1f9c1f7385
This refactors the nova-novncproxy configuration in TripleO
so that it can be used in a cloud where the compute hosts
don't each have to be assigned a public IP. With the
new configuration the nova-novncproxy is configured
to run on the nova-api (controller) node which
typically will have a public IP assigned to it.
Specific changes include:
-move novnc element to nova-novncproxy
-move all the nova-novncproxy stuff out of nova-kvm into
the new nova-novncproxy element
-configures Fedora iptables rules to support noVNC
connections on 6080
-Adds a new nova.public_ip parameter which is used
to construct the novncproxy_base_url. This may be
returned directly to cloud users via
'nova get-vnc-console' or consumed by Horizon.
-Set vncserver_proxyclient_address to the real
local(internal) IP address for the compute host.
Previously we used 127.0.0.1 which won't work
unless nova-novncproxy runs on the Nova compute host.
To do this we use {{local-ipv4}} which should always
be the Heat provided local management IP address
for the compute host.
Change-Id: I18cdc36248749390f0ab1e2c454741dbf2736e7f
Closes-bug: #1309510
1.6 KiB
Install nova software from git.
Configuration
nova: verbose: False - Print more verbose output (set logging level to INFO instead of default WARNING level). debug: False - Print debugging output (set logging level to DEBUG instead of default WARNING level). baremetal: pxe_deploy_timeout: "1200" - the duration in seconds for pxe deployment timeouts. virtual_power: type: "virsh" - what virtual power driver to use. "virsh" or "vbox" compute_libvirt_type: "qemu" - what libvirt compute type. Unset will use the nova default. image_cache_manager_interval: - Number of seconds to wait between runs of the image cache manager. resize_fs_using_block_device: BoolOpt - Attempt to resize the filesystem by accessing the image over a block device. resume_guests_state_on_host_boot: BoolOpt - Whether to start guests that were running before the host rebooted. running_deleted_instance_action: - Action to take if a running deleted instance is detected. Valid options are: 'noop', 'log', 'shutdown', or 'reap'. Set to 'noop' to take no action. virt_mkfs: - Name of the mkfs commands for ephemeral device. The format is <os_type>= e.g. 'linux-ext4=mkfs -t ext4 -F -L %(fs_label)s %(target)s' compute_manager: "ironic.nova.compute.manager.ClusterComputeManager" - set to override the compute manager class used by Nova-Compute. public_ip: - public IP address (if any) assigned to this node. Used for VNC proxy connections so this is typically only required on controller nodes.