[Xen-devel] xm save and restore problem

[Xuehai Zhang <hai@xxxxxxxxxxxxxxx>]

Hi,

I tried to do some experiments to test if the interrupted network traffic
(for example, unfinished scp or http session) gets replayed between two
vms (hosted in different machines) when I tried to freeze them first
(using "xm save") and then resume them after a while (using "xm restore").

I ran the tests multiple times and found the traffic usually gets replayed
after both "frozen" vms get restored (that is what i am excited to see).
However, in many rounds I also met vm kernel crash problem, especially I
tried to restore them from the "frozen" disk files. I attached an example
crash image at the end of the email, when I tried to resume both of the
client and server vms after I froze them when there was an unfinished scp
session between them (client just transfered 24% of the remote file).
Since some previous threads in the mailing list show some other people
also get such restore crash problems
(http://sourceforge.net/mailarchive/message.php?msg_id=10113264), I am
wondering if the crashes I met just solely come from Xen itself or also
relate to the network traffic I injected.

Thanks for any help.

Xuehai

***********
Client VM:
***********

hamachi:/home/hai/vm# xm console haivm1
************ REMOTE CONSOLE: CTRL-] TO QUIT ********
Unable to handle kernel paging request at virtual address 26991f7f
 printing eip:
c030f000
*pde = ma 00000000 pa 55555000
 [<c010a008>] __do_suspend+0x19f/0x1e0

 [<c012cc25>] worker_thread+0x22b/0x32f

 [<c010a110>] __shutdown_handler+0x0/0x48

 [<c0119882>] default_wake_function+0x0/0x12

 [<c0119882>] default_wake_function+0x0/0x12

 [<c012c9fa>] worker_thread+0x0/0x32f

 [<c0130d49>] kthread+0xa5/0xab

 [<c0130ca4>] kthread+0x0/0xab

 [<c010f705>] kernel_thread_helper+0x5/0xb

Oops: 0002 [#1]
PREEMPT
Modules linked in:
CPU:    0
EIP:    0061:[<c030f000>]    Not tainted VLI
EFLAGS: 00010297   (2.6.9-xenU)
EIP is at free_all_bootmem_core+0x1c5/0x274
eax: c02c5bd4   ebx: 0000c000   ecx: fbffc000   edx: 00000001
esi: 00000010   edi: c0102000   ebp: 00000000   esp: c10ebf04
ds: 0069   es: 0069   ss: 0069
Process events/0 (pid: 3, threadinfo=c10ea000 task=c10d9020)
Stack: c010ee65 00000000 c010a008 fbffc000 000001df 00000063 93de810d
00000180
       c02c2000 c3f79000 c02c4b80 00000000 c10ea000 00000000 c012cc25
00000000
       c10ebf74 00000000 c10ad278 c10ea000 c10ea000 c10ea000 c010a110
c10ea000
Call Trace:
 [<c010ee65>] time_resume+0x12/0x51

 [<c010a008>] __do_suspend+0x19f/0x1e0

 [<c012cc25>] worker_thread+0x22b/0x32f

 [<c010a110>] __shutdown_handler+0x0/0x48

 [<c0119882>] default_wake_function+0x0/0x12

 [<c0119882>] default_wake_function+0x0/0x12

 [<c012c9fa>] worker_thread+0x0/0x32f

 [<c0130d49>] kthread+0xa5/0xab

 [<c0130ca4>] kthread+0x0/0xab

 [<c010f705>] kernel_thread_helper+0x5/0xb

Code: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 <00> b0 ab c3 6c 66 12 c1 bc 00 00 00 bc f0 30 c0 29 00 00 00 45
/home/hai/ftp.img                              24%   50MB   0.0KB/s -
stalled -



***********
SERVER VM:
***********

gardenia2:/home/hai/vm# xm console haivm1
************ REMOTE CONSOLE: CTRL-] TO QUIT ********
------------[ cut here ]------------
kernel BUG at arch/xen/i386/kernel/time.c:685!
invalid operand: 0000 [#1]
PREEMPT
Modules linked in:
CPU:    0
EIP:    0061:[<c010e8e9>]    Not tainted VLI
EFLAGS: 00010286   (2.6.9-xenU)
EIP is at time_resume+0x16/0x51
eax: c02c57d4   ebx: 0000c000   ecx: fbffc000   edx: 00000001
esi: 00000010   edi: c0102000   ebp: 00000000   esp: c10ec544
ds: 0069   es: 0069   ss: 0069
Unable to handle kernel paging request at virtual address ad84006f
 printing eip:
c01164e2
*pde = ma 00000000 pa 55555000
 [<c0224fd1>] __xencons_tx_flush+0x21b/0x244

 [<c010dbf8>] page_fault+0x38/0x40

 [<c01164e2>] do_page_fault+0x8d/0x67a

 [<c011d210>] release_console_sem+0x73/0x11d

 [<c011d0d0>] vprintk+0x121/0x1aa

 [<c011cfab>] printk+0x17/0x1b

 [<c013545f>] __print_symbol+0x86/0xd2

 [<c010dbf8>] page_fault+0x38/0x40

 [<c01164e2>] do_page_fault+0x8d/0x67a

 [<c0224b09>] kcons_write+0x71/0xcd

 [<c011cd28>] __call_console_drivers+0x5b/0x5d

 [<c011ce1a>] call_console_drivers+0x69/0x11f

 [<c010dbf8>] page_fault+0x38/0x40

 [<c010a7f1>] show_registers+0x100/0x1b8

 [<c010af14>] do_invalid_op+0x0/0x10e

 [<c010aa4a>] die+0x103/0x1aa

 [<c010e8e9>] time_resume+0x16/0x51

 [<c0118606>] fixup_exception+0x16/0x34

 [<c010b020>] do_invalid_op+0x10c/0x10e

 [<c010e8e9>] time_resume+0x16/0x51

 [<c010da11>] error_code+0x2d/0x38

 [<c010e8e9>] time_resume+0x16/0x51

 [<c024a115>] ip_push_pending_frames+0x2b5/0x411

 [<c02405ff>] netlink_ack+0x116/0x1bb

 [<c02005c7>] cap_task_post_setuid+0xa2/0x11f

 [<c02005c7>] cap_task_post_setuid+0xa2/0x11f

 [<c02005c7>] cap_task_post_setuid+0xa2/0x11f

 [<c02005c7>] cap_task_post_setuid+0xa2/0x11f

 [<c02005c7>] cap_task_post_setuid+0xa2/0x11f

 [<c02005c7>] cap_task_post_setuid+0xa2/0x11f

 [<c020053b>] cap_task_post_setuid+0x16/0x11f

 [<c020a308>] read_zero+0x1af/0x204

 [<c02005c7>] cap_task_post_setuid+0xa2/0x11f

 [<c020a308>] read_zero+0x1af/0x204

 [<c02005c7>] cap_task_post_setuid+0xa2/0x11f

 [<c02005c7>] cap_task_post_setuid+0xa2/0x11f

 [<c02005c7>] cap_task_post_setuid+0xa2/0x11f

 [<c024a110>] ip_push_pending_frames+0x2b0/0x411

 [<c02005c7>] cap_task_post_setuid+0xa2/0x11f

 [<c02005c7>] cap_task_post_setuid+0xa2/0x11f

 [<c02005c7>] cap_task_post_setuid+0xa2/0x11f

 [<c0310000>] biovec_init_pools+0x87/0x108

 =======================








-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now. 
http://productguide.itmanagersjournal.com/
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxxx
https://lists.sourceforge.net/lists/listinfo/xen-devel