openstack/training-guides
Code Issues Proposed changes

labs: take additional snapshots of controller node

Browse Source 
Take a snapshot of the controller VM after installing the compute and
network nodes.

The configuration of compute and network nodes is largely saved in the
database running on the controller node, but until now these database
changes were not part of any snapshots. With this changeset, resetting
all nodes to the most recent snapshots restores a working cluster.

Change-Id: Ifd2ca144d176280bc6677a2e6381afb63a9feea3
This commit is contained in:
Roger Luethi 2014-10-03 12:02:10 +02:00
parent 1614ac0e02
commit 578c593607
2 changed files with 10 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
labs/config/scripts.compute
View File

@ -8,4 +8,9 @@ cmd snapshot_cycle nova-compute_installed
cmd queue setup_neutron_compute.sh
cmd queue setup_cinder_volumes.sh
cmd snapshot_cycle compute_node_installed
# Take snapshot of changes on controller VM, too
cmd queue shutdown_controller.sh
cmd boot
cmd wait_for_shutdown -n controller
cmd snapshot -n controller compute_node_installed
cmd boot -n controller

5
labs/config/scripts.network
View File

@ -5,4 +5,9 @@ cmd queue osbash/enable_vagrant_ssh_keys.sh
cmd snapshot_cycle network_configured
cmd queue setup_neutron_network.sh
cmd snapshot_cycle network_node_installed
# Take snapshot of changes on controller VM, too
cmd queue shutdown_controller.sh
cmd boot
cmd wait_for_shutdown -n controller
cmd snapshot -n controller network_node_installed
cmd boot -n controller