Re: kernel BUG at arch/x86/xen/mmu.c:1860!

[Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>]

[-- Attachment #1: Type: text/plain, Size: 2582 bytes --]

> Yes.  All are set in my kernel config and I believed Konrad has my kernel
> config.  I didn't use nomodeset as my servers don't have those graphic
> cards... ...

I am having a hard-time reproducing this. I made six LV's:
[root@tst011 ~]# lvs
  LV                   VG         Attr   LSize  Origin      Snap%  Move Log Copy%  Convert
  LogVol00             VolGroup00 -wi-ao 17.56G                                           
  LogVol01             VolGroup00 -wi-ao  5.81G                                           
  data-1.ext3          XenGroup   -wi-ao 10.00G                                           
  data-2.ext3          XenGroup   owi-ao 10.00G                                           
  data-2.ext3-snapshot XenGroup   swi-a-  1.00G data-2.ext3   2.41                        
  data-3.ext3          XenGroup   -wi-a- 10.00G                                           
  scratch-1.ext3       XenGroup   -wi-ao 10.00G                                           
  scratch-2.ext3       XenGroup   -wi-ao 10.00G                                           
  scratch-3.ext3       XenGroup   -wi-a- 10.00G                                           
[root@tst011 ~]# 

Where each scratch-X/data-X was prepared with 'mkfs.ext3'

And ran two guests, where each guest configuration looks as so:

kernel="/mnt/lab/latest/vmlinuz"
ramdisk="/mnt/lab/latest/initramfs.cpio.gz"
extra="console=hvc0 debug"
memory=768
vcpus=4
on_crash="preserve"
#vif = [ 'mac=00:0f:4b:00:00:68, bridge=switch' ]
vfb = [ 'vnc=1, vnclisten=0.0.0.0,vncunused=1']
disk = [ 'phy:/dev/XenGroup/scratch-1.ext3,xvda,w', 'phy:/dev/XenGroup/data-1.ext3,xvdb,w']

(the other is using -2 obviously).

And each guest is running 'mount /dev/xvda /mnt-1;(cd /mnt-1;fio iometer-file-access-server)'
to produce I/Os on the xvdb/xvda disks.

As those guests are chugging along, I ran your script with:

root      6250  4438  0 10:12 pts/4    00:00:00 /bin/sh ./lvm-test.sh loop 100 1G 5

(I tried 100 1G 0)

and so far it is running.... how long should I wait until I hit this problem?

This is on CentOS 5.5 and also on Fedora Core 13. Dom0 and DomU are all x86_64.

Dom0 is:

commit 892d2f052e979cf1916647c752b94cf62ec1c6dc
Merge: 35e2e28... 376faec...
Author: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Date:   Fri Feb 11 13:31:03 2011 -0800

    Merge commit 'v2.6.32.28' into xen/next-2.6.32

(plus one patch I just posted - but that is to fix the serial console, so
it is not relevant to this problem).

DomU is 2.6.38 kernel, but I can swap over to the same as Dom0..

Attached is the script I am using.

[-- Attachment #2: iometer-file-access-server --]
[-- Type: text/plain, Size: 439 bytes --]

# This job file tries to mimic the Intel IOMeter File Server Access Pattern
[global]
description=Emulation of Intel IOmeter File Server Access Pattern

[iometer]
bssplit=512/10:1k/5:2k/5:4k/60:8k/2:16k/4:32k/4:64k/10
rw=randrw
rwmixread=80
direct=1
size=4g
ioengine=libaio
# IOMeter defines the server loads as the following:
# iodepth=1	Linear
# iodepth=4	Very Light
# iodepth=8	Light
# iodepth=64	Moderate
# iodepth=256	Heavy
iodepth=64

[-- Attachment #3: lvm-test.sh --]
[-- Type: application/x-sh, Size: 3237 bytes --]

[-- Attachment #4: Type: text/plain, Size: 138 bytes --]

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