[Xen-users] Attempt to allocate order 5 skbuff. Increase MAX_SKBUFF_ORDER.

[David Young <david@xxxxxxxxxxxxxxx>]

Hi folks,

We have several CentOS 5.2 Xen Dom0s which have recently started misbehaving... xenconsoled mysteriously dies, and when I restart it, I see a bunch of these errors on the console of the DomU:

Sample:

Attempt to allocate order 5 skbuff. Increase MAX_SKBUFF_ORDER.

Attempt to allocate order 5 skbuff. Increase MAX_SKBUFF_ORDER.

Attempt to allocate order 5 skbuff. Increase MAX_SKBUFF_ORDER.

BUG: soft lockup - CPU#0 stuck for 209s! [swapper:0]

Pid: 0, comm: swapper

EIP: 0061:[<c0401227>] CPU: 0

EIP is at 0xc0401227

EFLAGS: 00000246 Not tainted (2.6.18-92.1.10.el5xen #1)

EAX: 00030001 EBX: 00000000 ECX: 00000000 EDX: f5416000

ESI: 00000220 EDI: c0772f22 EBP: 00000042 DS: 007b ES: 007b

CR0: 8005003b CR2: 08d6a7dc CR3: 00895000 CR4: 00000660

[<c054940c>] force_evtchn_callback+0xa/0xc

[<c0421f70>] vprintk+0x2b1/0x2bb

[<ee183cb1>] __ip_ct_refresh_acct+0xf6/0x129 [ip_conntrack]

[<c06094b8>] _spin_lock_irqsave+0x8/0x28

My first thought is that it's a recent Xen / kernel-xen update, since this is fairly recent behaviour. Any ideas?

----------

My setup:

Kernel:

Linux 2.6.18-92.1.10.el5xen 

xm info:

#1 SMP Tue Aug 5 08:46:32 EDT 2008 i686 i686 i386 GNU/Linux

release                : 2.6.18-92.1.10.el5xen

version                : #1 SMP Tue Aug 5 08:46:32 EDT 2008

machine                : i686

nr_cpus                : 2

nr_nodes               : 1

sockets_per_node       : 2

cores_per_socket       : 1

threads_per_core       : 1

cpu_mhz                : 930

hw_caps                : 0383fbff:00000000:00000000:00000040

total_memory           : 3071

free_memory            : 0

node_to_cpu            : node0:0-1

xen_major              : 3

xen_minor              : 1

xen_extra              : .2-92.1.10.el5

xen_caps               : xen-3.0-x86_32p 

xen_pagesize           : 4096

platform_params        : virt_start=0xf5800000

xen_changeset          : unavailable

cc_compiler            : gcc version 4.1.2 20071124 (Red Hat 4.1.2-42)

cc_compile_by          : mockbuild

cc_compile_domain      : centos.org

cc_compile_date        : Tue Aug  5 07:34:25 EDT 2008

xend_config_format     : 2

And RPM version:

# rpm -q xen

xen-3.0.3-64.el5_2.1

David Young systems engineer prophecy networks		industrial strength network & security solutions
email skype office mobile	david@xxxxxxxxxxxxxxx davidyoungnz +64 (6) 824-3904 +64 (21) 966-716

On 4/09/2008, at 10:48 AM, John Hannfield wrote:

Hello,
I have Ubuntu installed as a domU on the debian dom0, and connect via VNC
in dom0. This is great for doing installs, and seeing the console
output. However
the VNC session doesn't seem to keep the active connection during a domU reboot.
Has anyone got this working?
If so, how?

--

John

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

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