Can you get the template of the VM after it is deployed? Do a "onevm show <vmid>" or click in the running VM and get the template from the template tab. The image template will be also useful.<div><br></div>
<div><br><div><div class="gmail_quote">On Wed, Mar 27, 2013 at 4:53 PM, Mark Farragher <span dir="ltr"><<a href="mailto:mark@firstfocus.eu" target="_blank">mark@firstfocus.eu</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div style="word-wrap:break-word"><div><div style="word-wrap:break-word">Hi everyone, <div><br></div><div>I am trying to start up my first virtual machine on my OpenNebula configuration but I keep getting errors. As soon as the VM enters the RUNNING state I see start-up errors inside the VM that the boot device cannot be found. </div>
<div><br></div><div>Here's my setup: </div><div><br></div><div>I have 2 servers: an OpenNebula controller (running Sunstone) and an OpenNebula host. My oned.conf file contains this setting: </div><div><br></div><div><div>
<span style="white-space:pre-wrap"> </span># MDF 270313: changed datastore location</div><div><span style="white-space:pre-wrap"> </span>DATASTORE_LOCATION = /opt/one/datastores</div></div><div><br></div><div>.. so all images are instantiated in the /opt/one folder instead of the /var/lib/one folder. I needed to do this because of the partitioning scheme of my host. </div>
<div><br></div><div>On the controller I installed the disk image "Ubuntu Server 12.04 (Precise Pangolin) - kvm" from the Marketplace. This is an 11GB image containing a contextualized version of Ubuntu Server 12.04. I did not specify a device prefix or target so the image will use "hd" and "hda". </div>
<div><br></div><div>My Virtual Machine Template looks like this:</div><div><br></div><div><div>CPU="0.10"</div><div>DISK=[</div><div> BUS="virtio",</div><div> IMAGE_ID="3" ]</div><div>FEATURES=[</div>
<div> ACPI="yes",</div><div> PAE="yes" ]</div><div>GRAPHICS=[</div><div> LISTEN="localhost",</div><div> PORT="5911",</div><div> TYPE="vnc" ]</div><div>MEMORY="2000"</div>
<div>NAME="Ubuntu 12.04 Test"</div><div>NIC=[</div><div> IP="10.0.0.11",</div><div> MODEL="virtio",</div><div> NETWORK="virtual machine lan" ]</div><div>OS=[</div><div> ARCH="x86_64" ]</div>
<div>RAW=[</div><div> DATA="<devices><serial type=\"pty\"><source path=\"/dev/pts/5\"/><target port=\"0\"/></serial><console type=\"pty\" tty=\"/dev/pts/5\"><source path=\"/dev/pts/5\"/><target port=\"0\"/></console></devices>",</div>
<div> TYPE="kvm" ]</div><div>TEMPLATE_ID="1"</div></div><div><br></div><div>Note that I added a "BUS" argument in the DISK section and a "MODEL" argument in the NIC section to make the VM use the virtio driver. </div>
<div><br></div><div>When I start a new VM using this template I see the VM going from PENDING -> PROLOG -> RUNNING. But if I connect to the host using Nomachine and then use virt-manager to inspect the console of the virtual machine, I see this: </div>
<div><br></div><div><img height="419" width="740" src="cid:F47259B4-96D3-4D61-9490-604CD4E87ACF@fritz.box"></div><div><br></div><div>In case the screenshot does not appear correctly in this e-mail, the start up messages are: </div>
<div><br></div><div>TCP cubic registered</div><div>NET: Registered protocol family 10</div><div>Mobile IPv6</div><div>NET: Registered protocol family 17</div><div>registered taskstats version 1</div><div>rtc_cmos 00:01: setting system clock to ......</div>
<div>Initializing network drop monitor service</div><div>List of all partitions:</div><div>No filesystem could mount root, tried:</div><div>Kernel panic - not syncing : VFS: Unable to mount root fs on unknown block(254,32)</div>
<div>Pid: 1, comm: swapper Not tainted 2.6.32-5-amd64 #1</div><div>Call trace .....</div><div><br></div><div>Nothing I have tried so far has made this problem go away. I already tried : </div><div><br></div><div><ul><li>using a different image file</li>
<li>adding and removing the BUS argument in the DISK section</li><li>using different DEVICE PREFIX and disk target settings</li><li>experimenting with persistent and non-persistent disks</li></ul></div><div><br></div><div>
I know my OpenNebula configuration is working because :</div><div><br></div><div><ul><li>I can see the host CPU load and memory use in Sunstone</li><li>The image files transfer correctly from the front end to the host and appear in /opt/one/datastores in the correct folder</li>
<li>The virtual machine ends up in the RUNNING state</li><li>I am able to manually start VM's on the host by using virt-manager</li></ul><div><br></div><div>Right now I have a working KVM installation and I am able to manually create KVM images and run them in virt-manager. But when I try to use OpenNebula to create and deploy KVM images I see a new VM appear in virt-manager but it does not boot. </div>
</div><div><br></div><div>What could this be? </div><div><br></div><div>Thanks for any support on this! </div><div><br></div><div>Regards,</div><div><br></div><div><br></div><div> - Mark Farragher</div><div><br></div><div>
<br></div><div><br></div><div><br></div></div></div><br></div><br>_______________________________________________<br>
Users mailing list<br>
<a href="mailto:Users@lists.opennebula.org">Users@lists.opennebula.org</a><br>
<a href="http://lists.opennebula.org/listinfo.cgi/users-opennebula.org" target="_blank">http://lists.opennebula.org/listinfo.cgi/users-opennebula.org</a><br>
<br></blockquote></div><br><br clear="all"><div><br></div>-- <br>Javier Fontán Muiños<br>Project Engineer<br>OpenNebula - The Open Source Toolkit for Data Center Virtualization<br><a href="http://www.OpenNebula.org" target="_blank">www.OpenNebula.org</a> | <a href="mailto:jfontan@opennebula.org" target="_blank">jfontan@opennebula.org</a> | @OpenNebula
</div></div>