Multi-Node Compute Installation
Hi,
I was in the process of setting up a production Openstack Cluster. I have some very major doubts in my mind due to which I am not able to proceed any further. Please clear some of the
1) I have a cloud controller noode which runs all nova services. But on this node I don't want nova-compute to run, though I have installed it as I was following a tutorial. 2) This node also runs the MySQL as the database and Keystone as the Identity server, RabbitMQ, nova-api etc and the openstack dashboard 3) I have two Xen Cloud Platform Hosts. I have installed a domU guest on each of them which runs the nova-compute only. 4) In addition one of these domUs also runs the glance imaging service with swift as the backend object store. 5) I was successfully able to integrate glance and keystone with swift and was able to view images in the openstack dashboard.
Now the questions start
1) If I run nova-network on all compute nodes and not on the cloud-controller node then would this setup work? To rephrase, is it necessary or suggested to run the netwok services from the compute node, considering i use FlatNetworkManager. I mean one network manager on every compute node and none on the cloud-controller if it does not run the nova-compute 2) If the compute node runs only nova-compute, then how does it talk to the rest of the cloud and conversely how does the cloud controller come to know of the compute nodes in its cluster. 3) Since I am using keystone, do I need to create projects exclusively using euca2ools or keystone profiles will work 4) The nova-compute service also refuses to start on the domU compute node.
I am attaching the configuration files here. Kindly suggest the changes to meet my requirements
(I) The cloud-controller node ----- All services except nova-compute
------------------------------------------nova.conf-------------------------------------------------------------------
DATABASE
--sql_connection=mysql://nova:notnova@10.1.27.112:3306/nova
LOGS/STATE
--verbose --dhcpbridge_flagfile=/etc/nova/nova.conf --dhcpbridge=/usr/bin/nova-dhcpbridge --logdir=/var/log/nova --state_path=/var/lib/nova --lock_path=/var/lock/nova
RABBITMQ
--rabbit_password=guest --rabbit_port=5672 --rabbit_host=10.1.27.112
SCHEDULER
--scheduler_driver=nova.scheduler.simple.SimpleScheduler
NETWORK
--network_manager=nova.network.manager.FlatDHCPManager --fixed_range=192.168.100.0/24 --flat_network_dhcp_start=192.168.100.2 --public_interface=eth0 --flat_interface=eth0 --flat_network_bridge=br100
GLANCE
--image_service=nova.image.glance.GlanceImageService --glance_api_servers=10.1.27.113:9292
COMPUTE
--compute_manager=nova.compute.manager.ComputeManager --libvirt_type=qemu
VNCPROXY
#--vncproxy_url=http://192.168.206.130:6080 #--vncproxy_wwwroot=/var/lib/nova/noVNC
MISC
--use_deprecated_auth=false --allow_admin_api=true --enable_zone_routing=true
KEYSTONE
--keystone_ec2_url=http://10.1.27.112:5000/v2.0/ec2tokens
-----------------------------------api-paste.ini--------------------------------------------------------------------------------
#######
EC2
#######
[composite:ec2] use = egg:Paste#urlmap /: ec2versions /services/Cloud: ec2cloud /services/Admin: ec2admin /latest: ec2metadata /2007-01-19: ec2metadata /2007-03-01: ec2metadata /2007-08-29: ec2metadata /2007-10-10: ec2metadata /2007-12-15: ec2metadata /2008-02-01: ec2metadata /2008-09-01: ec2metadata /2009-04-04: ec2metadata /1.0: ec2metadata
[pipeline:ec2cloud] #pipeline = logrequest ec2noauth cloudrequest authorizer ec2executor #pipeline = logrequest authenticate cloudrequest authorizer ec2executor #pipeline = logrequest ec2lockout authenticate cloudrequest ...