OpenStack Compute, codenamed Nova, is a cloud computing fabric controller. In
addition to its "native" API (the OpenStack API), it also supports the Amazon
- openstack ›
This charm provides Openstack Nova compute for Hyper-V.
juju deploy nova-hyperv juju add-relation nova-hyperv nova-cloud-controller juju add-relation nova-hyperv rabbitmq-server juju add-relation nova-hyperv glance
Ideally you will deploy this charm to a machine that has at least 2 network cards. If only one is present, this charm will add that interface to a vmswitch and enable management OS on the bridge itself.
If more then one NIC is present, you will have to specify the data-port using the charm config. When adding a new node, make sure to update the data-port field using:
juju set nova-hyperv data-port="aa:aa:aa:aa:aa:aa aa:aa:aa:aa:aa:ab"
where aa:aa:aa:aa:aa:ab is the second machine.
- (string) Computer group that will be joined after adding AD relation
- (string) User that will created after adding AD relation
- (string) The name of the bond interface
- (string) MAC addresses of ports participating in bond link. A new Net-LbfoTeam called bond0 will be created in LACP mode. This interface can be used as a data-port.
- (string) Ethernet inerface to use for vmswitch. This can be a MAC address or an interface name. This option is mandatory if more then one interface is present on the target system. In cases where there is only one interface on the system, the vmswitch will be created using that interface and OSManagement will be set to True, regardless of the vmswitch-management option. Avoid using the management interface as a data interface where system has more then one NIC. If you would like to use bonding, make sure you supply valid MAC addresses in the bond-ports config field, and use "bond0" as a data-port name.
- (boolean) Enable debug logging
- (boolean) Enables the Hyper-V compute driver for clustering. This requires at least two Hyper-V compute joined to a cluster. NOTE: Available only in releases >= Mitaka
- (string) overwrite default installer location. Take care to set the correct openstack-version as that will be used by the charm to pick the correct config templates. If you use this option, you may enable file integrity checking by specifying an md5sum at the end of the URL as follows: https://example.com/installer.msi#md5=<hash>
- (string) Location in which to store VM disks
- (string) Log folder location
- (string) Yaml with the additional checks to be set on 'local-monitors' relation with NSClient++
- (string) Neutron agent to use. Valid options are: * hyperv * ovs Default is hyperv. Note: ovs is available for Kilo and above
- (string) Version of Openstack nova compute to install. Choices are: - juno - kilo - liberty
- (string) Version of Openstack nova compute to install. This option overrides openstack-origin. Choices are: - juno - kilo - liberty
- (string) The IP address and netmask of the OpenStack Data network (e.g., 192.168.0.0/24) . This network will be used for tenant network traffic in overlay networks. NOTE: If ovs is selected as the network type, this option overrides data-port. OVS on Windows requires that the os data network be set on the vmswitch itself. That means that if an interface has an IP in the os-data-network range, that interface will be added to the vmswitch. There is a caveat here. If os-data-network is on a different interface then data-port, it might result in the hyper-v compute nodes not being able to use VLAN network segregation.
- (string) Location of the OVS MSI installer
- (string) Username used to access rabbitmq queue
- (string) Rabbitmq vhost
- (boolean) Enable verbose logging
- (boolean) Allow management OS on this vm switch. If True, the switch will be set in DHCP mode.
- (string) Virtual switch to use