openstack_grizzly_install
openstack_grizzly_install copied to clipboard
Installation bash script for OpenStack Grizzly Release.
OpenStack Grizzly Installation Script
OpenStack Grizzly Installation Bash Script for Ubuntu Server 12.04 LTS.
Author
Tomokazu Hirai @jedipunkz
Twitter : https://twitter.com/jedipunkz Blog : http://jedipunkz.github.io
Notice
This script was tested ..
- all in one node with quantum
- separated nodes (controller node, network node, compute x n) with quantum
- all in one node with nova-network
- separated nodes (controller node, compute x n) with nova-network
- GRE Tunnel, VLAN
so, now I do not support separated nodes for each service (keystone, glance, nova, etc...). If you want to do this with separated nodes mode, please tell me for fork it.
Motivation
devstack is very usefull for me. I am using devstack for understanding openstack, especially Quantum ! ;) but when I reboot devstack node, all of openstack compornents was not booted. That is not good for me. and I wanted to use Ubuntu Cloud Archive packages.
Require Environment
Cinder Device
If you want use REAL disk device for cinder such as /dev/sdb, please input disk device name to $CINDER_VOLUME in setup.conf. If you do not have any additional disk for cinder, you can use loopback device. So please input loopback device name such as /dev/loop3.
In All in ne node mode
You need 2 NICs (management network, public network). You can run this script via management network NIC. VM can access to the internet via public network NIC (default : eth0, You can change device on setup.conf).
In separated nodes mode
You need 3 NICs for ..
- management network
- public network / API network (default: eth0)
- data network (default: eth1)
for more details, please see this doc.
http://docs.openstack.org/trunk/openstack-network/admin/content/app_demo_single_router.html
Quantum was designed on 4 networks (public, data, managememt, api) so You can 3 NICs on separated nodes mode. API network and Public network can share same network segment or you can separate these networks. This README's configuration of the premise is sharing a segment with API and Public network (default NIC : eth0).
How to use on All in One Node with quantum
Architecture
+------------------- Public/API Network
|
+------------+
|vm|vm|... |
+------------+
| all in one |
+------------+
| |
+-----)------------- Management/API Network
|
+------------- Data Network
- all of compornetns are on same node.
Setup network interfaces
Please setup network interfaces just like this.
% sudo ${EDITOR} /etc/network/interfaces
auto lo
iface lo inet loopback
# this NIC will be used for VM traffic to the internet
auto eth0
iface eth0 inet static
up ifconfig $IFACE 0.0.0.0 up
up ip link set $IFACE promisc on
down ip link set $IFACE promisc off
down ifconfig $IFACE down
address 10.200.9.10
netmask 255.255.255.0
dns-nameservers 8.8.8.8 8.8.4.4
dns-search example.com
# this NIC must be on management network
auto eth1
iface eth1 inet static
address 10.200.10.10
netmask 255.255.255.0
gateway 10.200.10.1
dns-nameservers 8.8.8.8 8.8.4.4
login and use this script via eth1 on management network. eth0 will be lost connectivity when you run this script. and make sure hostname resolv at /etc/hosts. in this sample, your host need resolv self fqdn in 10.200.10.10
Get this script
git clone this script from github.
% git clone git://github.com/jedipunkz/openstack_grizzly_install.git
% cd openstack_grizzly_install
% cp setup.conf.samples/setup.conf.allinone.quantum setup.conf
Edit parameters on setup.conf
There are many paramaters on setup.conf, but in 'allinone' mode, parameters which you need to edit is such things.
HOST_IP='10.200.10.10'
HOST_PUB_IP='10.200.9.10'
PUBLICNETWORK_NIC_NETWORK_NODE='eth0'
NETWORK_COMPONENT='quantum'
If you want to change other parameters such as DB password, admin password, please change these.
Run script
Run this script, all of conpornents will be built.
% sudo ./setup.sh allinone
That's all and You've done. :D Now you can access to Horizon (http://${HOST_IP}/horizon/) with user 'demo', password 'demo'.
How to use on All in One Node with nova-network
Architecture
+------------------- Public Network
|
+------------+
|vm|vm|... |
+------------+
| all in one |
+------------+
|
+------------------- Management/API Network
- all of compornetns are on same node.
Setup network interfaces
Please setup network interfaces just like this.
% sudo ${EDITOR} /etc/network/interfaces
auto lo
iface lo inet loopback
# this NIC will be used for VM traffic to the internet
auto eth0
iface eth0 inet static
address 10.200.9.10
netmask 255.255.255.0
gateway 10.200.9.1
dns-nameservers 8.8.8.8 8.8.4.4
dns-search example.com
# this NIC must be on management network
auto eth1
iface eth1 inet static
address 10.200.10.10
netmask 255.255.255.0
dns-nameservers 8.8.8.8 8.8.4.4
Get this script
git clone this script from github.
% git clone git://github.com/jedipunkz/openstack_grizzly_install.git
% cd openstack_grizzly_install
% cp setup.conf.samples/setup.conf.allinone.nova-network setup.conf
Edit parameters on setup.conf
There are many paramaters on setup.conf, but in 'allinone' mode, parameters which you need to edit is such things.
HOST_IP='10.200.10.10'
HOST_PUB_IP='10.200.9.10'
FLAT_INTERFACE='eth0'
NETWORK_COMPONENT='nova-network'
If you want to change other parameters such as DB password, admin password, please change these.
Run script
Run this script, all of conpornents will be built.
% sudo ./setup.sh allinone
That's all and You've done. :D Now you can access to Horizon (http://${HOST_IP}/horizon/) with user 'demo', password 'demo'.
How to use on separated nodes mode with quatum
Architecture
+-------------+-------------+------------------------------ Public/API Network
| | |
+-----------+ +-----------+ +-----------+ +-----------+ +-----------+
| | | | | | |vm|vm|.. | |vm|vm|.. |
| controller| | network | | network | +-----------+ +-----------+
| | | | | additional| | compute | | compute |
| | | | | | | | | additional|
+-----------+ +-----------+ +-----------+ +-----------+ +-----------+
| | | | | | | |
+-------------+-----)-------+-----)-------+-----)-------)-- Management/API Network
| | | |
+-------------+-------------+---------- Data Network
- minimum architecture : 3 nodes (controller node x 1, network node x 1, compute node x1)
- You can add some network nodes and compute nodes.
- additional network node(s) make you be able to have duplication of each agent
- additional compute node(s) make you be able to have more VMs.
Get this script
git clone this script from github on controller node.
controller% git clone git://github.com/jedipunkz/openstack_grizzly_install.git
controller% cd openstack_grizzly_install
controller% cp setup.conf.samples/setup.conf.separated.quantum setup.conf
Edit parameters on setup.conf
There are many paramaters on setup.conf, but in 'allinone' mode, parameters which you need to edit is such things.
CONTROLLER_NODE_IP='10.200.10.10'
CONTROLLER_NODE_PUB_IP='10.200.9.10'
NETWORK_NODE_IP='10.200.10.11'
COMPUTE_NODE_IP='10.200.10.12'
DATANETWORK_NIC_NETWORK_NODE='eth1'
DATANETWORK_NIC_COMPUTE_NODE='eth0'
PUBLICNETWORK_NIC_NETWORK_NODE='eth0'
NETWORK_COMPONENT='quantum'
If you want to change other parameters such as DB password, admin password, please change these.
copy to other nodes
copy directory to network node and compute node.
controller% scp -r openstack_grizzly_install <network_node_ip>:~/
controller% scp -r openstack_grizzly_install <compute_node_ip>:~/
Controller Node's network interfaces
Set up NICs for controller node.
controller% sudo ${EDITOR} /etc/network/interfaces
# The loopback network interface
auto lo
iface lo inet loopback
# for API network
auto eth0
iface eth0 inet static
address 10.200.9.10
netmask 255.255.255.0
gateway 10.200.9.1
dns-nameservers 8.8.8.8 8.8.4.4
dns-search example.com
# for management network
auto eth1
iface eth1 inet static
address 10.200.10.10
netmask 255.255.255.0
dns-nameservers 8.8.8.8 8.8.4.4
dns-search example.com
and login to controller node via eth0 (public network) for executing this script. Other NIC will lost connectivity. and make sure hostname resolv at /etc/hosts. in this sample, your host need resolv self fqdn in 10.200.9.10
Network Node's network interfaces
Set up NICs for network node.
network% sudo ${EDITOR} /etc/network/interfaces
# The loopback network interface
auto lo
iface lo inet loopback
# for API network
auto eth0
iface eth0 inet static
up ifconfig $IFACE 0.0.0.0 up
up ip link set $IFACE promisc on
down ip link set $IFACE promisc off
down ifconfig $IFACE down
address 10.200.9.11
netmask 255.255.255.0
dns-nameservers 8.8.8.8 8.8.4.4
dns-search example.com
# for VM traffic to the internet
auto eth1
iface eth1 inet static
address 172.16.1.11
netmask 255.255.255.0
# for management network
auto eth2
iface eth2 inet static
address 10.200.10.11
netmask 255.255.255.0
gateway 10.200.10.1
dns-nameservers 8.8.8.8 8.8.4.4
dns-search example.com
and login to network node via eth2 (management network) for executing this script. Other NIC will lost connectivity.
Compute Node's network interfaces
Set up NICs for network node.
compute% sudo ${EDITOR} /etc/network/interfaces
# The loopback network interface
auto lo
iface lo inet loopback
# for VM traffic to the internet
auto eth0
iface eth0 inet static
address 172.16.1.12
netmask 255.255.255.0
# for management network
auto eth1
iface eth1 inet static
address 10.200.10.12
netmask 255.255.255.0
gateway 10.200.10.1
dns-nameservers 8.8.8.8 8.8.4.4
dns-search example.com
and login to compute node via eth2 (mangement network) for executing this script. Other NIC will lost connectivity.
Run script
Run this script, all of conpornents will be built.
controller% sudo ./setup.sh controller
network % sudo ./setup.sh network
compute % sudo ./setup.sh comupte
That's all and You've done. :D Now you can access to Horizon (http://${CONTROLLER_NODE_PUB_IP}/horizon/) with user 'demo', password 'demo'.
Additional Compute Node
If you want to have additional compute node(s), please setup network interfaces as noted before for compute node and execute these commands.
Edit setup.conf (COPUTE_NODE_IP parameter) and execute setup.sh.
compute % scp -r ~/openstack_grizzly_install <add_compute_node>:~/
add_compute% cd openstack_grizzly_install
add_compute% ${EDITOR} setup.conf
COMPUTE_NODE_IP='<your additional compute node's ip>'
add_compute% sudo ./setup.sh compute
add_compute% sudo nova-manage service list # check nodes list
Additional Network Node
If you want to have additional network node(s), please setup network interfaces as noted before for network node and execute these commands.
Edit setup.conf (NETWORK_NODE_IP parameter) and execute setup.sh.
network % scp -r ~/openstack_grizzly_install <add_network_node>:~/
add_network% cd openstack_grizzly_install
add_network% ${EDITOR} setup.conf
NETWORK_NODE_IP='<your additional network node's ip>'
add_network% sudo ./setup.sh network
add_network% source ~/openstackrc
add_network% quantum agent-list # check agent list
How to use on separated nodes mode with nova-network
Architecture
+-------------+-------------+------------------------------ Public/API Network
| | |
+-----------+ +-----------+ +-----------+
| | |vm|vm|.. | |vm|vm|.. |
| controller| +-----------+ +-----------+
| | | compute | | compute |
| | | | | additional|
+-----------+ +-----------+ +-----------+
| | |
+-------------+-------------+------------------------------ Management Network
- minimum architecture : 2 nodes (controller node x 1, compute node x1)
- additional compute node(s) make you be able to have more VMs.
Get this script
git clone this script from github on controller node.
controller% git clone git://github.com/jedipunkz/openstack_grizzly_install.git
controller% cd openstack_grizzly_install
controller% cp setup.conf.samples/setup.conf.separated.nova-network setup.conf
Edit parameters on setup.conf
There are many paramaters on setup.conf, but in 'allinone' mode, parameters which you need to edit is such things.
CONTROLLER_NODE_IP='10.200.10.10'
CONTROLLER_NODE_PUB_IP='10.200.9.10'
COMPUTE_NODE_IP='10.200.10.11'
FLAT_INTERFACE='eth0'
NETWORK_COMPONENT='nova-network'
If you want to change other parameters such as DB password, admin password, please change these.
copy to other nodes
copy directory to network node and compute node.
controller% scp -r openstack_grizzly_install <network_node_ip>:~/
controller% scp -r openstack_grizzly_install <compute_node_ip>:~/
Controller Node's network interfaces
Set up NICs for controller node.
controller% sudo ${EDITOR} /etc/network/interfaces
# The loopback network interface
auto lo
iface lo inet loopback
# for API network
auto eth0
iface eth0 inet static
address 10.200.9.10
netmask 255.255.255.0
gateway 10.200.9.1
dns-nameservers 8.8.8.8 8.8.4.4
dns-search example.com
# for management network
auto eth1
iface eth1 inet static
address 10.200.10.10
netmask 255.255.255.0
dns-nameservers 8.8.8.8 8.8.4.4
dns-search example.com
Compute Node's network interfaces
Set up NICs for network node.
compute% sudo ${EDITOR} /etc/network/interfaces
# The loopback network interface
auto lo
iface lo inet loopback
# for API network
auto eth0
iface eth0 inet static
address 10.200.9.11
netmask 255.255.255.0
gateway 10.200.9.1
dns-nameservers 8.8.8.8 8.8.4.4
dns-search example.com
# for management network
auto eth1
iface eth1 inet static
address 10.200.10.11
netmask 255.255.255.0
dns-nameservers 8.8.8.8 8.8.4.4
dns-search example.com
and login to compute node via eth2 (mangement network) for executing this script. Other NIC will lost connectivity.
Run script
Run this script, all of conpornents will be built.
controller% sudo ./setup.sh controller
compute % sudo ./setup.sh comupte
That's all and You've done. :D Now you can access to Horizon (http://${CONTROLLER_NODE_PUB_IP}/horizon/) with user 'demo', password 'demo'.
Additional Compute Node
If you want to have additional compute node(s), please setup network interfaces as noted before for compute node and execute these commands.
Edit setup.conf (COPUTE_NODE_IP parameter) and execute setup.sh.
compute % scp -r ~/openstack_grizzly_install <add_compute_node>:~/
add_compute% cd openstack_grizzly_install
add_compute% ${EDITOR} setup.conf
COMPUTE_NODE_IP='<your additional compute node's ip>'
add_compute% sudo ./setup.sh compute
add_compute% sudo nova-manage service list # check nodes list
Parameters
These are Meaning of parameters.
- HOST_IP : IP addr on management network with 'allinone' node
- HOST_PUB_IP : IP addr on public network with 'allinone' node
- PUBLIC_NIC : NIC name on public network with 'allinone' node
- CONTROLLER_NODE_IP : IP addr on management network with controller node
- CONTROLLER_NODE_PUB_IP : IP addr on public network with controller node
- NETWORK_NODE_IP : IP addr on management network with network node
- COMPUTE_NODE_IP : IP addr on management network with compute node
- DATA_NIC_CONTROLLER : NIC name on data network with controller node
- DATA_NIC_COMPUTE : NIC name on data network with compute nod
- PUBLIC_NIC : NIC name on public network on network node
- CINDER_VOLUME : Disk device name for Cinder Volume
- MYSQL_PASS : root password of MySQL
- DB_KEYSTONE_USER : MySQL user for Keystone
- DB_KEYSTONE_PASS : MySQL password for Keystone
- DB_GLANCE_USER : MySQL user for Glance
- DB_GLANCEPASS : MySQL password for Glance
- DB_QUANTUM_USER : MySQL user for Quantum
- DB_QUANTUM_PASS : MySQL password for Quantum
- DB_NOVA_USER : MySQL user for Nova
- DB_NOVA_PASS : MySQL password for Nova
- DB_CINDER_USER : MySQL user for Cinder
- DB_CINDER_PASS : MySQL password for CInder
- ADMIN_PASSWORD : Keystone password for admin user
- SERVICE_PASSWORD : Keystone password for service user
- OS_TENANT_NAME : OS tenant name
- OS_USERNAME : OS username
- OS_PASSWORD : OS password
- DEMO_USER : first user for DEMO
- DEMO_PASSWORD : first user's password for DEMO
- INT_NET_GATEWAY : Gateway address of internal network
- INT_NET_RANGE : Range of external network
- EXT_NET_GATEWAY : Gateway address of external network
- EXT_NET_START : Starging address of external network
- EXT_NET_END : Ending address of external network
- EXT_NET_RANGE : Range of external network
- OS_IMAGE_URL : URL for downloading OS image file
- OS_IMAGE_NAME : Name of OS IMAGE name for Glance service
Licensing
This Script is licensed under a Creative Commons Attribution 3.0 Unported License.
To view a copy of this license, visit [ http://creativecommons.org/licenses/by/3.0/deed.en_US ].
Credits
This work has been based on: mseknibilel's guide.
https://github.com/mseknibilel/OpenStack-Grizzly-Install-Guide
Version and Change log
- 2013/06/19 : version 0.6 : supported nova-network
- 2013/05/27 : version 0.5 : supported vlan mode
- 2013/05/24 : version 0.4 : enabled to use loopback device for cinder.
- 2013/05/13 : version 0.3 : enabled nova live-migration
- 2013/04/18 : version 0.2 : enabled LBaaS, fixed a problem to access metadata server from vm.
- 2013/04/17 : version 0.1 : First release.