VEM charm installs the Cisco Nexus 1000v virtual switch onto the compute/network nodes.
VEM charm is designed as a subordinate charm. The aim is to have this charm installed on the nova-compute and the quantum-gateway hosts.
In order to use Cisco Openstack solution we would need to install VEM on the nova-compute and quantum-gateway hosts. We need to have nova-compute deployed first and we would have vem charm as subordinate it.
In the config.yaml you can provide general config that will be common to all VEM hosts in environement. If you need to configure host-specific config to each host depending on its fqdn, a mapping file can be provided as a string to the variable called mapping.
juju deploy nova-compute
juju deploy --config=config.yaml vem
juju add-relation nova-compute vem
juju set vem mapping="$(cat mapping.yaml)"
Here is a sample of the mapping file:
maas-node-1:
host_mgmt_intf: eth0
uplink_profile: phys eth3 profile sys-uplink
node_type: compute
maas-node-3:
host_mgmt_intf: eth0
uplink_profile: phys eth4 profile sys-uplink
node_type: network
vtep_config: 'virt vtep1-nw2 profile vxlan-vtep mode static address 6.0.8.253 netmask 255.255.0.0 mac 0e:1f:56:cf:82:53'
In this way, the hosts in the mapping mentioned in the mapping file will get these specific config which will overwrite the generate config provided in the config.yaml
Contact Information
Author: Shiva Prasad Rao shivrao@cisco.com
Report bugs at: http://bugs.launchpad.net/charms/+source/vem
Location: http://jujucharms.com/charms/distro/vem