5408a4f556
Nova uses a data container. Data containers are expected to use the VOLUME keyword and then use volumes_from to access those volumes. Throwing in a bindmount to the host filesystem results in unpredicble behavior, usually in the form of files being stored on the host filesystem that should be persisted in the compute data volume. Change-Id: I60a7873995c9397368ed4df61d5a6d4e8250f3c2 |
||
|---|---|---|
| .. | ||
| glance-api-registry.yml | ||
| heat-api-engine.yml | ||
| horizon.yml | ||
| keystone.yml | ||
| mariadb.yml | ||
| neutron-agents.yml | ||
| neutron-server.yml | ||
| nova-api-compute-network.yml | ||
| nova-api-conductor-scheduler.yml | ||
| nova-compute-network.yml | ||
| nova-compute.yml | ||
| rabbitmq.yml | ||
| README.md | ||
Docker compose
These scripts and docker compose files can be used to stand up a simple installation of openstack. Running the 'tools/genenv' script creates an 'openstack.env' suitable for running on a single host system as well as an 'openrc' to allow access to the installation.
Once you have run that you can either manually start the containers using the 'docker-compose' command or try the 'tools/kolla start' script which tries to start them all in a reasonable order, waiting at key points for services to become available. Once stood up you can issue the typical openstack commands to use the installation:
# source openrc
# nova network-create vmnet --fixed-range-v4=10.0.0.0/24 --bridge=br100 --multi-host=T
# nova secgroup-add-rule default tcp 22 22 0.0.0.0/0
# nova secgroup-add-rule default icmp -1 -1 0.0.0.0/0
#
# nova keypair-add mykey > mykey.pem
# chmod 600 mykey.pem
# nova boot --flavor m1.medium --key_name mykey --image puffy_clouds instance_name
# ssh -i mykey.pem cirros@<ip>