Xen users - May 2008 - Xen breaks under high I/O load

Hi,

When performing high I/O operations on the guests, xen just breaks and
the guest dies.

 login: Unable to handle kernel NULL pointer dereference a
t virtual address 00000b54
 printing eip:
c01493b1
0c735000 -> *pde = 00000000:73639027
0b19c000 -> *pme = 00000000:71421067
0f62b000 -> *pte = 00000000:00000000
Oops: 0002 [#1]
SMP
Modules linked in: libafs(U) autofs4 sunrpc md5 ipv6 dm_mirror dm_mod xennet xen
blk ext3 jbd
CPU:    0
EIP:    0061:[<c01493b1>]    Tainted: PF     VLI
EFLAGS: 00010202   (2.6.9-67.0.4.EL.cernxenU)
EIP is at copy_page_range+0x491/0x52c
eax: 00000b50   ebx: 4bf99065   ecx: cd098300   edx: 4bf99045
esi: 00000000   edi: c1156180   ebp: 00010800   esp: cb199ec0
ds: 007b   es: 007b   ss: 0068
Process yum (pid: 2100, threadinfo=cb199000 task=ce250cd0)
Stack: 4ffc5067 00000000 00000b50 c9438b50 cc5bb250 cb19c250 00000001 0b504000
       0956a000 cc5ba000 cc735000 cd098344 cce5064c c93bf494 cd098300 c011a6b8
       cd098300 cd098080 cce5064c 00001f9a 00000000 cce50668 cce50670 cce50658
Call Trace:
 [<c011a6b8>] copy_mm+0x2d9/0x396
 [<c011b26a>] copy_process+0x6b5/0xb0b
 [<c011b7ad>] do_fork+0x8a/0x16b
 [<c012736e>] sigprocmask+0xb6/0xd0
 [<c012741f>] sys_rt_sigprocmask+0x97/0x14c
 [<c0105cff>] sys_clone+0x24/0x28
 [<c010734f>] syscall_call+0x7/0xb
Code: 89 f9 83 e2 df f6 c4 80 74 03 8b 4f 0c f0 ff 41 04 8b 4c 24 40 ff 81 80 00
 00 00 f6 47 10 01 74 06 ff 81 84 00 00 00 8b 44 24 08 <89> 70 04 89 10 f0
ff 47
 08 81 44 24 20 00 10 00 00 8b 54 24 1c
 <0>Fatal exception: panic in 5 seconds
Kernel panic - not syncing: Fatal exception

Using this xen version : https://rhn.redhat.com/errata/RHSA-2008-0194.html

Any ideas on why this happens?

Thank you,
Ricardo

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

Hrmm.  Not sure off the top of my head, no.  You could try reporting this to 
the xen-devel mailing list where it might get developer attention.

Try to use a more descriptive subject line such as "domU panics under High
I/O
load".

Have you tried reporting this to the RHEL developers?  (or if you''re
using a
RHEL derivative like Scientific Linux or CentOS, to their developers).  Have 
you searched to see if other people have had a similar problem anywhere.

Sorry not to be more specific.

Cheers,
Mark
> When performing high I/O operations on the guests, xen just breaks and
> the guest dies.
>
>  login: Unable to handle kernel NULL pointer dereference a
> t virtual address 00000b54
>  printing eip:
> c01493b1
> 0c735000 -> *pde = 00000000:73639027
> 0b19c000 -> *pme = 00000000:71421067
> 0f62b000 -> *pte = 00000000:00000000
> Oops: 0002 [#1]
> SMP
> Modules linked in: libafs(U) autofs4 sunrpc md5 ipv6 dm_mirror dm_mod
> xennet xen blk ext3 jbd
> CPU:    0
> EIP:    0061:[<c01493b1>]    Tainted: PF     VLI
> EFLAGS: 00010202   (2.6.9-67.0.4.EL.cernxenU)
> EIP is at copy_page_range+0x491/0x52c
> eax: 00000b50   ebx: 4bf99065   ecx: cd098300   edx: 4bf99045
> esi: 00000000   edi: c1156180   ebp: 00010800   esp: cb199ec0
> ds: 007b   es: 007b   ss: 0068
> Process yum (pid: 2100, threadinfo=cb199000 task=ce250cd0)
> Stack: 4ffc5067 00000000 00000b50 c9438b50 cc5bb250 cb19c250 00000001
> 0b504000 0956a000 cc5ba000 cc735000 cd098344 cce5064c c93bf494 cd098300
> c011a6b8 cd098300 cd098080 cce5064c 00001f9a 00000000 cce50668 cce50670
> cce50658 Call Trace:
>  [<c011a6b8>] copy_mm+0x2d9/0x396
>  [<c011b26a>] copy_process+0x6b5/0xb0b
>  [<c011b7ad>] do_fork+0x8a/0x16b
>  [<c012736e>] sigprocmask+0xb6/0xd0
>  [<c012741f>] sys_rt_sigprocmask+0x97/0x14c
>  [<c0105cff>] sys_clone+0x24/0x28
>  [<c010734f>] syscall_call+0x7/0xb
> Code: 89 f9 83 e2 df f6 c4 80 74 03 8b 4f 0c f0 ff 41 04 8b 4c 24 40 ff 81
> 80 00 00 00 f6 47 10 01 74 06 ff 81 84 00 00 00 8b 44 24 08 <89> 70
04 89
> 10 f0 ff 47 08 81 44 24 20 00 10 00 00 8b 54 24 1c
>  <0>Fatal exception: panic in 5 seconds
> Kernel panic - not syncing: Fatal exception
>
> Using this xen version : https://rhn.redhat.com/errata/RHSA-2008-0194.html
>
> Any ideas on why this happens?
>
> Thank you,
> Ricardo
>
> _______________________________________________
> Xen-users mailing list
> Xen-users@lists.xensource.com
> http://lists.xensource.com/xen-users


-- 
Push Me Pull You - Distributed SCM tool (http://www.cl.cam.ac.uk/~maw48/pmpu/)

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