We are trying to install openstack charms, as a part of it we installed MAAS and attempted to create a MAAS cluster. We tried adding a Virtual Machine created on KVM to MAAS. But we faced an error, "Failed to power on node - Power on for the node failed: Failed talking to node's BMC: Failed to login to virsh console".
Systems used:
For MAAS → KVM Virtual Machine, Ubuntu 20.04, 8GB RAM, 50GB, 2 vcpus
For JUJU → KVM Virtual Machine, PXE boot, 4GB RAM, 50GB, 2vcpus
Steps followed:
Installed KVM on Ubuntu 20.04 server.
Created a VM for MAAS.
Installed MAAS. Followed https://docs.openstack.org/project-deploy-guide/charm-deployment-guide/yoga/install-maas.html .
Enabled DHCP for MAAS.
Created another VM for juju installation and set boot option to pxe boot.
Added the machine to MAAS using virsh power type. Followed https://ubuntu.com/blog/quick-add-kvms-for-maas this article.
When we try to commission the node, its failing and we are facing an error.
We are stuck here and would appreciate any help.
Same here … Same setup with MaaS 3.1 works as expected but switching to 3.2 made VMs not to complete comissioning.
Same issue, fails on commissioning.
EDIT:Marking node failed - Node operation 'Commissioning' timed out after 30 minutes.
Could you please provide more details about your hardware setup? Are you trying to do this all on one server or do you have multiple bare metal machines?
Have you tried registering the VM host as a virsh host in maas? There’s currently a bug that prevents you from deploying a machine as a LXD/KVM host, but you should still be able to manually add the host under the KVM section.
Hi,
I am trying de deploy a node as a LXD vmhost, with
"maas myprofile machine deploy $LINUX_MACHINE_ID register_vmhost=True"
The node eventually enters deployment failed state after the deployment timeout. The node actually deploys and works, but...
My best guess is that your maas install doesn’t have access to manage your vm server. Try going to <maas-ip>:5240/MAAS/r/kvm/virsh and adding your vm host into maas. (you don’t have to provision the actual host with maas). I would also recommend you try lxd if you have trouble with virsh.
@mounika-alavala , can you share some more info:
The output of snap list maas or apt list maas (depending on which you’re using)
I think you’re using the UI, but please confirm.
Add log file output from:
maas.log
regiond.log
rackd.log
the rsyslog file of the affected machine(s), if it exists.
On snap, these files are located as follows (you’ll have to sudo su - to read them}:
/var/snap/maas/common/log/maas.log
/var/snap/maas/common/log/regiond.log
/var/snap/maas/common/log/rackd.log
/var/snap/maas/common/log/rsyslog/$MACHINE_NAME/$RELEVANT_DATE/messages
If you’re using packages, you’ll find the files in these locations:
/var/log/maas/maas.log
/var/log/maas/regiond.log
/var/log/maas/rackd.log
/var/log/maas/rsyslog/$MACHINE_NAME/$RELEVANT_DATE/messages
Hi @billwear , Thanks for responding. Below are the responses.
Output of snap list maas
Name Version Rev Tracking Publisher Notes
maas 3.2.6-12016-g.19812b4da 23947 3.2/stable canonical** -
Yes I am using UI
maas.log
2022-10-18T10:23:50.754401+05:30 maaskvm-Standard-PC-Q35-ICH9-2009 maas.import-images: [info] Downloading image descriptions from http:/>http://image >
regiond.log
2022-10-18 04:53:58 sstreams: [info] com.ubuntu.maas:stable:v3:download/com.ubuntu.maas.stable:v3:boot:22.04:amd64:ga-22.04-lowlatency: >
rackd.log
2022-10-18 09:33:41 provisioningserver.rackdservices.dhcp_probe_service: [info] Probe for external DHCP servers started on interfaces: e>
rsyslog file doesn’t exist.
Hi @ben-ihelputech , Thanks for responding. KVM section. I directly tried adding VM to Machines using virsh. Will try registering the host first.
thanks, @mounika-alavala ! i’ll take a look this week.
@mounika-alavala , your log files don’t show anything about the KVM interactions, and there’s no rsyslog file. can you pull up MAAS and see if it has actually has enlisted any machines?
