Bridge interfaces not created on lxc hosts - openstack ansible

asked 2019-11-09 23:57:25 -0500

Josh Ploude gravatar image

Running Openstack-ansible on Ubuntu 18.04 hosts with Openstack Train, containers are created but each of the three hosts fail to have br-mgmt configured on it. I can create br-mgmt manually and containers will then get IP addresses and start but root problem is that the hosts don't get br-mgmt and br-vxlan bridge interfaces as part of the setup-hosts.yml process.

Hosts to get lxcbr0 interfaces but not br-mgmt or any other bridges after that.

Here is the openstack_user_config.yml file first part:

global_overrides: internal_lb_vip_address: external_lb_vip_address: management_bridge: "br-mgmt" tunnel_bridge: "br-vxlan" provider_networks: - network: group_binds: - all_containers - hosts type: "raw" container_bridge: "br-mgmt" container_interface: "eth1" container_type: "veth" ip_from_q: "container" is_container_address: true - network: group_binds: - glance_api - cinder_api - cinder_volume - nova_compute type: "raw" container_bridge: "br-storage" container_type: "veth" container_interface: "eth2" container_mtu: "9000" ip_from_q: "storage" - network: group_binds: - neutron_linuxbridge_agent container_bridge: "br-vxlan" container_type: "veth" container_interface: "eth10" container_mtu: "9000" ip_from_q: "tunnel" type: "vxlan" range: "1:1000" net_name: "vxlan"

I am tracing the openstack ansible process to determine why br-mgmt wouldn't be created. Where is best place to look? Does anybody have ideas as to something I am doing wrong on the config file so that br-mgmt shouldn't be created?

Thank you for assistance and any direction in advance.

edit retag flag offensive close merge delete

1 answer

Sort by ยป oldest newest most voted

answered 2019-11-11 14:05:09 -0500

Josh Ploude gravatar image

Turns out only lxcbr0 is created by OSA. All other bridges have to be created manually unless you create playbook and/or tasks to do this. Out of the box the bridges have to be configured on each target host prior to running the setup-hosts playbook.

Did this in netplan and all containters are up and running. The info was quite clear in the deployment guide, I just missed it.

In case it helps anyone, netplan configuration on the target hosts that was used is found below.

"""network: version: 2 ethernets: eno4: addresses: [] dhcp4: false gateway4: nameservers: addresses: - - search: - ens2f0: {} vlans: vlan.302: id: 302 link: ens2f0 vlan.401: id: 401 link: ens2f0 vlan.402: id: 402 link: ens2f0 vlan.403: id: 403 link: ens2f0 bridges: br-mgmt: interfaces: [vlan.401] addresses: [] mtu: 1500 br-tunnel: interfaces: [vlan.402] addresses: [] mtu: 1500 br-storage: interfaces: [vlan.403] addresses: [] br-flat: interfaces: [vlan.302] addresses: []"""

edit flag offensive delete link more

Get to know Ask OpenStack

Resources for moderators

Question Tools

1 follower


Asked: 2019-11-09 23:53:23 -0500

Seen: 153 times

Last updated: Nov 09 '19