Re: [Xen-users] XEN 3.4.2 /proc/xen/balloon doesn't exist and grub2 trouble

[Gabor Szilagyi <szilagyi@xxxxxxxxxxxx>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Pasi Kärkkäinen wrote:
> On Tue, Jan 26, 2010 at 11:49:43AM -0500, Gabor Szilagyi wrote:
>>>> So I am trying to make it myself.
>>>>
>>>> I have installed xen 3.4.2 from source, did make world and installed.
>>>>
>>>> Got the kernel source from from the git repository with the xen patches
>>>> following the xen wiki.... compiled dom0 kernel just fine.
>>>>
>>> So you're using pv_ops dom0 kernel?
>> Yes I am pretty sure.. I can boot it on the bare hardware not problem
>>
> 
> Ok.
> 
>>>> I can boot xen, with grup (old something went odd with grub2 for me) and
>>>> then it boots my dom0 kernel fine.
>>>>
>>> For GRUB2 see:
>>> http://wiki.xensource.com/xenwiki/XenCommonProblems
>> Well, I have looked at this and went a head re-install Debian Squeeze
>> with grub2 ...
>>
>> Now, I can not even boot xen-3.4.2.gz which I built again from source....
>>
> 
> Uhm.. what error do you get? 

I can copy from the console only the last few all line start with (XEN)
so I will not copy those

Panic on CPU 0:
Cannot access memory beyond end of bootstrap direct-map area
******************

Reboot in five second
Early fatal page fault at e008:ff146c9 (cr2=fffde020,ec=0000
Stack dump ... and a whole bunch of numbers


I see this exact some other post

> 
> Can you set up a serial console to log the full boot process?
I could if I had the right cable.
> 
> See:
> http://wiki.xensource.com/xenwiki/XenSerialConsole

I would try if I can find a null-modem cable db9-db9 !!! I may have that
somewhere .... but have not used such things for ages... the one came
with cisco terminal thing.... doesn't connect

> 
> 
>> I tried the debian package (squeeze has 3.4.2 as well) I can not boot
>> even the hypervisor (and can not downgrade grub2 to the old grub)...
>>
>> I can see that booting xen with the new grub gives a lot of people grief.
>>
> 
> Yeah, grub2 has been pretty problematic..
> 
>> I tried all the suggestions like dummy=dummy, double the file name on
>> booth multiboot and module lines I also compiled the command line into
>> the dom0 kernel (the 2.6.31.6 from the Jeremy tree). I know that command
>> line is OK since I can boot it again on the bare hardware no problem....
>>
>> I am not sure what  I am doing wrong
>>
> 
> We need more information about what exactly fails.. if you cannot set up
> a serial console, you could always record a video of the boot process..

It is pretty fast but will try ....

> 
>>>> Here I got stuck trying to create a hvm instance for installation of OS.
>>>>
>>>> root@rasnew:/etc/xen# xm create -c /etc/xen/test1.cfg
>>>> Using config file "/etc/xen/test1.cfg".
>>>> Error: [Errno 2] No such file or directory: '/proc/xen/balloon'
>>>>
>>> Did you mount /proc/xen ? You need to have xenfs mounted.
>> No, /proc/xen was there (see my earlier post) except the
>> /proc/xen/ballon file was not there (five others were there)
>>
> 
> Ok.
> 
>>> See:
>>> http://wiki.xensource.com/xenwiki/XenParavirtOps
>> I can probably recite this already nothing there solved my problem. I
>> really like to know if anyone can actually boot the hypervisor built
>> from source with kernel on Debian Squeeze with grub2 .... the kernel
>> included in the source 2.6.18 doesn't support my ethernet card and
>> pretty old even udev is too old for some stuff (OK I can run everything
>> in domU if I ever can create one :-))
>>
>> I am not new to linux (started 1997) this maybe the first project I will
>> have to give up on ... virtbox and vmware just worked but I would prefer
>>
> 
> There are many people who have been using Xen with GRUB2.. so it's possible.

Well, most example only has grub-legacy example I am not sure what to
try to eliminate the possibility of grub2.

If anyone made squeeze with grub2 boot xen-3.4.2 with the jeremy's
kernel work on x86_32 I would really love to hear about that...

Is it possible to know if the xen-3.4.2.gz file is OK  ??? there is no
config option during compile...

The dom0 kernel definitely boots on the bare hardware here some excerpts
from the dmesg with the dom0 capable (I hope)

[    0.000000] Allocating PCI resources starting at e0000000 (gap:
e0000000:18000000)
[    0.000000] NR_CPUS:8 nr_cpumask_bits:8 nr_cpu_ids:2 nr_node_ids:1
[    0.000000] PERCPU: Embedded 14 pages at c4fd0000, static data 35292
bytes
[    0.000000] Built 1 zonelists in Zone order, mobility grouping on.
Total pages: 1908572
[    0.000000] Kernel command line: root=/dev/md1 ro
BOOT_IMAGE=/boot/vmlinuz-2.6.31.6-xen-2 root=/dev/md1 ro quiet
[    0.000000] PID hash table entries: 4096 (order: 12, 16384 bytes)
[    0.000000] Dentry cache hash table entries: 131072 (order: 7, 524288
bytes)
[    0.000000] Inode-cache hash table entries: 65536 (order: 6, 262144
bytes)
[    0.000000] Enabling fast FPU save and restore... done.
[    0.000000] Enabling unmasked SIMD FPU exception support... done.
[    0.000000] Initializing CPU#0
[    0.000000] xsave/xrstor: enabled xstate_bv 0x3, cntxt size 0x240
[    0.000000] Initializing HighMem for node 0 (000379fe:001fe000)
[    0.000000] Memory: 7618936k/8355840k available (2486k kernel code,
79724k reserved, 1171k data, 376k init, 6788616k highmem)
[    0.000000] virtual kernel memory layout:
[    0.000000]     fixmap  : 0xfff1e000 - 0xfffff000   ( 900 kB)
[    0.000000]     pkmap   : 0xffa00000 - 0xffc00000   (2048 kB)
[    0.000000]     vmalloc : 0xf81fe000 - 0xff9fe000   ( 120 MB)
[    0.000000]     lowmem  : 0xc0000000 - 0xf79fe000   ( 889 MB)
[    0.000000]       .init : 0xc0493000 - 0xc04f1000   ( 376 kB)
[    0.000000]       .data : 0xc036d99d - 0xc04927d0   (1171 kB)
[    0.000000]       .text : 0xc0100000 - 0xc036d99d   (2486 kB)
[    0.000000] Checking if this processor honours the WP bit even in
supervisor mode...Ok.
........ some left out...

[    0.067951] CPU0: Intel(R) Core(TM)2 Duo CPU     E8400  @ 3.00GHz
stepping 0a
[    0.068001] Booting processor 1 APIC 0x1 ip 0x6000
[    0.004000] Initializing CPU#1
[    0.004000] Calibrating delay using timer specific routine.. 5984.96
BogoMIPS (lpj=11969930)
[    0.004000] CPU: L1 I cache: 32K, L1 D cache: 32K
[    0.004000] CPU: L2 cache: 6144K
[    0.004000] CPU: Physical Processor ID: 0
[    0.004000] CPU: Processor Core ID: 1
[    0.004000] mce: CPU supports 6 MCE banks
[    0.004000] CPU1: Thermal monitoring enabled (TM2)
[    0.004000] x86 PAT enabled: cpu 1, old 0x7040600070406, new
0x7010600070106
[    0.153423] CPU1: Intel(R) Core(TM)2 Duo CPU     E8400  @ 3.00GHz
stepping 0a
[    0.153432] checking TSC synchronization [CPU#0 -> CPU#1]: passed.
[    0.156024] Brought up 2 CPUs
[    0.156026] Total of 2 processors activated (11969.17 BogoMIPS).
[    0.156077] CPU0 attaching sched-domain:
[    0.156079]  domain 0: span 0-1 level MC
[    0.156081]   groups: 0 1
[    0.156084] CPU1 attaching sched-domain:
[    0.156085]  domain 0: span 0-1 level MC
[    0.156087]   groups: 1 0
[    0.156131] Booting paravirtualized kernel on bare hardware
[    0.156131] regulator: core version 0.5

Any suggestion ... to further investigate this??

Gabor



> 
> -- Pasi
> 

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iEYEARECAAYFAktfVTkACgkQcYMbOTV6LVrg9ACgvFFQIrToEDmwMVyD2FaxEvPc
e2AAnjqQhcB8JV+3EoUk9ApqKCdhZRIU
=kzXG
-----END PGP SIGNATURE-----

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