潇湘夜雨移动版

三、安装配置Nova

 

3.1 准备工作

 

3.1.1 启用网络接口的PROMISC功能

 

本示例将基于FlatDHCP网络模型，因此其所有工作可以仅通过一个网络接口完成。在所有的节点上，需要为此接口开启“promisc”标志，这可通过如下命令实现。

 

# ip link set eth0 promisc on

 

确认其已经启用“PROMISC”功能。

# ip link show eth0

[root@node1 stackimages(keystone_admin)]# ip link show eth0

2: eth0: <BROADCAST,MULTICAST,PROMISC,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000

link/ether 00:0c:29:f8:c2:31 brd ff:ff:ff:ff:ff:ff

[root@node1 stackimages(keystone_admin)]# service libvirtd status

libvirtd (pid  10765) 正在运行...

[root@node1 stackimages(keystone_admin)]# virsh iface-bridge eth0 br100

Created bridge br100 with attached device eth0

Bridge interface br100 started

 

[root@node1 stackimages(keystone_admin)]# ip a

1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN 

link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00

inet 127.0.0.1/8 scope host lo

inet6 ::1/128 scope host 

  valid_lft forever preferred_lft forever

2: eth0: <BROADCAST,MULTICAST,PROMISC,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000

link/ether 00:0c:29:f8:c2:31 brd ff:ff:ff:ff:ff:ff

523: br-ex: <BROADCAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN 

link/ether 4e:09:99:ec:c4:48 brd ff:ff:ff:ff:ff:ff

inet6 fe80::a078:2aff:fe49:154a/64 scope link 

  valid_lft forever preferred_lft forever

704: br100: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN 

link/ether 00:0c:29:f8:c2:31 brd ff:ff:ff:ff:ff:ff

inet 172.18.109.235/26 brd 172.18.109.255 scope global br100

inet6 fe80::20c:29ff:fef8:c231/64 scope link 

  valid_lft forever preferred_lft forever

461: ovs-system: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN 

link/ether 86:5b:81:7f:46:1f brd ff:ff:ff:ff:ff:ff

462: br-int: <BROADCAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state UNKNOWN 

link/ether a2:e8:7f:79:95:49 brd ff:ff:ff:ff:ff:ff

inet6 fe80::d0ba:3dff:fe48:83cf/64 scope link 

  valid_lft forever preferred_lft forever

查看桥接接口的信息。

# brctl show

bridge name bridge id STP enabled interfaces

br100 8000.000c29ed4c10 no eth0

virbr0 8000.5254003bee3a yes virbr0-nic

 

3.1.2 启动messagbus服务

# service messagebus start

# chkconfig messagebus on

[root@node1 stackimages(keystone_admin)]# service messagebus status

messagebus (pid  1590) 正在运行...

 

3.2 安装nova

# yum install openstack-utils memcached qpid-cpp-server 

# yum install openstack-nova

 

初始化nova数据库，同时创建其服务同名的用户，并为其指定密码，这里选择使用与服务名同名的密码nova。

[root@node1 stackimages(keystone_admin)]# openstack-db --init --service nova --password nova

nova default DB is not mysql. Would you like to reset to mysql now? (y/n): y

Please enter the password for the 'root' MySQL user: 

Verified connectivity to MySQL.

Creating 'nova' database.

Initializing the nova database, please wait...

Complete!

如果不想使用默认的nova用户访问其数据，也可以使用下面的命令为nova创建数据库及相关的用户。

# mysql -uroot -p

mysql> CREATE DATABASE nova;

mysql> GRANT ALL ON nova.* TO 'NOVA_USER'@'%' IDENTIFIED BY 'NOVA_PASS';

mysql> GRANT ALL ON nova.* TO 'NOVA_USER'@'localhost' IDENTIFIED BY 'NOVA_PASS';

mysql> FLUSH PRIVILEGES;

 

 

3.3 配置nova

 

3.3.1 编辑配置文件/etc/nova/nova.conf，修改其内容如下所示：

 

[DEFAULT]

my_ip=172.18.109.235

state_path=/var/lib/nova

network_manager=nova.network.manager.FlatDHCPManager

api_paste_config=/etc/nova/api-paste.ini

auth_strategy=keystone

novncproxy_host=0.0.0.0

novncproxy_port=6080

instance_name_template=instance-%08x

dhcpbridge_flagfile=/etc/nova/nova.conf

public_interface=eth0

dhcpbridge=/usr/bin/nova-dhcpbridge

flat_network_bridge=br100

flat_interface=eth0

vlan_interface=eth0

fixed_range=172.18.109.192/26

force_dhcp_release=True

log_dir=/var/log/nova

rpc_backend=nova.openstack.common.rpc.impl_qpid

qpid_hostname=172.18.109.235

qpid_port=5672

firewall_driver=nova.virt.libvirt.firewall.IptablesFirewallDriver

libvirt_type=qemu

live_migration_uri=qemu+tcp://%s/system

live_migration_flag=VIR_MIGRATE_UNDEFINE_SOURCE, VIR_MIGRATE_PEER2PEER

novncproxy_base_url=http://172.18.109.235:6080/vnc_auto.html

vncserver_listen=172.18.109.235

vncserver_proxyclient_address=172.18.109.235

vnc_enabled=true

connection=mysql://nova:nova@172.18.109.235/nova

volume_driver=nova.virt.baremetal.volume_driver.LibvirtVolumeDriver

[keystone_authtoken]

admin_tenant_name=service

admin_user=nova

admin_password=nova

auth_host=172.18.109.235

auth_port=35357

auth_protocol=http

auth_version=v2.0

signing_dir=/var/lib/nova/keystone-signing

 

如果要开启api还需添加认证方式和密码到文件/etc/nova/api-paste.ini 

[filter:authtoken]

paste.filter_factory = keystoneclient.middleware.auth_token:filter_factory

auth_host = 172.18.109.235

auth_port = 35357

auth_protocol = http

auth_uri = http://172.18.109.235:5000/v2.0

admin_tenant_name = service

admin_user = nova

admin_password = nova

 

说明：KVM虚拟化要求CPU支持svm或vmx，因此，这里的Openstack所在的系统RHEL6.4必须直接运行于支持这两种特性之一的CPU平台上。如果整个过程是在虚拟机上进行测试的话，比如Openstack所安装的系统RHEL6.4本身就运行于虚拟上，那么，虚拟化技术只能使用qemu而不能使用KVM。此种情形下，需要执行如下步骤，在RHEL6.4启动对qemu的支持。

(1)安装libguestfs-tools

# yum -y install libguestfs-tools

 

(2)设定libvirt类型为qemu

openstack-config --set /etc/nova/nova.conf DEFAULT libvirt_type qemu

 

(3)为qemu-kvm创建所需要链接

# ln -sv  /usr/libexec/qemu-kvm /usr/bin/qemu

 

(4)重启libvirtd服务

# service libvirtd restart

 

3.3.2 导入或迁移nova数据库

# nova-manage db sync

 

3.2.3 安装配置qpid队列服务

 

# yum install -y qpid-cpp-server

# sed -i -e 's/auth=.*/auth=no/g' /etc/qpidd.conf

# service qpidd start 

# chkconfig qpidd on

 

3.3.4 启动服务

首先创建nova服务的锁文件目录，并将其权限赋予nova用户。

# mkdir /var/lock/nova

# chown -R nova.nova /var/lock/nova/

而后启动nova相关的服务，并将其设置为开机自动启动。

# for svc in api compute network cert console scheduler; do service openstack-nova-$svc start; chkconfig openstack-nova-$svc on; done

 

查看服务启动状况：

[root@node1 ~(keystone_admin)]# nova-manage service list

Binary           Host                                 Zone             Status     State Updated_At

nova-console     node1.test.com                       internal         enabled    :-)   2016-10-06 05:23:09 (责任编辑：liangzh)