WARNING - OLD ARCHIVES

This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
   
 
 
Xen 
 
Home Products Support Community News
 
   
 

xen-users

[Xen-users] Xen 4.0 - prerequisites for succesfull live migration?

To: xen-users@xxxxxxxxxxxxxxxxxxx
Subject: [Xen-users] Xen 4.0 - prerequisites for succesfull live migration?
From: Danilo Godec <danilo.godec@xxxxxxxxx>
Date: Wed, 06 Jul 2011 15:19:03 +0200
Delivery-date: Wed, 06 Jul 2011 06:20:14 -0700
Envelope-to: www-data@xxxxxxxxxxxxxxxxxxx
List-help: <mailto:xen-users-request@lists.xensource.com?subject=help>
List-id: Xen user discussion <xen-users.lists.xensource.com>
List-post: <mailto:xen-users@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/mailman/listinfo/xen-users>, <mailto:xen-users-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/mailman/listinfo/xen-users>, <mailto:xen-users-request@lists.xensource.com?subject=unsubscribe>
Sender: xen-users-bounces@xxxxxxxxxxxxxxxxxxx
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:5.0) Gecko/20110624 Thunderbird/5.0
Hi,

I have three Xen hosts running Xen 4.0.2 (OpenSuSE 11.4 based). I also have one 'NFS' server with an NFS export holding VM images and configuration files.

Each host has a dedicated LAN link directly to the NFS server. I have another separated NFS export for VM locking (but I had the same issues before).

When I attempt to live migrate a VM, it *looks'*like it's OK, but I end up with a non-responsive VM on the target host.

There are no fail or error lines in the source host xend.log file.

The xend.log file on the target host these 'fail' or 'error' lines:

[2011-07-06 14:57:11 26204] INFO (XendCheckpoint:485) read_exact_timed failed (select returned 0) [2011-07-06 14:57:11 26204] INFO (XendCheckpoint:485) ERROR Internal error: Error when reading batch size [2011-07-06 14:57:11 26204] INFO (XendCheckpoint:485) ERROR Internal error: error when buffering batch, finishing

I can also see that the VM lock file is changed and shows the VM is running on target host as if everything is OK.

Yet the VM is unresponsive - xm console, reboot or shutdown don't work - the only thing I can do is 'destroy' and re-create the VM.

My typical VM config looks like this:

name="ab1"
description="Apache Bench Server 1"
uuid="5383bef7-0794-4372-8c3d-e211e65e22d9"
memory=1024                    # It should start with 1G
maxmem=16384                # but we want to be able to add memory if needed
vcpu_avail=2                        # this VM should start with 2 VCPUs
vcpus=8 # but 8 cores are available and we want to be able to add some if needed
cpus="1-8"                           # core 0 is 'reserved' for Domain 0
on_poweroff="destroy"
on_reboot="restart"
on_crash="destroy"
localtime=0
builder="linux"
bootloader="/usr/lib/xen/boot/domUloader.py"
bootargs="--entry=xvda1:/boot/vmlinuz-xen,/boot/initrd-xen"
extra=" 3 mem=4096m xencons=tty apparmor=0"
disk=[ 'tap:aio:/var/lib/xen/images/ab1/ab1.raw,xvda,w', ]
vif=[ 'mac=00:16:3E:02:00:43,bridge=br0' ]
vfb=['type=vnc,vncunused=1']


What am I missing? Anyone using a similar NFS approach and has live migration working OK?


  Danilo


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