Jump to: navigation, search

Difference between revisions of "NeutronDevstack"

Line 1: Line 1:
 
__NOTOC__
 
__NOTOC__
 
== Basic Setup ==
 
== Basic Setup ==
 
'''Note:''' This description covers only Quantum's v2 API (i.e. Folsom code).
 
  
 
In order to use Quantum with devstack (http://devstack.org) a single node setup, you'll need the following settings in your localrc (see [http://devstack.org/stack.sh.html  this page ] for more details on localrc).
 
In order to use Quantum with devstack (http://devstack.org) a single node setup, you'll need the following settings in your localrc (see [http://devstack.org/stack.sh.html  this page ] for more details on localrc).
Line 12: Line 10:
 
enable_service q-agt
 
enable_service q-agt
 
enable_service q-dhcp
 
enable_service q-dhcp
 +
enable_service q-l3
 
enable_service quantum
 
enable_service quantum
Q_PLUGIN=openvswitch
 
 
</nowiki></pre>
 
</nowiki></pre>
  
Line 19: Line 17:
 
Then run stack.sh as normal.
 
Then run stack.sh as normal.
  
'''Note:''' For more information about creating networks/subnets and VMs attached to them, see: [[RunningQuantumV2Api]]
+
See the Quantum Admin Guide for details on interacting with Quantum: [[ http://docs.openstack.org/trunk/openstack-network/admin/content/index.html ]]
  
 
== Multi-Node Setup ==
 
== Multi-Node Setup ==
Line 33: Line 31:
 
enable_service q-agt
 
enable_service q-agt
 
enable_service q-dhcp
 
enable_service q-dhcp
 +
enable_service q-l3
 
enable_service quantum
 
enable_service quantum
Q_PLUGIN=openvswitch
 
NOVA_USE_QUANTUM_API=v2
 
 
</nowiki></pre>
 
</nowiki></pre>
  
Line 52: Line 49:
 
<pre><nowiki>
 
<pre><nowiki>
 
ENABLED_SERVICES=n-cpu,rabbit,g-api,quantum,q-agt,q-dhcp
 
ENABLED_SERVICES=n-cpu,rabbit,g-api,quantum,q-agt,q-dhcp
Q_PLUGIN=openvswitch
 
NOVA_USE_QUANTUM_API=v2
 
 
SERVICE_HOST=[IP of controller node]
 
SERVICE_HOST=[IP of controller node]
 
MYSQL_HOST=$SERVICE_HOST
 
MYSQL_HOST=$SERVICE_HOST
Line 62: Line 57:
  
 
'''Note:''' the need to include 'g-api' and 'rabbit' here seems to be a bug, which may have been fixed by the time you're reading this.  Without it, nova-compute dies because it can't import the glance.common library.  This process does not actually need to be running on this host, we just need a way to make sure the glance.common library is installed.  If 'rabbit' is not specified, nova-compute also will try to connect to rabbit on localhost, not the "controller host".  See the following link for info on both issues: https://answers.launchpad.net/devstack/+question/197749
 
'''Note:''' the need to include 'g-api' and 'rabbit' here seems to be a bug, which may have been fixed by the time you're reading this.  Without it, nova-compute dies because it can't import the glance.common library.  This process does not actually need to be running on this host, we just need a way to make sure the glance.common library is installed.  If 'rabbit' is not specified, nova-compute also will try to connect to rabbit on localhost, not the "controller host".  See the following link for info on both issues: https://answers.launchpad.net/devstack/+question/197749
 
== Using Quantum ==
 
 
Now you have Quantum up and running, look at [[RunningQuantumV2Api]] to find out how to make of v2 Quantum.  The commands you need to use to create a network in v2 have changed since the Essex documentation.
 

Revision as of 20:03, 26 September 2012

Basic Setup

In order to use Quantum with devstack (http://devstack.org) a single node setup, you'll need the following settings in your localrc (see this page for more details on localrc).


disable_service n-net
enable_service q-svc
enable_service q-agt
enable_service q-dhcp
enable_service q-l3
enable_service quantum


Then run stack.sh as normal.

See the Quantum Admin Guide for details on interacting with Quantum:  http://docs.openstack.org/trunk/openstack-network/admin/content/index.html 

Multi-Node Setup

A more interesting setup involves running multiple compute nodes, with Quantum networks connecting VMs on different compute nodes.

You should run at least one "controller node", which should have a stackrc that includes at least:


disable_service n-net
enable_service q-svc
enable_service q-agt
enable_service q-dhcp
enable_service q-l3
enable_service quantum


You likely want to change your localrc to run a scheduler that will balance VMs across hosts:


SCHEDULER=nova.scheduler.simple.SimpleScheduler


You can then run many compute nodes, each of which should have a stackrc which includes the following, with the IP address of the above controller node:


ENABLED_SERVICES=n-cpu,rabbit,g-api,quantum,q-agt,q-dhcp
SERVICE_HOST=[IP of controller node]
MYSQL_HOST=$SERVICE_HOST
RABBIT_HOST=$SERVICE_HOST
Q_HOST=$SERVICE_HOST


Note: the need to include 'g-api' and 'rabbit' here seems to be a bug, which may have been fixed by the time you're reading this. Without it, nova-compute dies because it can't import the glance.common library. This process does not actually need to be running on this host, we just need a way to make sure the glance.common library is installed. If 'rabbit' is not specified, nova-compute also will try to connect to rabbit on localhost, not the "controller host". See the following link for info on both issues: https://answers.launchpad.net/devstack/+question/197749