I have solved my similar situation by adding vif plugin entries in the nova.conf under [DEFAULT] section

vif_plugging_is_fatal: false

vif_plugging_timeout: 0

Hope this helps

I was getting the same errors on a 3-node setup, created with packstack. I had configured one of the nodes as controller and running compute and network services on all 3 nodes. Every other instance creationg would log Timeout waiting for vif plugging callback for instance in nova-compute.log and the throw VirtualInterfaceCreateException: Virtual Interface creation failed exception immediately after that.

Setting vif_plugging_is_fatal: false seems to work around the problem, but then I was getting random non-VM network errors, like router ports not being properly attached to private networks after creation (status DOWN).

I turned out I was missing the nova interactions settings in /etc/neutron/neutron.conf on the 2 non-controller nodes:

notify_nova_on_port_status_changes = True
notify_nova_on_port_data_changes = True
nova_url = http://<controller-ip>:8774/v2
nova_region_name = RegionOne
nova_admin_username = nova
nova_admin_tenant_name = services
nova_admin_password = <your-password>
nova_admin_auth_url = http://<controller-ip>:5000/v2.0
send_events_interval = 2

Packstack configured this only on the first (controller) node. Adding them to the other 2 nodes sovled the issue immediately.

I too got this error With Icehouse in Ubuntu 14.04

2014-04-19 19:35:19.470 20217 INFO nova.virt.libvirt.driver [req-b061faf1-32bc-42d1-a1f4-822fcc9aaca5 3e7ba37aa6cf4472b823c8057a29c217 93a7bf9a8cb64c0da27486e722ce07c1] [instance: bf2d05ff-1de4-459b-8b18-3fc2a8ed44e6] Deleting instance files /var/lib/nova/instances/bf2d05ff-1de4-459b-8b18-3fc2a8ed44e6 2014-04-19 19:35:19.472 20217 INFO nova.virt.libvirt.driver [req-b061faf1-32bc-42d1-a1f4-822fcc9aaca5 3e7ba37aa6cf4472b823c8057a29c217 93a7bf9a8cb64c0da27486e722ce07c1] [instance: bf2d05ff-1de4-459b-8b18-3fc2a8ed44e6] Deletion of /var/lib/nova/instances/bf2d05ff-1de4-459b-8b18-3fc2a8ed44e6 complete 2014-04-19 19:35:19.623 20217 ERROR nova.compute.manager [req-b061faf1-32bc-42d1-a1f4-822fcc9aaca5 3e7ba37aa6cf4472b823c8057a29c217 93a7bf9a8cb64c0da27486e722ce07c1] [instance: bf2d05ff-1de4-459b-8b18-3fc2a8ed44e6] Instance failed to spawn 2014-04-19 19:35:19.623 20217 TRACE nova.compute.manager [instance: bf2d05ff-1de4-459b-8b18-3fc2a8ed44e6] Traceback (most recent call last): 2014-04-19 19:35:19.623 20217 TRACE nova.compute.manager [instance: bf2d05ff-1de4-459b-8b18-3fc2a8ed44e6] File "/usr/lib/python2.7/dist-packages/nova/compute/manager.py", line 1720, in _spawn 2014-04-19 19:35:19.623 20217 TRACE nova.compute.manager [instance: bf2d05ff-1de4-459b-8b18-3fc2a8ed44e6] block_device_info) 2014-04-19 19:35:19.623 20217 TRACE nova.compute.manager [instance: bf2d05ff-1de4-459b-8b18-3fc2a8ed44e6] File "/usr/lib/python2.7/dist-packages/nova/virt/libvirt/driver.py", line 2253, in spawn 2014-04-19 19:35:19.623 20217 TRACE nova.compute.manager [instance: bf2d05ff-1de4-459b-8b18-3fc2a8ed44e6] block_device_info) 2014-04-19 19:35:19.623 20217 TRACE nova.compute.manager [instance: bf2d05ff-1de4-459b-8b18-3fc2a8ed44e6] File "/usr/lib/python2.7/dist-packages/nova/virt/libvirt/driver.py", line 3663, in _create_domain_and_network 2014-04-19 19:35:19.623 20217 TRACE nova.compute.manager [instance: bf2d05ff-1de4-459b-8b18-3fc2a8ed44e6] raise exception.VirtualInterfaceCreateException() 2014-04-19 19:35:19.623 20217 TRACE nova.compute.manager [instance: bf2d05ff-1de4-459b-8b18-3fc2a8ed44e6] VirtualInterfaceCreateException: Virtual Interface creation failed 2014-04-19 19:35:19.623 20217 TRACE nova.compute.manager [instance: bf2d05ff-1de4-459b-8b18-3fc2a8ed44e6]

Any workarounds for this issue.

I'm experiencing the same issue here but even weird. Sometimes I can spawn several VMs without any issues, sometimes I can spawn only one and only services restart helps. Behaviour is like unpredictable. I'm using openvswitch plugin with vlans. I've tried both linuxnet_interface_driver=nova.network.linux_net.LinuxOVSInterfaceDriver and deprecated libvirt_vif_driver=nova.virt.libvirt.vif.LibvirtHybridOVSBridgeDriver

A workaroung mentioned above: vif_plugging_timeout = 10 vif_plugging_is_fatal = False

seems helps and networking is ok. There is interface attached with assigned ip. May 1 05:01:32 cirros kern.info kernel: [ 0.958597] e1000: Intel(R) PRO/1000 Network Driver - version 7.3.21-k8-NAPI May 1 05:01:32 cirros kern.info kernel: [ 0.958600] e1000: Copyright (c) 1999-2006 Intel Corporation.

But that's all looks like a bug. I'm on Wheezy though.

I think this was the problem: https://bugs.launchpad.net/neutron/+bug/1301449

(and I had neutron-server running on controller and network...stupid me :) )

Hello

I had the same problem ("Timeout waiting for vif plugging callback for instance"), and it can be caused by an invalid Neutron configuration. For me the solution was to fix a typo in /etc/neutron/plugins/ml2/ml2_conf.ini:

I replaced: firewall_driver=neutron.agent.linux.iptables_firewall.OVSHybridIpTablesFirewallDriver

With: firewall_driver=neutron.agent.linux.iptables_firewall.OVSHybridIptablesFirewallDriver

But your problem can be caused by anything that prevents Neutron from working properly. To find the error, you could go to /var/log/neutron on the misbehaving node, and do:

tail -f *

And wait for errors to appear. My error was something like "Module OVSHybridIpTablesFirewallDriver doesn't exist", because of the typo.

Hi, I have similar problem, I'm using devstack on ubuntu 12.04, so how to do? Thanks

Message Virtual Interface creation failed Code 500 Details File "/opt/stack/nova/nova/compute/manager.py", line 301, in decorated_function return function(self, context, args, *kwargs) File "/opt/stack/nova/nova/compute/manager.py", line 2120, in run_instance do_run_instance() File "/opt/stack/nova/nova/openstack/common/lockutils.py", line 249, in inner return f(args, *kwargs) File "/opt/stack/nova/nova/compute/manager.py", line 2119, in do_run_instance legacy_bdm_in_spec) File "/opt/stack/nova/nova/compute/manager.py", line 1237, in _run_instance notify("error", fault=e) # notify that build failed File "/opt/stack/nova/nova/openstack/common/excutils.py", line 82, in __exit__ six.reraise(self.type_, self.value, self.tb) File "/opt/stack/nova/nova/compute/manager.py", line 1221, in _run_instance instance, image_meta, legacy_bdm_in_spec) File "/opt/stack/nova/nova/compute/manager.py", line 1364, in _build_instance network_info.wait(do_raise=False) File "/opt/stack/nova/nova/openstack/common/excutils.py", line 82, in __exit__ six.reraise(self.type_, self.value, self.tb) File "/opt/stack/nova/nova/compute/manager.py", line 1341, in _build_instance set_access_ip=set_access_ip) File "/opt/stack/nova/nova/compute/manager.py", line 401, in decorated_function return function(self, context, args, *kwargs) File "/opt/stack/nova/nova/compute/manager.py", line 1758, in _spawn LOG.exception(_('Instance failed to spawn'), instance=instance) File "/opt/stack/nova/nova/openstack/common/excutils.py", line 82, in __exit__ six.reraise(self.type_, self.value, self.tb) File "/opt/stack/nova/nova/compute/manager.py", line 1755, in _spawn block_device_info) File "/opt/stack/nova/nova/virt/libvirt/driver.py", line 2287, in spawn block_device_info) File "/opt/stack/nova/nova/virt/libvirt/driver.py", line 3699, in _create_domain_and_network raise exception.VirtualInterfaceCreateException()

I ran into the same issue on a 3 node (controller, network, compute) icehouse install on physical servers. In my case, the neturon error logs displayed "ERROR neutron.notifiers.nova Failed to notify nova on events: ['status':'completed', 'name':''network-vif-unplugged', 'server_uuid':'u'f4a..']" The above was due to a configuration error in the neutron.conf file in my controller node. My entry for the field "nova_url" was "http://10.0.0.35:8774/v2/v2" instead of "http://10.0.0.35:8774/v2". Took me a couple of days over the weekend to figure this out.

I got this error in the nova logs of compute node while creating a new instance in openstack liberty. It was because neutron was not able to convey nova that a virtual interface has been created. I made sure I had the following conf in nova section of neutron.conf on controller node so that neutron identifies where the nova service is and conveys of the networking events.

[nova]
auth_url = http://controller:35357
auth_plugin =password
project_domain_id = default
user_domain_id = default
region_name = RegionOne
project_name = service
username = nova
password = nova