux/core/5/i386/os/


Starting install...
Using config file "/etc/xen/FC5_1".
Error: Error creating domain: The privileged domain did not balloon!
If your install has exited, you can restart your guest by running
''xm create -c FC5_1''.  Otherwise, you can reconnect to the
console
by running ''xm console FC5_1''
[root@localhost ~]# xm create -c FC5_1
Using config file "/etc/xen/FC5_1".
Traceback (most recent call last):
  File "/usr/bin/pygrub", line 256, in ?
    cf = get_config(file, isconfig)
  File "/usr/bin/pygrub", line 139, in get_config
    raise RuntimeError, "Unable to read filesystem"
RuntimeError: Unable to read filesystem
Error: Boot loader didn''t return any data!
[root@localhost ~]# cat /etc/xen/
FC5_1            scripts/         xmexample.vti
auto/            xend-config.sxp  xmexample1
qemu-ifup        xmexample.hvm    xmexample2
[root@localhost ~]# cat /etc/xen/FC5_1
# Automatically generated xen config file
name = "FC5_1"
memory = "256"
disk = [ ''file:/XEN/FC5_1.img,xvda,w'' ]
vif = [ ''mac=00:16:3e:60:8f:ea'' ]
bootloader="/usr/bin/pygrub"

on_reboot   = ''restart''
on_crash    = ''restart''
[root@localhost ~]# cat /boot/grub/menu.lst
# grub.conf generated by anaconda
#
# Note that you do not have to rerun grub after making changes to this file
# NOTICE:  You have a /boot partition.  This means that
#          all kernel and initrd paths are relative to /boot/, eg.
#          root (hd0,0)
#          kernel /vmlinuz-version ro root=/dev/VolGroup00/LogVol00
#          initrd /initrd-version.img
#boot=/dev/hda
default=2
timeout=5
splashimage=(hd0,0)/grub/splash.xpm.gz
hiddenmenu
title Fedora Core (2.6.16-1.2080_FC5xenU)
        root (hd0,0)
        kernel /vmlinuz-2.6.16-1.2080_FC5xenU ro root=/dev/VolGroup00/LogVol00
rhgb quiet
        initrd /initrd-2.6.16-1.2080_FC5xenU.img
title Fedora Core (2.6.16-1.2080_FC5xen0)
        root (hd0,0)
        kernel /xen.gz-2.6.16-1.2080_FC5
        module /vmlinuz-2.6.16-1.2080_FC5xen0 ro root=/dev/VolGroup00/LogVol00
rhgb quiet
        module /initrd-2.6.16-1.2080_FC5xen0.img
title Fedora Core (2.6.16-1.2080_FC5)
        root (hd0,0)
        kernel /vmlinuz-2.6.16-1.2080_FC5 ro root=/dev/VolGroup00/LogVol00 rhgb
quiet
        initrd /initrd-2.6.16-1.2080_FC5.img
title Fedora Core (2.6.15-1.2054_FC5)
        root (hd0,0)
        kernel /vmlinuz-2.6.15-1.2054_FC5 ro root=/dev/VolGroup00/LogVol00 rhgb
quiet
        initrd /initrd-2.6.15-1.2054_FC5.img
[root@localhost ~]# cat /proc/partitions
major minor  #blocks  name

   3     0    8388608 hda
   3     1     104391 hda1
   3     2    8281507 hda2
 253     0    7831552 dm-0
 253     1     393216 dm-1
[root@localhost ~]# xm dmesg
 __  __            _____  ___   ___
 \ \/ /___ _ __   |___ / / _ \ / _ \
  \  // _ \ ''_ \    |_ \| | | | | | |
  /  \  __/ | | |  ___) | |_| | |_| |
 /_/\_\___|_| |_| |____(_)___(_)___/

 http://www.cl.cam.ac.uk/netos/xen
 University of Cambridge Computer Laboratory

 Xen version 3.0.0 (bhcompile@build.redhat.com) (gcc version 4.1.0 20060304 (Red
Hat 4.1.0-3)) Tue
Mar 28 03:35:03 EST 2006
 Latest ChangeSet: unavailable

(XEN) Physical RAM map:
(XEN)  0000000000000000 - 000000000009f800 (usable)
(XEN)  000000000009f800 - 00000000000a0000 (reserved)
(XEN)  00000000000dc000 - 0000000000100000 (reserved)
(XEN)  0000000000100000 - 000000000bef0000 (usable)
(XEN)  000000000bef0000 - 000000000befc000 (ACPI data)
(XEN)  000000000befc000 - 000000000bf00000 (ACPI NVS)
(XEN)  000000000bf00000 - 000000000c000000 (usable)
(XEN)  00000000fec00000 - 00000000fec10000 (reserved)
(XEN)  00000000fee00000 - 00000000fee01000 (reserved)
(XEN)  00000000fffe0000 - 0000000100000000 (reserved)
(XEN) System RAM: 191MB (196156kB)
(XEN) Xen heap: 10MB (10344kB)
(XEN) Using scheduler: Simple EDF Scheduler (sedf)
(XEN) PAE disabled.
(XEN) found SMP MP-table at 000f6ce0
(XEN) DMI present.
(XEN) Using APIC driver default
(XEN) ACPI: RSDP (v000 PTLTD                                 ) @ 0x000f6c70
(XEN) ACPI: RSDT (v001 PTLTD    RSDT   0x06040000  LTP 0x00000000) @ 0x0bef7c02
(XEN) ACPI: FADT (v001 INTEL  440BX    0x06040000 PTL  0x000f4240) @ 0x0befbf14
(XEN) ACPI: MADT (v001 PTLTD     APIC   0x06040000  LTP 0x00000000) @ 0x0befbf88
(XEN) ACPI: BOOT (v001 PTLTD  $SBFTBL$ 0x06040000  LTP 0x00000001) @ 0x0befbfd8
(XEN) ACPI: DSDT (v001 PTLTD  Custom   0x06040000 MSFT 0x0100000d) @ 0x00000000
(XEN) ACPI: Local APIC address 0xfee00000
(XEN) ACPI: LAPIC (acpi_id[0x00] lapic_id[0x00] enabled)
(XEN) Processor #0 15:12 APIC version 17
(XEN) ACPI: LAPIC_NMI (acpi_id[0x00] high edge lint[0x1])
(XEN) ACPI: IOAPIC (id[0x01] address[0xfec00000] gsi_base[0])
(XEN) IOAPIC[0]: apic_id 1, version 17, address 0xfec00000, GSI 0-23
(XEN) ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 2 high edge)
(XEN) ACPI: IRQ0 used by override.
(XEN) ACPI: IRQ2 used by override.
(XEN) Enabling APIC mode:  Flat.  Using 1 I/O APICs
(XEN) Using ACPI (MADT) for SMP configuration information
(XEN) Initializing CPU#0
(XEN) Detected 1599.572 MHz processor.
(XEN) CPU0: AMD Flush Filter disabled
(XEN) CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line)
(XEN) CPU: L2 Cache: 128K (64 bytes/line)
(XEN) Intel machine check architecture supported.
(XEN) Intel machine check reporting enabled on CPU#0.
(XEN) CPU0: AMD Sempron(tm) Processor 2600+ stepping 02
(XEN) Total of 1 processors activated.
(XEN) ENABLING IO-APIC IRQs
(XEN) ..TIMER: vector=0x31 apic1=0 pin1=2 apic2=-1 pin2=-1
(XEN) Platform timer is 1.193MHz PIT
(XEN) Brought up 1 CPUs
(XEN) Machine check exception polling timer started.
(XEN) *** LOADING DOMAIN 0 ***
(XEN) Domain 0 kernel supports features = { 0000001f }.
(XEN) Domain 0 kernel requires features = { 00000000 }.
(XEN) PHYSICAL MEMORY ARRANGEMENT:
(XEN)  Dom0 alloc.:   0a000000->0b000000 (37859 pages to be allocated)
(XEN) VIRTUAL MEMORY ARRANGEMENT:
(XEN)  Loaded kernel: c0100000->c0455b7c
(XEN)  Init. ramdisk: c0456000->c07e2200
(XEN)  Phys-Mach map: c07e3000->c080bf8c
(XEN)  Start info:    c080c000->c080d000
(XEN)  Page tables:   c080d000->c0811000
(XEN)  Boot stack:    c0811000->c0812000
(XEN)  TOTAL:         c0000000->c0c00000
(XEN)  ENTRY ADDRESS: c0100000
(XEN) Dom0 has maximum 1 VCPUs
(XEN) Initrd len 0x38c200, start at 0xc0456000
(XEN) Scrubbing Free RAM: ..done.
(XEN) Xen trace buffers: disabled
(XEN) *** Serial input -> DOM0 (type ''CTRL-a'' three times
to switch input to Xen).

[root@localhost ~]# cat /etc/redhat-release
Fedora Core release 5 (Bordeaux)
[root@localhost ~]#

I believe the error "Error: Error creating domain: The privileged  
domain did not balloon!" happens when you don''t have enough
available
memory for the domU to boot.  The xenguest-install.py script tells  
you to restart the domain by running "xm create -c xxxx'', but I
think
the error message is just a generic one that is printed if the script  
crashes out, and it it is the wrong thing to do in this case.

When I had the same problem, I had to allocate less memory to dom0  
and re-run the xenguest-install script.  I see you have about 196MB  
of RAM.  Since the FC5 script needs 256MB of RAM, I don''t think  
you''ll be able to install it unless you increase the amount of RAM  
you have.  But since the machine is a virtual one in VMWare, this  
shouldn''t be too difficult.


--Winston



On Apr 2, 2006, at 7:49 AM, Jun OKAJIMA wrote:
>
> Hello, FC5 Xen folks.
>
> When I tried to install Xen3.0 on FC5, it has failed in making a  
> guest OS image.
> It seems to be because that pygrub.py can not recognize my dom0  
> menu.lst correctly.
> Or, maybe it is wrong that I used LVM, but this is default of  
> current FC5 (Why?).
>
> BTW, I am testing FC5 Xen dom0 on VMware 5.5.1, and it runs, although
> many guys have reported that it has a lot of trouble.
> For me, the problem is not VMware or dom0, but installer of domU.
>
> Any help?
>
>                  --- Okajima, Jun. Tokyo, Japan.
>
>
>
> ---------------------------
>
> [root@localhost ~]# xm list
> Name                              ID Mem(MiB) VCPUs State  Time(s)
> Domain-0                           0      164     1 r-----   427.8
> [root@localhost ~]# xenguest-install.py
> What is the name of your virtual machine? FC5_1
>  How much RAM should be allocated (in megabytes)? 256
>  What would you like to use as the disk (path)? /XEN/FC5_1.img
>  What is the install location? http://ring.sakura.ad.jp/archives/ 
> linux/fedora/linux/core/5/i386/os/
>
>
> Starting install...
> Using config file "/etc/xen/FC5_1".
> Error: Error creating domain: The privileged domain did not balloon!
> If your install has exited, you can restart your guest by running
> ''xm create -c FC5_1''.  Otherwise, you can reconnect to
the console
> by running ''xm console FC5_1''
> [root@localhost ~]# xm create -c FC5_1
> Using config file "/etc/xen/FC5_1".
> Traceback (most recent call last):
>   File "/usr/bin/pygrub", line 256, in ?
>     cf = get_config(file, isconfig)
>   File "/usr/bin/pygrub", line 139, in get_config
>     raise RuntimeError, "Unable to read filesystem"
> RuntimeError: Unable to read filesystem
> Error: Boot loader didn''t return any data!
> [root@localhost ~]# cat /etc/xen/
> FC5_1            scripts/         xmexample.vti
> auto/            xend-config.sxp  xmexample1
> qemu-ifup        xmexample.hvm    xmexample2
> [root@localhost ~]# cat /etc/xen/FC5_1
> # Automatically generated xen config file
> name = "FC5_1"
> memory = "256"
> disk = [ ''file:/XEN/FC5_1.img,xvda,w'' ]
> vif = [ ''mac=00:16:3e:60:8f:ea'' ]
> bootloader="/usr/bin/pygrub"
>
> on_reboot   = ''restart''
> on_crash    = ''restart''
> [root@localhost ~]# cat /boot/grub/menu.lst
> # grub.conf generated by anaconda
> #
> # Note that you do not have to rerun grub after making changes to  
> this file
> # NOTICE:  You have a /boot partition.  This means that
> #          all kernel and initrd paths are relative to /boot/, eg.
> #          root (hd0,0)
> #          kernel /vmlinuz-version ro root=/dev/VolGroup00/LogVol00
> #          initrd /initrd-version.img
> #boot=/dev/hda
> default=2
> timeout=5
> splashimage=(hd0,0)/grub/splash.xpm.gz
> hiddenmenu
> title Fedora Core (2.6.16-1.2080_FC5xenU)
>         root (hd0,0)
>         kernel /vmlinuz-2.6.16-1.2080_FC5xenU ro root=/dev/ 
> VolGroup00/LogVol00 rhgb quiet
>         initrd /initrd-2.6.16-1.2080_FC5xenU.img
> title Fedora Core (2.6.16-1.2080_FC5xen0)
>         root (hd0,0)
>         kernel /xen.gz-2.6.16-1.2080_FC5
>         module /vmlinuz-2.6.16-1.2080_FC5xen0 ro root=/dev/ 
> VolGroup00/LogVol00 rhgb quiet
>         module /initrd-2.6.16-1.2080_FC5xen0.img
> title Fedora Core (2.6.16-1.2080_FC5)
>         root (hd0,0)
>         kernel /vmlinuz-2.6.16-1.2080_FC5 ro root=/dev/VolGroup00/ 
> LogVol00 rhgb quiet
>         initrd /initrd-2.6.16-1.2080_FC5.img
> title Fedora Core (2.6.15-1.2054_FC5)
>         root (hd0,0)
>         kernel /vmlinuz-2.6.15-1.2054_FC5 ro root=/dev/VolGroup00/ 
> LogVol00 rhgb quiet
>         initrd /initrd-2.6.15-1.2054_FC5.img
> [root@localhost ~]# cat /proc/partitions
> major minor  #blocks  name
>
>    3     0    8388608 hda
>    3     1     104391 hda1
>    3     2    8281507 hda2
>  253     0    7831552 dm-0
>  253     1     393216 dm-1
> [root@localhost ~]# xm dmesg
>  __  __            _____  ___   ___
>  \ \/ /___ _ __   |___ / / _ \ / _ \
>   \  // _ \ ''_ \    |_ \| | | | | | |
>   /  \  __/ | | |  ___) | |_| | |_| |
>  /_/\_\___|_| |_| |____(_)___(_)___/
>
>  http://www.cl.cam.ac.uk/netos/xen
>  University of Cambridge Computer Laboratory
>
>  Xen version 3.0.0 (bhcompile@build.redhat.com) (gcc version 4.1.0  
> 20060304 (Red Hat 4.1.0-3)) Tue
> Mar 28 03:35:03 EST 2006
>  Latest ChangeSet: unavailable
>
> (XEN) Physical RAM map:
> (XEN)  0000000000000000 - 000000000009f800 (usable)
> (XEN)  000000000009f800 - 00000000000a0000 (reserved)
> (XEN)  00000000000dc000 - 0000000000100000 (reserved)
> (XEN)  0000000000100000 - 000000000bef0000 (usable)
> (XEN)  000000000bef0000 - 000000000befc000 (ACPI data)
> (XEN)  000000000befc000 - 000000000bf00000 (ACPI NVS)
> (XEN)  000000000bf00000 - 000000000c000000 (usable)
> (XEN)  00000000fec00000 - 00000000fec10000 (reserved)
> (XEN)  00000000fee00000 - 00000000fee01000 (reserved)
> (XEN)  00000000fffe0000 - 0000000100000000 (reserved)
> (XEN) System RAM: 191MB (196156kB)
> (XEN) Xen heap: 10MB (10344kB)
> (XEN) Using scheduler: Simple EDF Scheduler (sedf)
> (XEN) PAE disabled.
> (XEN) found SMP MP-table at 000f6ce0
> (XEN) DMI present.
> (XEN) Using APIC driver default
> (XEN) ACPI: RSDP (v000 PTLTD                                 ) @  
> 0x000f6c70
> (XEN) ACPI: RSDT (v001 PTLTD    RSDT   0x06040000  LTP 0x00000000)  
> @ 0x0bef7c02
> (XEN) ACPI: FADT (v001 INTEL  440BX    0x06040000 PTL  0x000f4240)  
> @ 0x0befbf14
> (XEN) ACPI: MADT (v001 PTLTD     APIC   0x06040000  LTP 0x00000000)  
> @ 0x0befbf88
> (XEN) ACPI: BOOT (v001 PTLTD  $SBFTBL$ 0x06040000  LTP 0x00000001)  
> @ 0x0befbfd8
> (XEN) ACPI: DSDT (v001 PTLTD  Custom   0x06040000 MSFT 0x0100000d)  
> @ 0x00000000
> (XEN) ACPI: Local APIC address 0xfee00000
> (XEN) ACPI: LAPIC (acpi_id[0x00] lapic_id[0x00] enabled)
> (XEN) Processor #0 15:12 APIC version 17
> (XEN) ACPI: LAPIC_NMI (acpi_id[0x00] high edge lint[0x1])
> (XEN) ACPI: IOAPIC (id[0x01] address[0xfec00000] gsi_base[0])
> (XEN) IOAPIC[0]: apic_id 1, version 17, address 0xfec00000, GSI 0-23
> (XEN) ACPI: INT_SRC_OVR (bus 0 bus_irq 0 global_irq 2 high edge)
> (XEN) ACPI: IRQ0 used by override.
> (XEN) ACPI: IRQ2 used by override.
> (XEN) Enabling APIC mode:  Flat.  Using 1 I/O APICs
> (XEN) Using ACPI (MADT) for SMP configuration information
> (XEN) Initializing CPU#0
> (XEN) Detected 1599.572 MHz processor.
> (XEN) CPU0: AMD Flush Filter disabled
> (XEN) CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/ 
> line)
> (XEN) CPU: L2 Cache: 128K (64 bytes/line)
> (XEN) Intel machine check architecture supported.
> (XEN) Intel machine check reporting enabled on CPU#0.
> (XEN) CPU0: AMD Sempron(tm) Processor 2600+ stepping 02
> (XEN) Total of 1 processors activated.
> (XEN) ENABLING IO-APIC IRQs
> (XEN) ..TIMER: vector=0x31 apic1=0 pin1=2 apic2=-1 pin2=-1
> (XEN) Platform timer is 1.193MHz PIT
> (XEN) Brought up 1 CPUs
> (XEN) Machine check exception polling timer started.
> (XEN) *** LOADING DOMAIN 0 ***
> (XEN) Domain 0 kernel supports features = { 0000001f }.
> (XEN) Domain 0 kernel requires features = { 00000000 }.
> (XEN) PHYSICAL MEMORY ARRANGEMENT:
> (XEN)  Dom0 alloc.:   0a000000->0b000000 (37859 pages to be allocated)
> (XEN) VIRTUAL MEMORY ARRANGEMENT:
> (XEN)  Loaded kernel: c0100000->c0455b7c
> (XEN)  Init. ramdisk: c0456000->c07e2200
> (XEN)  Phys-Mach map: c07e3000->c080bf8c
> (XEN)  Start info:    c080c000->c080d000
> (XEN)  Page tables:   c080d000->c0811000
> (XEN)  Boot stack:    c0811000->c0812000
> (XEN)  TOTAL:         c0000000->c0c00000
> (XEN)  ENTRY ADDRESS: c0100000
> (XEN) Dom0 has maximum 1 VCPUs
> (XEN) Initrd len 0x38c200, start at 0xc0456000
> (XEN) Scrubbing Free RAM: ..done.
> (XEN) Xen trace buffers: disabled
> (XEN) *** Serial input -> DOM0 (type ''CTRL-a'' three
times to switch
> input to Xen).
>
> [root@localhost ~]# cat /etc/redhat-release
> Fedora Core release 5 (Bordeaux)
> [root@localhost ~]#
>
> --
> Fedora-xen mailing list
> Fedora-xen@redhat.com
> https://www.redhat.com/mailman/listinfo/fedora-xen

>I believe the error "Error: Error creating domain: The privileged  
>domain did not balloon!" happens when you don''t have enough
available
>memory for the domU to boot.  The xenguest-install.py script tells  
>you to restart the domain by running "xm create -c xxxx'', but
I think
>the error message is just a generic one that is printed if the script  
>crashes out, and it it is the wrong thing to do in this case.
>
>When I had the same problem, I had to allocate less memory to dom0  
>and re-run the xenguest-install script.  I see you have about 196MB  
>of RAM.  Since the FC5 script needs 256MB of RAM, I don''t think  
>you''ll be able to install it unless you increase the amount of RAM
>you have.  But since the machine is a virtual one in VMWare, this  
>shouldn''t be too difficult.
>

Thank you for replying.
Unfortunately, I tried same stuff with 512MB mem but failed.
What I suspect now is, a disk image would have to be formatted.
xenguest-install.py just makes plain empty sparse file, but is this
enough? I suspect it should be formatted and partitioned and so on.
You succeeded to do an install on FC5? How you did it?

Check the code of pygrub.
See this:
   116      fs = None
   117      for fstype in grub.fsys.fstypes.values():
   118          if fstype.sniff_magic(fn, offset):
   119              fs = fstype.open_fs(fn, offset)
   120              break
   121
   122      if fs is not None:
(snip)
   138      else:
   139          raise RuntimeError, "Unable to read filesystem"

What happened here is, in short, like this:

   # Check diskimage formatted as described in grub config
   117      for fstype in grub.fsys.fstypes.values():
   118          if fstype.sniff_magic(fn, offset):
   # If not formatted properly, raise error.
   138      else:
   139          raise RuntimeError, "Unable to read filesystem"

                                --- Okajima, Jun. Tokyo, Japan.


-------------
[root@localhost ~]# cat -n /usr/bin/pygrub
     1  #!/usr/bin/python
     2  #
     3  # pygrub - simple python-based bootloader for Xen
     4  #
     5  # Copyright 2005 Red Hat, Inc.
     6  # Jeremy Katz <katzj@redhat.com>
     7  #
     8  # This software may be freely redistributed under the terms of the GNU
     9  # general public license.
    10  #
    11  # You should have received a copy of the GNU General Public License
    12  # along with this program; if not, write to the Free Software
    13  # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
    14  #
    15
    16  import os, sys, string, struct, tempfile
    17  import logging
    18
    19  import curses, _curses, curses.wrapper
    20  import getopt
    21
    22  sys.path = [ ''/usr/lib/python'' ] + sys.path
    23
    24  import grub.GrubConf
    25  import grub.fsys
    26
    27  PYGRUB_VER = 0.3
    28
    29
    30  def draw_window():
    31      stdscr = curses.initscr()
    32      if hasattr(curses, ''use_default_colors''):
    33          curses.use_default_colors()
    34      try:
    35          curses.curs_set(0)
    36      except _curses.error:
    37          pass
    38
    39      stdscr.addstr(1, 4, "pyGRUB  version %s" %(PYGRUB_VER,))
    40
    41      win = curses.newwin(10, 74, 2, 1)
    42      win.box()
    43      win.refresh()
    44
    45      stdscr.addstr(12, 5, "Use the U and D keys to select which
entry is highlighted.")
    46      stdscr.addstr(13, 5, "Press enter to boot the selected OS.
''e'' to edit the")
    47      stdscr.addstr(14, 5, "commands before booting,
''a'' to modify the kernel arguments ")
    48      stdscr.addstr(15, 5, "before booting, or ''c''
for a command line.")
    49      stdscr.addch(12, 13, curses.ACS_UARROW)
    50      stdscr.addch(12, 19, curses.ACS_DARROW)
    51      (y, x) = stdscr.getmaxyx()
    52      stdscr.move(y - 1, x - 1)
    53
    54      stdscr.refresh()
    55      return (stdscr, win)
    56
    57  def fill_entries(win, cfg, selected):
    58      y = 0
    59
    60      for i in cfg.images:
    61          if (0, y) > win.getmaxyx():
    62              break
    63          if y == selected:
    64              attr = curses.A_REVERSE
    65          else:
    66              attr = 0
    67          win.addstr(y + 1, 2, i.title.ljust(70), attr)
    68          y += 1
    69      win.refresh()
    70
    71  def select(win, line):
    72      win.attron(curses.A_REVERSE)
    73      win.redrawln(line + 1, 1)
    74      win.refresh()
    75
    76  def is_disk_image(file):
    77      fd = os.open(file, os.O_RDONLY)
    78      buf = os.read(fd, 512)
    79      os.close(fd)
    80
    81      if len(buf) >= 512 and struct.unpack("H", buf[0x1fe:
0x200]) == (0xaa55,):
    82          return True
    83      return False
    84
    85  SECTOR_SIZE=512
    86  def get_active_offset(file):
    87      """Find the offset for the start of the first active
partition in the
    88      disk image file."""
    89      fd = os.open(file, os.O_RDONLY)
    90      buf = os.read(fd, 512)
    91      for poff in (446, 462, 478, 494): # partition offsets
    92          # active partition has 0x80 as the first byte
    93          if struct.unpack("<c", buf[poff:poff+1]) ==
(''\x80'',):
    94              return struct.unpack("<L",
buf[poff+8:poff+12])[0] * SECTOR_SIZE
    95      return -1
    96
    97  def get_config(fn, isconfig = False):
    98      if not os.access(fn, os.R_OK):
    99          raise RuntimeError, "Unable to access %s" %(fn,)
   100
   101      cf = grub.GrubConf.GrubConfigFile()
   102
   103      if isconfig:
   104          # set the config file and parse it
   105          cf.filename = fn
   106          cf.parse()
   107          return cf
   108
   109      offset = 0
   110      if is_disk_image(fn):
   111          offset = get_active_offset(fn)
   112          if offset == -1:
   113              raise RuntimeError, "Unable to find active partition on
disk"
   114
   115      # open the image and read the grub config
   116      fs = None
   117      for fstype in grub.fsys.fstypes.values():
   118          if fstype.sniff_magic(fn, offset):
   119              fs = fstype.open_fs(fn, offset)
   120              break
   121
   122      if fs is not None:
   123          grubfile = None
   124          for f in ("/boot/grub/menu.lst",
"/boot/grub/grub.conf",
   125                    "/grub/menu.lst",
"/grub/grub.conf"):
   126              if fs.file_exist(f):
   127                  grubfile = f
   128                  break
   129          if grubfile is None:
   130              raise RuntimeError, "we couldn''t find
/boot/grub{menu.lst,grub.conf} " + \
   131                                  "in the image provided. halt!"
   132          f = fs.open_file(grubfile)
   133          buf = f.read()
   134          f.close()
   135          fs.close()
   136          # then parse the grub config
   137          cf.parse(buf)
   138      else:
   139          raise RuntimeError, "Unable to read filesystem"
   140
   141      return cf
   142
   143  def get_entry_idx(cf, entry):
   144      # first, see if the given entry is numeric
   145      try:
   146          idx = string.atoi(entry)
   147          return idx
   148      except ValueError:
   149          pass
   150
   151      # it''s not, now check the labels for a match
   152      for i in range(len(cf.images)):
   153          if entry == cf.images[i].title:
   154              return i
   155
   156      return None
   157
   158  def main(cf = None):
   159      mytime = 0
   160      timeout = int(cf.timeout)
   161
   162      (stdscr, win) = draw_window()
   163      stdscr.timeout(1000)
   164      selected = cf.default
   165
   166      while (timeout == -1 or mytime < int(timeout)):
   167          if timeout != -1 and mytime != -1:
   168              stdscr.addstr(20, 5, "Will boot selected entry in %2d
seconds"
   169                            %(int(timeout) - mytime))
   170          else:
   171              stdscr.addstr(20, 5, " " * 80)
   172
   173          fill_entries(win, cf, selected)
   174          c = stdscr.getch()
   175          if mytime != -1:
   176              mytime += 1
   177  #        if c == ord(''q''):
   178  #            selected = -1
   179  #            break
   180          if c == ord(''c''):
   181              # FIXME: needs to go to command line mode
   182              continue
   183          elif c == ord(''a''):
   184              # FIXME: needs to go to append mode
   185              continue
   186          elif c == ord(''e''):
   187              # FIXME: needs to go to edit mode
   188              continue
   189          elif c in (curses.KEY_ENTER, ord(''\n''),
ord(''\r'')):
   190              break
   191          elif c == curses.KEY_UP:
   192              mytime = -1
   193              selected -= 1
   194          elif c == curses.KEY_DOWN:
   195              mytime = -1
   196              selected += 1
   197          else:
   198              pass
   199
   200          # bound at the top and bottom
   201          if selected < 0:
   202              selected = 0
   203          elif selected >= len(cf.images):
   204              selected = len(cf.images) - 1
   205
   206      if selected >= 0:
   207          return selected
   208
   209  if __name__ == "__main__":
   210      sel = None
   211
   212      def run_main(scr, *args):
   213          global sel
   214          sel = main(cf)
   215
   216      def usage():
   217          print >> sys.stderr, "Usage: %s [-q|--quiet]
[--output=] [--entry=] <image>" %
(sys.argv[0],)
   218
   219      try:
   220          opts, args = getopt.gnu_getopt(sys.argv[1:],
''qh::'',
   221                                     ["quiet", "help",
"output=", "entry=",
   222                                      "isconfig"])
   223      except getopt.GetoptError:
   224          usage()
   225          sys.exit(1)
   226
   227      if len(args) < 1:
   228          usage()
   229          sys.exit(1)
   230      file = args[0]
   231
   232      output = None
   233      entry = None
   234      interactive = True
   235      isconfig = False
   236      for o, a in opts:
   237          if o in ("-q", "--quiet"):
   238              interactive = False
   239          elif o in ("-h", "--help"):
   240              usage()
   241              sys.exit()
   242          elif o in ("--output",):
   243              output = a
   244          elif o in ("--entry",):
   245              entry = a
   246              # specifying the entry to boot implies non-interactive
   247              interactive = False
   248          elif o in ("--isconfig",):
   249              isconfig = True
   250
   251      if output is None or output == "-":
   252          fd = sys.stdout.fileno()
   253      else:
   254          fd = os.open(output, os.O_WRONLY)
   255
   256      cf = get_config(file, isconfig)
   257      if interactive:
   258          curses.wrapper(run_main)
   259      else:
   260          sel = cf.default
   261
   262      # set the entry to boot as requested
   263      if entry is not None:
   264          idx = get_entry_idx(cf, entry)
   265          if idx is not None and idx > 0 and idx < len(cf.images):
   266              sel = idx
   267
   268      img = cf.images[sel]
   269      print "Going to boot %s" %(img.title)
   270      print "  kernel: %s" %(img.kernel[1],)
   271      if img.initrd:
   272          print "  initrd: %s" %(img.initrd[1],)
   273
   274      offset = 0
   275      if is_disk_image(file):
   276          offset = get_active_offset(file)
   277          if offset == -1:
   278              raise RuntimeError, "Unable to find active partition on
disk"
   279
   280      # read the kernel and initrd onto the hostfs
   281      fs = None
   282      for fstype in grub.fsys.fstypes.values():
   283          if fstype.sniff_magic(file, offset):
   284              fs = fstype.open_fs(file, offset)
   285              break
   286
   287      if fs is None:
   288          raise RuntimeError, "Unable to open filesystem"
   289
   290      kernel = fs.open_file(img.kernel[1],).read()
   291      (tfd, fn) = tempfile.mkstemp(prefix="vmlinuz.",
dir="/var/lib/xen")
   292      os.write(tfd, kernel)
   293      os.close(tfd)
   294      sxp = "linux (kernel %s)" %(fn,)
   295
   296      if img.initrd:
   297          initrd = fs.open_file(img.initrd[1],).read()
   298          (tfd, fn) = tempfile.mkstemp(prefix="initrd.",
dir="/var/lib/xen")
   299          os.write(tfd, initrd)
   300          os.close(tfd)
   301          sxp += "(ramdisk %s)" %(fn,)
   302      else:
   303          initrd = None
   304      sxp += "(args ''%s'')" %(img.args,)
   305
   306      sys.stdout.flush()
   307      os.write(fd, sxp)
   308
[root@localhost ~]#

On Po, dub 03, 2006 at 01:40:01 +0900, Jun OKAJIMA
wrote:
> Unfortunately, I tried same stuff with 512MB mem but failed.
> What I suspect now is, a disk image would have to be formatted.
> xenguest-install.py just makes plain empty sparse file, but is this
> enough? I suspect it should be formatted and partitioned and so on.
 No. The empty file is enough, because xenguest-install.py doesn''t use
 pygrub for domU installation boot. You need prepare your virtual disk
 by anaconda which is started by xenguest-install.py. The pygrub
 script is possible use for already __successfully__ installed domU
 only.

    Karel

-- 
 Karel Zak  <kzak@redhat.com>

The installer will create the disk image and partition it.  If you do  
not successfully run to the installer, "xm create -c xxxx" will not  
work.  The script gives bad instructions when an error occurs.

The Xen hypervisor takes 32MB of RAM (I think), so you have to make  
sure that your dom0 gets less than 224MB.  I gave mine 192MB, with  
the following in grub.conf:
   kernel /xen.gz-2.6.15-1.2054_FC5 dom0_mem=196608


In my opinion, the xenguest-install.py method isn''t very good at this  
point.
- The installer gives poor feedback when something goes wrong, like  
when memory is too low.
- pygrub crashes if your terminal emulation is a little weird
- It creates a whole disk image instead of importing partitions.   
This makes it difficult to access domU filesystems from dom0, but  
this is often necessary for management.
- It doesn''t provide an option for creating images on LVM volumes.



An easier solution might be to use the FC5 image from jailtime.org.   
If you use the images from their site, keep in mind that you have to  
use a kernel from outside of the image; the kernel would have to be  
readable from your dom0.

If you use the stock FC5-xenU kernel, it does not have the xennet  
driver compiled in.   You would have to install the kernel-xenU  
package onto the domU disk image to provide the necessary driver  
(assuming the disk image is mounted at /mnt/tmp):
   yum --installroot=/mnt/tmp install kernel-xenU
And add the following to the domU''s /etc/modprobe.conf:
   alias eth0 xennet

Or you could use a kernel from xensource that has the xennet driver  
compiled into the kernel.  I don''t know why the developers decided to  
take the driver out of the kernel.  Pretty much everyone would want  
networking on their domU machines, and adding it in this way is just  
more complicated.



--Winston


On Apr 2, 2006, at 12:40 PM, Jun OKAJIMA wrote:
>
>> I believe the error "Error: Error creating domain: The privileged
>> domain did not balloon!" happens when you don''t have
enough available
>> memory for the domU to boot.  The xenguest-install.py script tells
>> you to restart the domain by running "xm create -c xxxx'',
but I think
>> the error message is just a generic one that is printed if the script
>> crashes out, and it it is the wrong thing to do in this case.
>>
>> When I had the same problem, I had to allocate less memory to dom0
>> and re-run the xenguest-install script.  I see you have about 196MB
>> of RAM.  Since the FC5 script needs 256MB of RAM, I don''t
think
>> you''ll be able to install it unless you increase the amount of
RAM
>> you have.  But since the machine is a virtual one in VMWare, this
>> shouldn''t be too difficult.
>>
>
>
> Thank you for replying.
> Unfortunately, I tried same stuff with 512MB mem but failed.
> What I suspect now is, a disk image would have to be formatted.
> xenguest-install.py just makes plain empty sparse file, but is this
> enough? I suspect it should be formatted and partitioned and so on.
> You succeeded to do an install on FC5? How you did it?
>
> Check the code of pygrub.
> See this:
>    116      fs = None
>    117      for fstype in grub.fsys.fstypes.values():
>    118          if fstype.sniff_magic(fn, offset):
>    119              fs = fstype.open_fs(fn, offset)
>    120              break
>    121
>    122      if fs is not None:
> (snip)
>    138      else:
>    139          raise RuntimeError, "Unable to read filesystem"
>
> What happened here is, in short, like this:
>
>    # Check diskimage formatted as described in grub config
>    117      for fstype in grub.fsys.fstypes.values():
>    118          if fstype.sniff_magic(fn, offset):
>    # If not formatted properly, raise error.
>    138      else:
>    139          raise RuntimeError, "Unable to read filesystem"
>
>                                 --- Okajima, Jun. Tokyo, Japan.
>
>
> -------------
> [root@localhost ~]# cat -n /usr/bin/pygrub
>      1  #!/usr/bin/python
>      2  #
>      3  # pygrub - simple python-based bootloader for Xen
>      4  #
>      5  # Copyright 2005 Red Hat, Inc.
>      6  # Jeremy Katz <katzj@redhat.com>
>      7  #
>      8  # This software may be freely redistributed under the terms  
> of the GNU
>      9  # general public license.
>     10  #
>     11  # You should have received a copy of the GNU General Public  
> License
>     12  # along with this program; if not, write to the Free Software
>     13  # Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
>     14  #
>     15
>     16  import os, sys, string, struct, tempfile
>     17  import logging
>     18
>     19  import curses, _curses, curses.wrapper
>     20  import getopt
>     21
>     22  sys.path = [ ''/usr/lib/python'' ] + sys.path
>     23
>     24  import grub.GrubConf
>     25  import grub.fsys
>     26
>     27  PYGRUB_VER = 0.3
>     28
>     29
>     30  def draw_window():
>     31      stdscr = curses.initscr()
>     32      if hasattr(curses, ''use_default_colors''):
>     33          curses.use_default_colors()
>     34      try:
>     35          curses.curs_set(0)
>     36      except _curses.error:
>     37          pass
>     38
>     39      stdscr.addstr(1, 4, "pyGRUB  version %s"
%(PYGRUB_VER,))
>     40
>     41      win = curses.newwin(10, 74, 2, 1)
>     42      win.box()
>     43      win.refresh()
>     44
>     45      stdscr.addstr(12, 5, "Use the U and D keys to select  
> which entry is highlighted.")
>     46      stdscr.addstr(13, 5, "Press enter to boot the selected  
> OS. ''e'' to edit the")
>     47      stdscr.addstr(14, 5, "commands before booting,
''a'' to
> modify the kernel arguments ")
>     48      stdscr.addstr(15, 5, "before booting, or
''c'' for a
> command line.")
>     49      stdscr.addch(12, 13, curses.ACS_UARROW)
>     50      stdscr.addch(12, 19, curses.ACS_DARROW)
>     51      (y, x) = stdscr.getmaxyx()
>     52      stdscr.move(y - 1, x - 1)
>     53
>     54      stdscr.refresh()
>     55      return (stdscr, win)
>     56
>     57  def fill_entries(win, cfg, selected):
>     58      y = 0
>     59
>     60      for i in cfg.images:
>     61          if (0, y) > win.getmaxyx():
>     62              break
>     63          if y == selected:
>     64              attr = curses.A_REVERSE
>     65          else:
>     66              attr = 0
>     67          win.addstr(y + 1, 2, i.title.ljust(70), attr)
>     68          y += 1
>     69      win.refresh()
>     70
>     71  def select(win, line):
>     72      win.attron(curses.A_REVERSE)
>     73      win.redrawln(line + 1, 1)
>     74      win.refresh()
>     75
>     76  def is_disk_image(file):
>     77      fd = os.open(file, os.O_RDONLY)
>     78      buf = os.read(fd, 512)
>     79      os.close(fd)
>     80
>     81      if len(buf) >= 512 and struct.unpack("H",
buf[0x1fe:
> 0x200]) == (0xaa55,):
>     82          return True
>     83      return False
>     84
>     85  SECTOR_SIZE=512
>     86  def get_active_offset(file):
>     87      """Find the offset for the start of the first
active
> partition in the
>     88      disk image file."""
>     89      fd = os.open(file, os.O_RDONLY)
>     90      buf = os.read(fd, 512)
>     91      for poff in (446, 462, 478, 494): # partition offsets
>     92          # active partition has 0x80 as the first byte
>     93          if struct.unpack("<c", buf[poff:poff+1]) ==
(''\x80'',):
>     94              return struct.unpack("<L",
buf[poff+8:poff+12])
> [0] * SECTOR_SIZE
>     95      return -1
>     96
>     97  def get_config(fn, isconfig = False):
>     98      if not os.access(fn, os.R_OK):
>     99          raise RuntimeError, "Unable to access %s" %(fn,)
>    100
>    101      cf = grub.GrubConf.GrubConfigFile()
>    102
>    103      if isconfig:
>    104          # set the config file and parse it
>    105          cf.filename = fn
>    106          cf.parse()
>    107          return cf
>    108
>    109      offset = 0
>    110      if is_disk_image(fn):
>    111          offset = get_active_offset(fn)
>    112          if offset == -1:
>    113              raise RuntimeError, "Unable to find active  
> partition on disk"
>    114
>    115      # open the image and read the grub config
>    116      fs = None
>    117      for fstype in grub.fsys.fstypes.values():
>    118          if fstype.sniff_magic(fn, offset):
>    119              fs = fstype.open_fs(fn, offset)
>    120              break
>    121
>    122      if fs is not None:
>    123          grubfile = None
>    124          for f in ("/boot/grub/menu.lst",
"/boot/grub/
> grub.conf",
>    125                    "/grub/menu.lst",
"/grub/grub.conf"):
>    126              if fs.file_exist(f):
>    127                  grubfile = f
>    128                  break
>    129          if grubfile is None:
>    130              raise RuntimeError, "we couldn''t find
/boot/grub
> {menu.lst,grub.conf} " + \
>    131                                  "in the image provided.
halt!"
>    132          f = fs.open_file(grubfile)
>    133          buf = f.read()
>    134          f.close()
>    135          fs.close()
>    136          # then parse the grub config
>    137          cf.parse(buf)
>    138      else:
>    139          raise RuntimeError, "Unable to read filesystem"
>    140
>    141      return cf
>    142
>    143  def get_entry_idx(cf, entry):
>    144      # first, see if the given entry is numeric
>    145      try:
>    146          idx = string.atoi(entry)
>    147          return idx
>    148      except ValueError:
>    149          pass
>    150
>    151      # it''s not, now check the labels for a match
>    152      for i in range(len(cf.images)):
>    153          if entry == cf.images[i].title:
>    154              return i
>    155
>    156      return None
>    157
>    158  def main(cf = None):
>    159      mytime = 0
>    160      timeout = int(cf.timeout)
>    161
>    162      (stdscr, win) = draw_window()
>    163      stdscr.timeout(1000)
>    164      selected = cf.default
>    165
>    166      while (timeout == -1 or mytime < int(timeout)):
>    167          if timeout != -1 and mytime != -1:
>    168              stdscr.addstr(20, 5, "Will boot selected entry  
> in %2d seconds"
>    169                            %(int(timeout) - mytime))
>    170          else:
>    171              stdscr.addstr(20, 5, " " * 80)
>    172
>    173          fill_entries(win, cf, selected)
>    174          c = stdscr.getch()
>    175          if mytime != -1:
>    176              mytime += 1
>    177  #        if c == ord(''q''):
>    178  #            selected = -1
>    179  #            break
>    180          if c == ord(''c''):
>    181              # FIXME: needs to go to command line mode
>    182              continue
>    183          elif c == ord(''a''):
>    184              # FIXME: needs to go to append mode
>    185              continue
>    186          elif c == ord(''e''):
>    187              # FIXME: needs to go to edit mode
>    188              continue
>    189          elif c in (curses.KEY_ENTER, ord(''\n''),
ord(''\r'')):
>    190              break
>    191          elif c == curses.KEY_UP:
>    192              mytime = -1
>    193              selected -= 1
>    194          elif c == curses.KEY_DOWN:
>    195              mytime = -1
>    196              selected += 1
>    197          else:
>    198              pass
>    199
>    200          # bound at the top and bottom
>    201          if selected < 0:
>    202              selected = 0
>    203          elif selected >= len(cf.images):
>    204              selected = len(cf.images) - 1
>    205
>    206      if selected >= 0:
>    207          return selected
>    208
>    209  if __name__ == "__main__":
>    210      sel = None
>    211
>    212      def run_main(scr, *args):
>    213          global sel
>    214          sel = main(cf)
>    215
>    216      def usage():
>    217          print >> sys.stderr, "Usage: %s [-q|--quiet] [--
> output=] [--entry=] <image>" %
> (sys.argv[0],)
>    218
>    219      try:
>    220          opts, args = getopt.gnu_getopt(sys.argv[1:],
''qh::'',
>    221                                     ["quiet",
"help",
> "output=", "entry=",
>    222                                      "isconfig"])
>    223      except getopt.GetoptError:
>    224          usage()
>    225          sys.exit(1)
>    226
>    227      if len(args) < 1:
>    228          usage()
>    229          sys.exit(1)
>    230      file = args[0]
>    231
>    232      output = None
>    233      entry = None
>    234      interactive = True
>    235      isconfig = False
>    236      for o, a in opts:
>    237          if o in ("-q", "--quiet"):
>    238              interactive = False
>    239          elif o in ("-h", "--help"):
>    240              usage()
>    241              sys.exit()
>    242          elif o in ("--output",):
>    243              output = a
>    244          elif o in ("--entry",):
>    245              entry = a
>    246              # specifying the entry to boot implies non- 
> interactive
>    247              interactive = False
>    248          elif o in ("--isconfig",):
>    249              isconfig = True
>    250
>    251      if output is None or output == "-":
>    252          fd = sys.stdout.fileno()
>    253      else:
>    254          fd = os.open(output, os.O_WRONLY)
>    255
>    256      cf = get_config(file, isconfig)
>    257      if interactive:
>    258          curses.wrapper(run_main)
>    259      else:
>    260          sel = cf.default
>    261
>    262      # set the entry to boot as requested
>    263      if entry is not None:
>    264          idx = get_entry_idx(cf, entry)
>    265          if idx is not None and idx > 0 and idx < len 
> (cf.images):
>    266              sel = idx
>    267
>    268      img = cf.images[sel]
>    269      print "Going to boot %s" %(img.title)
>    270      print "  kernel: %s" %(img.kernel[1],)
>    271      if img.initrd:
>    272          print "  initrd: %s" %(img.initrd[1],)
>    273
>    274      offset = 0
>    275      if is_disk_image(file):
>    276          offset = get_active_offset(file)
>    277          if offset == -1:
>    278              raise RuntimeError, "Unable to find active  
> partition on disk"
>    279
>    280      # read the kernel and initrd onto the hostfs
>    281      fs = None
>    282      for fstype in grub.fsys.fstypes.values():
>    283          if fstype.sniff_magic(file, offset):
>    284              fs = fstype.open_fs(file, offset)
>    285              break
>    286
>    287      if fs is None:
>    288          raise RuntimeError, "Unable to open filesystem"
>    289
>    290      kernel = fs.open_file(img.kernel[1],).read()
>    291      (tfd, fn) = tempfile.mkstemp(prefix="vmlinuz.",
dir="/
> var/lib/xen")
>    292      os.write(tfd, kernel)
>    293      os.close(tfd)
>    294      sxp = "linux (kernel %s)" %(fn,)
>    295
>    296      if img.initrd:
>    297          initrd = fs.open_file(img.initrd[1],).read()
>    298          (tfd, fn) = tempfile.mkstemp(prefix="initrd.",  
> dir="/var/lib/xen")
>    299          os.write(tfd, initrd)
>    300          os.close(tfd)
>    301          sxp += "(ramdisk %s)" %(fn,)
>    302      else:
>    303          initrd = None
>    304      sxp += "(args ''%s'')" %(img.args,)
>    305
>    306      sys.stdout.flush()
>    307      os.write(fd, sxp)
>    308
> [root@localhost ~]#

>> Unfortunately, I tried same stuff with 512MB mem but failed.
>> What I suspect now is, a disk image would have to be formatted.
>> xenguest-install.py just makes plain empty sparse file, but is this
>> enough? I suspect it should be formatted and partitioned and so on.
>
> No. The empty file is enough, because xenguest-install.py doesn''t
use
> pygrub for domU installation boot. You need prepare your virtual disk
> by anaconda which is started by xenguest-install.py. The pygrub
> script is possible use for already __successfully__ installed domU
> only.
>

Ummm... You mean I have to start anaconda/XenU stuff first,
but I dont have felt that I have succeeded to do so, although
I dont have understanded what xenguest-install.py does yet.


             --- Okajima, Jun. Tokyo, Japan.

>The installer will create the disk image and partition it.  If you do  
>not successfully run to the installer, "xm create -c xxxx" will
not
>work.  The script gives bad instructions when an error occurs.
>
>The Xen hypervisor takes 32MB of RAM (I think), so you have to make  
>sure that your dom0 gets less than 224MB.  I gave mine 192MB, with  
>the following in grub.conf:
>   kernel /xen.gz-2.6.15-1.2054_FC5 dom0_mem=196608
>
>
>In my opinion, the xenguest-install.py method isn''t very good at
this
>point.
>- The installer gives poor feedback when something goes wrong, like  
>when memory is too low.
>- pygrub crashes if your terminal emulation is a little weird
>- It creates a whole disk image instead of importing partitions.   
>This makes it difficult to access domU filesystems from dom0, but  
>this is often necessary for management.
>- It doesn''t provide an option for creating images on LVM volumes.
>
>
>
>An easier solution might be to use the FC5 image from jailtime.org.   
>If you use the images from their site, keep in mind that you have to  
>use a kernel from outside of the image; the kernel would have to be  
>readable from your dom0.
>
>If you use the stock FC5-xenU kernel, it does not have the xennet  
>driver compiled in.   You would have to install the kernel-xenU  
>package onto the domU disk image to provide the necessary driver  
>(assuming the disk image is mounted at /mnt/tmp):
>   yum --installroot=/mnt/tmp install kernel-xenU
>And add the following to the domU''s /etc/modprobe.conf:
>   alias eth0 xennet
>
>Or you could use a kernel from xensource that has the xennet driver  
>compiled into the kernel.  I don''t know why the developers decided
to
>take the driver out of the kernel.  Pretty much everyone would want  
>networking on their domU machines, and adding it in this way is just  
>more complicated.
>
>
>

Thank you for suggestive advise.
but, I have been still working on the issue.
I tried to use a ready made image from jailtime.org,
but still it does not work.
Why os.path.isfile(self.kernel) fails?


         --- Okajima, Jun. Tokyo, Japan.





-------------
[root@localhost okajima]# xm create -c fedora.fc5.xen.cfg
Using config file "fedora.fc5.xen.cfg".
Warning: The nics option is deprecated.  Please use an empty vif entry instead:

  vif = [ '''' ]

Error: Error creating domain: Kernel image does not exist:
/boot/vmlinuz-2.6.16-1.2080_FC5xenU
[root@localhost okajima]# ls -la /boot/vmlinuz-2.6.16-1.2080_FC5xenU
-rw-r--r-- 1 root root 1232494 Mar 28 18:38 /boot/vmlinuz-2.6.16-1.2080_FC5xenU
[root@localhost okajima]# md5sum /boot/vmlinuz-2.6.16-1.2080_FC5xenU
242c2a1cfeb924481ba57e983cbe801e  /boot/vmlinuz-2.6.16-1.2080_FC5xenU
[root@localhost okajima]# ls -l /usr/lib/xen/boot/
total 160
-rwxr-xr-x 1 root root 153797 Mar  9 14:20 hvmloader
[root@localhost okajima]# md5sum /usr/lib/xen/boot/hvmloader
2bbd4cf936aa8bde3ab6b6fe31362ff0  /usr/lib/xen/boot/hvmloader
[root@localhost okajima]# cat fedora.fc5.xen.cfg
kernel = "/boot/vmlinuz-2.6.16-1.2080_FC5xenU"
#kernel = "/boot/vmlinuz-2.6-xenU"
memory = 128
name = "fedora.fc5"
nics = 1
dhcp = "dhcp"
disk = [''file:/home/okajima/fedora.fc5.img,sda1,w'',
''file:/home/okajima/fedora.swap,sda2,w'']
root = "/dev/sda1 ro"
[root@localhost okajima]# grep -3 -ir "Kernel image does not exist"
/usr/lib/python2.4/site-packages/
*
Binary file /usr/lib/python2.4/site-packages/xen/xend/image.pyo matches
--
/usr/lib/python2.4/site-packages/xen/xend/image.py-        # Set params and call
buildDomain().
/usr/lib/python2.4/site-packages/xen/xend/image.py-
/usr/lib/python2.4/site-packages/xen/xend/image.py-        if not
os.path.isfile(self.kernel):
/usr/lib/python2.4/site-packages/xen/xend/image.py:            raise
VmError(''Kernel image does not
exist: %s'' % self.kernel)
/usr/lib/python2.4/site-packages/xen/xend/image.py-        if self.ramdisk and
not os.path.isfile
(self.ramdisk):
/usr/lib/python2.4/site-packages/xen/xend/image.py-            raise
VmError(''Kernel ramdisk does not
exist: %s'' % self.ramdisk)
/usr/lib/python2.4/site-packages/xen/xend/image.py-        if len(self.cmdline)
>= MAX_GUEST_CMDLINE:
Binary file /usr/lib/python2.4/site-packages/xen/xend/image.pyc matches
[root@localhost okajima]#

On Mon, 3 Apr 2006, Jun OKAJIMA wrote:
> What happened here is, in short, like this:
> 
>    # Check diskimage formatted as described in grub config
>    117      for fstype in grub.fsys.fstypes.values():
>    118          if fstype.sniff_magic(fn, offset):
>    # If not formatted properly, raise error.
>    138      else:
>    139          raise RuntimeError, "Unable to read filesystem"
I''m also seeing an error raised here, after a successful anaconda over
vnc
install, without any memory issues.  I''ve since mounted the partition
from
the image and it seems ok.

Perhaps something not quite right in the fs image detection code?



- James
-- 
James Morris
<jmorris@redhat.com>

Hi,

On Mon, 2006-04-03 at 20:16 +0900, Jun OKAJIMA wrote:
> > No. The empty file is enough, because xenguest-install.py
doesn''t use
> > pygrub for domU installation boot. You need prepare your virtual disk
> > by anaconda which is started by xenguest-install.py. The pygrub
> > script is possible use for already __successfully__ installed domU
> > only.
> Ummm... You mean I have to start anaconda/XenU stuff first,
Yes, but...
> but I dont have felt that I have succeeded to do so, although
> I dont have understanded what xenguest-install.py does yet.
You don''t have to start it yourself: xenguest-install.py will do it for
you automatically.

--Stephen

Hi,

On Sun, 2006-04-02 at 22:09 -0400, Winston Chang wrote:
> The installer will create the disk image and partition it.  If you do  
> not successfully run to the installer, "xm create -c xxxx" will
not
> work.  The script gives bad instructions when an error occurs.
> 
> The Xen hypervisor takes 32MB of RAM (I think), so you have to make  
> sure that your dom0 gets less than 224MB.  I gave mine 192MB, with  
> the following in grub.conf:
>    kernel /xen.gz-2.6.15-1.2054_FC5 dom0_mem=196608
Actually, you should be able to do that manually with 

        xm mem-set 0 192
        
to balloon the dom0 down to 192MB.  It won''t balloon automatically
below
256MB, but can do so if explicitly told to.
> In my opinion, the xenguest-install.py method isn''t very good at
this
> point.
> - The installer gives poor feedback when something goes wrong, like  
> when memory is too low.
Agreed, it could well be improved.
> - pygrub crashes if your terminal emulation is a little weird
> - It creates a whole disk image instead of importing partitions.   
> This makes it difficult to access domU filesystems from dom0, but  
> this is often necessary for management.
That''s just part of the goal of making the domU environment as similar
as possible to the dom0.  Having a single disk image passed in means you
can keep boot info, root fs and swap all in the same place in the host
OS.
> - It doesn''t provide an option for creating images on LVM volumes.
If you already have an LVM volume, you can simply specify that and it
just works (most of my test xen installs are to LVM.)
> Or you could use a kernel from xensource that has the xennet driver  
> compiled into the kernel.  I don''t know why the developers decided
to
> take the driver out of the kernel.  
The default networking with Xen uses netfront/back and bridging, so
that''s what we compile in.  Is there really a strong need for xennet
too?
> Pretty much everyone would want  
> networking on their domU machines, and adding it in this way is just  
> more complicated.
With FC-5 default networking, things just work (though it definitely can
interact badly with more complex networking setups); what exactly is
complicated and needs fixed?

--Stephen

Hi,

On Mon, 2006-04-03 at 12:18 -0400, James Morris wrote:
> On Mon, 3 Apr 2006, Jun OKAJIMA wrote:
> 
> > What happened here is, in short, like this:
> > 
> >    # Check diskimage formatted as described in grub config
> >    117      for fstype in grub.fsys.fstypes.values():
> >    118          if fstype.sniff_magic(fn, offset):
> >    # If not formatted properly, raise error.
> >    138      else:
> >    139          raise RuntimeError, "Unable to read
filesystem"
> 
> I''m also seeing an error raised here, after a successful anaconda
over vnc
> install, without any memory issues.  I''ve since mounted the
partition from
> the image and it seems ok.
> 
> Perhaps something not quite right in the fs image detection code?
Hmm --- did you create a manual partition layout, or did you just use
the default one provided by the installer?

If it''s a manual layout, then the problem may be that you (or the
installer) didn''t set the "active" flag on the partition
containing the
kernel images.  pygrub looks for that to know which partition to try
booting from.

--Stephen

On Mon, 3 Apr 2006, Stephen C. Tweedie wrote:
> Hmm --- did you create a manual partition layout, or did you just use
> the default one provided by the installer?
It was a standard layout, although I did select minimal packages 
(shouldn''t affect it, I hope).
> If it''s a manual layout, then the problem may be that you (or the
> installer) didn''t set the "active" flag on the partition
containing the
> kernel images.  pygrub looks for that to know which partition to try
> booting from.
# fdisk -l xt4-disk
last_lba(): I don''t know how to handle files with mode 81ed
You must set cylinders.
You can do this from the extra functions menu.

Disk xt4-disk: 0 MB, 0 bytes
255 heads, 63 sectors/track, 0 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes

   Device Boot      Start         End      Blocks   Id  System
xt4-disk1   *           1          13      104391   83  Linux
xt4-disk2              14         261     1992060   8e  Linux LVM

# losetup -o 32256 /dev/loop0 xt4-disk
# fsck.ext3 -v /dev/loop0
e2fsck 1.38 (30-Jun-2005)
/boot: clean, 32/26104 files, 12598/104388 blocks



Look ok?


- James
-- 
James Morris
<jmorris@redhat.com>

Hi,

On Mon, 2006-04-03 at 15:57 -0400, James Morris wrote:
> # fdisk -l xt4-disk
> last_lba(): I don''t know how to handle files with mode 81ed
> You must set cylinders.
> You can do this from the extra functions menu.
> 
> Disk xt4-disk: 0 MB, 0 bytes
> 255 heads, 63 sectors/track, 0 cylinders
> Units = cylinders of 16065 * 512 = 8225280 bytes
That''s all normal;
> 
>    Device Boot      Start         End      Blocks   Id  System
> xt4-disk1   *           1          13      104391   83  Linux
> xt4-disk2              14         261     1992060   8e  Linux LVM
looks fine.  Hmm.
> # losetup -o 32256 /dev/loop0 xt4-disk
> # fsck.ext3 -v /dev/loop0
> e2fsck 1.38 (30-Jun-2005)
> /boot: clean, 32/26104 files, 12598/104388 blocks
Does it seem to contain the correct boot files?

--Stephen

On Mon, 3 Apr 2006, Stephen C. Tweedie wrote:
> > # losetup -o 32256 /dev/loop0 xt4-disk
> > # fsck.ext3 -v /dev/loop0
> > e2fsck 1.38 (30-Jun-2005)
> > /boot: clean, 32/26104 files, 12598/104388 blocks
> 
> Does it seem to contain the correct boot files?
Yes, as far as I can tell:

# ls -l
total 3456
-rw-r--r-- 1 root root   28213 Mar 14 16:58 config-2.6.15-1.2054_FC5xenU
drwxr-xr-x 2 root root    1024 Mar 30 19:32 grub
-rw-r--r-- 1 root root 1601303 Mar 30 19:31 initrd-2.6.15-1.2054_FC5xenU.img
drwx------ 2 root root   12288 Mar 30 19:26 lost+found
-rw-r--r-- 1 root root  637784 Mar 14 16:58 System.map-2.6.15-1.2054_FC5xenU
-rw-r--r-- 1 root root 1232540 Mar 14 16:58 vmlinuz-2.6.15-1.2054_FC5xenU

# ls -l grub
total 203
-rw-r--r-- 1 root root     64 Mar 30 19:32 device.map
-rw-r--r-- 1 root root   7424 Mar 30 19:32 e2fs_stage1_5
-rw-r--r-- 1 root root   7328 Mar 30 19:32 fat_stage1_5
-rw-r--r-- 1 root root   6592 Mar 30 19:32 ffs_stage1_5
-rw------- 1 root root    621 Mar 30 19:32 grub.conf
-rw-r--r-- 1 root root   6592 Mar 30 19:32 iso9660_stage1_5
-rw-r--r-- 1 root root   8032 Mar 30 19:32 jfs_stage1_5
lrwxrwxrwx 1 root root     11 Mar 30 19:32 menu.lst -> ./grub.conf
-rw-r--r-- 1 root root   6752 Mar 30 19:32 minix_stage1_5
-rw-r--r-- 1 root root   9056 Mar 30 19:32 reiserfs_stage1_5
-rw-r--r-- 1 root root   7028 Feb 28 23:41 splash.xpm.gz
-rw-r--r-- 1 root root    512 Mar 30 19:32 stage1
-rw-r--r-- 1 root root 102332 Mar 30 19:32 stage2
-rw-r--r-- 1 root root   6944 Mar 30 19:32 ufs2_stage1_5
-rw-r--r-- 1 root root   6144 Mar 30 19:32 vstafs_stage1_5
-rw-r--r-- 1 root root   8768 Mar 30 19:32 xfs_stage1_5


-- 
James Morris
<jmorris@redhat.com>

>> - It creates a whole disk image instead of importing partitions.
>> This makes it difficult to access domU filesystems from dom0, but
>> this is often necessary for management.
>
> That''s just part of the goal of making the domU environment as
similar
> as possible to the dom0.  Having a single disk image passed in  
> means you
> can keep boot info, root fs and swap all in the same place in the host
> OS.
I understand the motivation here, but it does make it much harder to  
manipulate the domU filesystem from dom0.  I personally couldn''t  
figure out how to mount a domU partition when using the whole-disk  
method.  I imagine that it''s even more complicated when the domU has  
its own LVM volumes.

This issue aside, I do like the idea of having a single disk image  
for each domU, though, since it keeps things simple.  It''s nice to  
have the kernel, swap partition, and main partition (and whatever  
else) all in one place.  But it seems very inelegant to use the  
ancient DOS partitioning scheme and then put LVM on top of it.

Things would be simpler if a domU could start out with LVM right on  
top of the disk image, like if you ran ''pvcreate /dev/sda'' in
the
domU.  Then you wouldn''t need a separate /boot partition; it could be  
just another LVM volume.  I know that this is problematic for "real"  
installations of Linux, because of the BIOS <1024 cylinder limitation  
(is that still an issue?), and because grub doesn''t understand LVM.   
But the first limitation simply doesn''t exist with a domU under Xen  
(I hope), and the second can be worked around -- I imagine that  
making pygrub understand LVM would be easier than making grub  
understand LVM, since there aren''t the same memory/space limitations.

In short, why emulate all crufty old mechanics of booting and  
partitioning a PC when Xen provides a way to (finally) avoid it?  You  
would lose the ability to image a domU disk image right onto a real  
disk and then boot a real machine with that disk, but I think that''s  
the only thing, and it''s pretty minor.



>> - It doesn''t provide an option for creating images on LVM
volumes.
>
> If you already have an LVM volume, you can simply specify that and it
> just works (most of my test xen installs are to LVM.)
That''s good to know... The script doesn''t make it clear that
it has
that capability.  If I remember right, I tried that, and the config  
file said file:/dev/VolGroup00/fc5_root, or something like that  
instead of phy:.., which seems like an error.  (I was still having  
other Xen problems so I never got it working this way.)  The LVM  
volume already has a size, so what happens when you specify a disk  
image size to the installer script?

>> Or you could use a kernel from xensource that has the xennet driver
>> compiled into the kernel.  I don''t know why the developers
decided to
>> take the driver out of the kernel.
>
> The default networking with Xen uses netfront/back and bridging, so
> that''s what we compile in.  Is there really a strong need for
xennet
> too?
I''m not an expert on Xen network drivers... All I know is that when I  
booted the jailtime.org FC5 image (and a Centos 4.2 image that I  
created myself) using the stock FC5 xenU kernel, that it would say  
there was no eth0 on boot.  I managed to get eth0 to show up by  
loading the xennet kernel module, as recommended by someone on the  
xensource xen-users list.  If there''s way to get a domU''s eth0
to
show up using the stock XenU kernel and without loading a module,  
that would be great.  How would I do this?

--Winston