前言：
openstack的部署非常简单，简单的前提建立在扎实的理论功底，本人一直觉得，玩技术一定是理论指导实践，网上遍布个种搭建方法都可以实现一个基本的私有云环境，但是诸位可曾发现，很多配置都是重复的，为何重复？到底什么位置该不该配？具体配置什么参数？很多作者本人都搞不清楚，今天本人就是要在这里正本清源。
介绍：本次案列为基本的三节点部署，集群案列后期有时间再整理
一：网络：
1.管理网络：172.16.209.0/24
2.数据网络：1.1.1.0/24
二：操作系统：CentOS Linux release 7.2.1511 (Core)
三：内核：3.10.0-327.el7.x86_64
四：openstack版本mitaka
效果图：
OpenStack mitaka部署
约定：
1.在修改配置的时候，切勿在某条配置后加上注释，可以在配置的上面或者下面加注释
2.相关配置一定是在标题后追加，不要在原有注释的基础上修改
PART1：环境准备（在所有节点执行）
一：
每台机器设置固定ip，每台机器添加hosts文件解析，为每台机器设置主机名，关闭firewalld，selinux
可选操作：在控制节点制作密钥登录其他节点（可以方便后面的操作，真实环境也极有必要准备一台单独的管理机），在控制节点修改/etc/hosts并scp到其他节点
/etc/hosts 127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4 ::1 localhost localhost.localdomain localhost6 localhost6.localdomain6 172.16.209.115 controller01 172.16.209.117 compute01 172.16.209.119 network02
二：获取软件包源（在所有节点执行），即配置yum源，下述两种方式请按个人情况选择一种，推荐方式一
方式一：自定义的yum源
本人从官网下载包后自定制的yum源，自定制源的好处是：严格地控制软件包版本，保持平台内主机版本的一致性和可控性，具体做法如下
- 找一个服务器，作为yum源（同时也可以作为cobbler或pxe主机）
- 上传openstack-mitaka-rpms.tar.gz
- tar xvf openstack-mitaka-rpms.tar.gz -C /
- 在这台机器上安装httpd并且启动，设置开机启动
- ln -s /mitaka-rpms /var/www/html/
然后每台机器配置yum源
[mitaka] name=mitaka repo baseurl=http://172.16.209.100/mitaka-rpms/ enabled=1 gpgcheck=0
方式二：下载安装官网的源
自定义yum源的包其实也都是来自于官网，只不过，官网经常会更新包，一个包的更新可能会导致一些版本兼容性问题，所以推荐使用方式一，但如果只是测试而非生产环境，方式二是一种稍微便捷的方式
基于centos系统，在所有节点上执行：
yum install centos-release-openstack-mitaka -y
基于redhat系统，在所有节点上执行：
yum install https://rdoproject.org/repos/rdo-release.rpm -y #红帽系统请去掉epel源，否则会跟这里的源有冲突
三 制作yum缓存并更新系统（在所有节点执行）
yum makecache && yum install vim net-tools -y&& yum update -y
小知识点：
yum -y update
升级所有包，改变软件设置和系统设置,系统版本内核都升级
yum -y upgrade
升级所有包，不改变软件设置和系统设置，系统版本升级，内核不改变
四 关闭yum自动更新（在所有节点执行）
CentOS7最小化安装后默认yum会自动下载更新，这对许多生产系统是不需要的，可以手动关闭它
[[email protected] cron.weekly]# cd /etc/yum [[email protected] yum]# ls fssnap.d pluginconf.d protected.d vars version-groups.conf yum-cron.conf yum-cron-hourly.conf
编辑yum-cron.conf，将download_updates = yes改为no即可
ps：yum install yum-plugin-priorities -y #如果不想关闭自动更新，那么可以安装这个插件，来设定升级的优先级，从官网下载更新而不是从一些乱七八糟的第三方源
五 预装包（在所有节点执行）
yum install python-openstackclient -y yum install openstack-selinux -y
六 时间服务部署
yum install chrony -y #（在所有节点执行）
控制节点：
修改配置：
/etc/chrony.conf
server ntp.staging.kycloud.lan iburst
allow 管理网络网段ip/24
启服务：
systemctl enable chronyd.service systemctl start chronyd.service
其余节点：
修改配置：
/etc/chrony.conf
server 控制节点ip iburst
启服务
systemctl enable chronyd.service
systemctl start chronyd.service
时区不是Asia/Shanghai需要改时区：
# timedatectl set-local-rtc 1 # 将硬件时钟调整为与本地时钟一致, 0 为设置为 UTC 时间
# timedatectl set-timezone Asia/Shanghai # 设置系统时区为上海
其实不考虑各个发行版的差异化, 从更底层出发的话, 修改时间时区比想象中要简单:
# cp /usr/share/zoneinfo/Asia/Shanghai /etc/localtime
验证：
每台机器执行：
chronyc sources
在S那一列包含*号，代表同步成功（可能需要花费几分钟去同步，时间务必同步）
七：部署mariadb数据库
控制节点：
yum install mariadb mariadb-server python2-PyMySQL -y
编辑：
/etc/my.cnf.d/openstack.cnf
[mysqld] bind-address = 控制节点管理网络ip default-storage-engine = innodb innodb_file_per_table max_connections = 4096 collation-server = utf8_general_ci character-set-server = utf8
启服务：
systemctl enable mariadb.service systemctl start mariadb.service mysql_secure_installation
八：为Telemetry 服务部署MongoDB
控制节点：
yum install mongodb-server mongodb -y
编辑：/etc/mongod.conf
bind_ip = 控制节点管理网络ip smallfiles = true
启动服务：
systemctl enable mongod.service systemctl start mongod.service
九：部署消息队列rabbitmq(验证方式:http://172.16.209.104:15672/ 用户:guest 密码:guest)
控制节点：
yum install rabbitmq-server -y
启动服务：
systemctl enable rabbitmq-server.service systemctl start rabbitmq-server.service
新建rabbitmq用户密码：
rabbitmqctl add_user openstack che001
为新建的用户openstack设定权限：
rabbitmqctl set_permissions openstack ".*" ".*" ".*"
十：部署memcached缓存（为keystone服务缓存tokens）
控制节点：
yum install memcached python-memcached -y
启动服务：
systemctl enable memcached.service systemctl start memcached.service
PART2：认证服务keystone部署
一：安装和配置服务
1.建库建用户
mysql -u root -p CREATE DATABASE keystone; GRANT ALL PRIVILEGES ON keystone.* TO ‘keystone‘@‘localhost‘ IDENTIFIED BY ‘che001‘; GRANT ALL PRIVILEGES ON keystone.* TO ‘keystone‘@‘%‘ IDENTIFIED BY ‘che001‘; flush privileges;
2.yum install openstack-keystone httpd mod_wsgi -y
3.编辑/etc/keystone/keystone.conf
[DEFAULT] admin_token = che001 #建议用命令制作token:openssl rand -hex 10 [database] connection = mysql+pymysql://keystone:[email protected]/keystone [token] provider = fernet #Token Provider：UUID, PKI, PKIZ, or Fernet #http://blog.csdn.net/miss_yang_cloud/article/details/49633719
4.同步修改到数据库
su -s /bin/sh -c "keystone-manage db_sync" keystone
5.初始化fernet keys
keystone-manage fernet_setup --keystone-user keystone --keystone-group keystone
6.配置apache服务
编辑：/etc/httpd/conf/httpd.conf
ServerName controller01
编辑：/etc/httpd/conf.d/wsgi-keystone.conf
新增配置
Listen 5000
Listen 35357
<VirtualHost *:5000>
WSGIDaemonProcess keystone-public processes=5 threads=1 user=keystone group=keystone display-name=%{GROUP}
WSGIProcessGroup keystone-public
WSGIScriptAlias / /usr/bin/keystone-wsgi-public
WSGIApplicationGroup %{GLOBAL}
WSGIPassAuthorization On
ErrorLogFormat "%{cu}t %M"
ErrorLog /var/log/httpd/keystone-error.log
CustomLog /var/log/httpd/keystone-access.log combined
<Directory /usr/bin>
Require all granted
</Directory>
</VirtualHost>
<VirtualHost *:35357>
WSGIDaemonProcess keystone-admin processes=5 threads=1 user=keystone group=keystone display-name=%{GROUP}
WSGIProcessGroup keystone-admin
WSGIScriptAlias / /usr/bin/keystone-wsgi-admin
WSGIApplicationGroup %{GLOBAL}
WSGIPassAuthorization On
ErrorLogFormat "%{cu}t %M"
ErrorLog /var/log/httpd/keystone-error.log
CustomLog /var/log/httpd/keystone-access.log combined
<Directory /usr/bin>
Require all granted
</Directory>
</VirtualHost>
7.启动服务:
systemctl enable httpd.service
systemctl restart httpd.service #因为之前自定义基于http协议的yum源时已经启动过了httpd，所以此处需要restart
二：创建服务实体和访问端点
1.实现配置管理员环境变量，用于获取后面创建的权限
export OS_TOKEN=che001 export OS_URL=http://controller01:35357/v3 export OS_IDENTITY_API_VERSION=3
2.基于上一步给的权限，创建认证服务实体（目录服务）
export OS_TOKEN=che001 export OS_URL=http://controller01:35357/v3 export OS_IDENTITY_API_VERSION=3
--name keystone --description "OpenStack Identity" identity
3.基于上一步建立的服务实体，创建访问该实体的三个api端点
openstack endpoint create --region RegionOne identity public http://controller01:5000/v3 openstack endpoint create --region RegionOne identity internal http://controller01:5000/v3 openstack endpoint create --region RegionOne identity admin http://controller01:35357/v3
三：创建域，租户，用户，角色，把四个元素关联到一起
建立一个公共的域名：
openstack domain create --description "Default Domain" default
管理员：admin
openstack project create --domain default --description "Admin Project" admin openstack user create --domain default --password-prompt admin openstack role create admin openstack role add --project admin --user admin admin
普通用户：demo
openstack project create --domain default --description "Demo Project" demo openstack user create --domain default --password-prompt demo openstack role create user openstack role add --project demo --user demo user
为后续的服务创建统一租户service
解释：后面每搭建一个新的服务都需要在keystone中执行四种操作：1.建租户 2.建用户 3.建角色 4.做关联
后面所有的服务公用一个租户service，都是管理员角色admin，所以实际上后续的服务安装关于keysotne
的操作只剩2，4
openstack project create --domain default --description "Service Project" service
四：验证操作：
编辑：/etc/keystone/keystone-paste.ini
在[pipeline:public_api], [pipeline:admin_api], and [pipeline:api_v3] 三个地方 移走：admin_token_auth unset OS_TOKEN OS_URL
openstack --os-auth-url http://controller01:35357/v3 --os-project-domain-name default --os-user-domain-name default --os-project-name admin --os-username admin token issue Password: +------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+ | Field | Value | +------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+ | expires | 2016-08-17T08:29:18.528637Z | | id | gAAAAABXtBJO-mItMcPR15TSELJVB2iwelryjAGGpaCaWTW3YuEnPpUeg799klo0DaTfhFBq69AiFB2CbFF4CE6qgIKnTauOXhkUkoQBL6iwJkpmwneMo5csTBRLAieomo4z2vvvoXfuxg2FhPUTDEbw-DPgponQO-9FY1IAEJv_QV1qRaCRAY0 | | project_id | 9783750c34914c04900b606ddaa62920 | | user_id | 8bc9b323a3b948758697cb17da304035 | +------------+-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------+
五：新建客户端脚本文件
管理员：admin-openrc
export OS_PROJECT_DOMAIN_NAME=default export OS_USER_DOMAIN_NAME=default export OS_PROJECT_NAME=admin export OS_USERNAME=admin export OS_PASSWORD=che001 export OS_AUTH_URL=http://controller01:35357/v3 export OS_IDENTITY_API_VERSION=3 export OS_IMAGE_API_VERSION=2
普通用户demo：demo-openrc
export OS_PROJECT_DOMAIN_NAME=default export OS_USER_DOMAIN_NAME=default export OS_PROJECT_NAME=demo export OS_USERNAME=demo export OS_PASSWORD=che001 export OS_AUTH_URL=http://controller01:5000/v3 export OS_IDENTITY_API_VERSION=3 export OS_IMAGE_API_VERSION=2
效果：
source admin-openrc [[email protected] ~]# openstack token issue
part3：部署镜像服务
一：安装和配置服务
1.建库建用户
mysql -u root -p CREATE DATABASE glance; GRANT ALL PRIVILEGES ON glance.* TO ‘glance‘@‘localhost‘ IDENTIFIED BY ‘che001‘; GRANT ALL PRIVILEGES ON glance.* TO ‘glance‘@‘%‘ IDENTIFIED BY ‘che001‘; flush privileges;
2.keystone认证操作：
上面提到过：所有后续项目的部署都统一放到一个租户service里，然后需要为每个项目建立用户，建管理员角色，建立关联
. admin-openrc
openstack user create --domain default --password-prompt glance openstack role add --project service --user glance admin
建立服务实体
openstack service create --name glance --description "OpenStack Image" image
建端点
openstack endpoint create --region RegionOne image public http://controller01:9292 openstack endpoint create --region RegionOne image internal http://controller01:9292 openstack endpoint create --region RegionOne image admin http://controller01:9292
3.安装软件
yum install openstack-glance -y
4.初始化存放镜像的存储设备，此处我们暂时用本地存储，但是无论哪种存储，都应该在glance启动前就已经有了，否则glance启动时通过驱动程序检索不到存储设备，即在glance启动后 新建的存储设备无法被glance识别到，需要重新启动glance才可以，因此我们将下面的步骤提到了最前面。
新建目录：
mkdir -p /var/lib/glance/images/ chown glance. /var/lib/glance/images/
5.修改配置：
编辑：/etc/glance/glance-api.conf
[database]
#这里的数据库连接配置是用来初始化生成数据库表结构,不配置无法生成数据库表结构
#glance-api不配置database对创建vm无影响，对使用metada有影响
#日志报错:ERROR glance.api.v2.metadef_namespaces
connection = mysql+pymysql://glance:[email protected]/glance [keystone_authtoken] auth_url = http://controller01:5000 memcached_servers = controller01:11211 auth_type = password project_domain_name = default user_domain_name = default project_name = service username = glance password = che001 [paste_deploy] flavor = keystone [glance_store] stores = file,http default_store = file filesystem_store_datadir = /var/lib/glance/images/
编辑：/etc/glance/glance-registry.conf
[database]
#这里的数据库配置是用来glance-registry检索镜像元数据
connection = mysql+pymysql://glance:[email protected]/glance
同步数据库：（此处会报一些关于future的问题，自行忽略）
su -s /bin/sh -c "glance-manage db_sync" glance
启动服务：
systemctl enable openstack-glance-api.service openstack-glance-registry.service systemctl start openstack-glance-api.service openstack-glance-registry.service
二：验证操作：
. admin-openrc
wget http://download.cirros-cloud.net/0.3.4/cirros-0.3.4-x86_64-disk.img
（本地下载：wget http://172.16.209.100/cirros-0.3.4-x86_64-disk.img）
openstack image create "cirros" --file cirros-0.3.4-x86_64-disk.img --disk-format qcow2 --container-format bare --public openstack image list
part4：部署compute服务
一：控制节点配置
1.建库建用户
CREATE DATABASE nova_api; CREATE DATABASE nova; GRANT ALL PRIVILEGES ON nova_api.* TO ‘nova‘@‘localhost‘ IDENTIFIED BY ‘che001‘; GRANT ALL PRIVILEGES ON nova_api.* TO ‘nova‘@‘%‘ IDENTIFIED BY ‘che001‘; GRANT ALL PRIVILEGES ON nova.* TO ‘nova‘@‘localhost‘ IDENTIFIED BY ‘che001‘; GRANT ALL PRIVILEGES ON nova.* TO ‘nova‘@‘%‘ IDENTIFIED BY ‘che001‘; flush privileges;
2.keystone相关操作
. admin-openrc
openstack user create --domain default --password-prompt nova openstack role add --project service --user nova admin openstack service create --name nova --description "OpenStack Compute" compute openstack endpoint create --region RegionOne compute public http://controller01:8774/v2.1/%\(tenant_id\)s openstack endpoint create --region RegionOne compute internal http://controller01:8774/v2.1/%\(tenant_id\)s openstack endpoint create --region RegionOne compute admin http://controller01:8774/v2.1/%\(tenant_id\)s
3.安装软件包：
yum install openstack-nova-api openstack-nova-conductor openstack-nova-console openstack-nova-novncproxy \openstack-nova-scheduler -y
4.修改配置：
编辑/etc/nova/nova.conf
[DEFAULT] enabled_apis = osapi_compute,metadata rpc_backend = rabbit auth_strategy = keystone #下面的为管理ip my_ip = 172.16.209.115 use_neutron = True firewall_driver = nova.virt.firewall.NoopFirewallDriver [api_database] connection = mysql+pymysql://nova:[email protected]/nova_api [database] connection = mysql+pymysql://nova:[email protected]/nova [oslo_messaging_rabbit] rabbit_host = controller01 rabbit_userid = openstack rabbit_password = che001 [keystone_authtoken] auth_url = http://controller01:5000 memcached_servers = controller01:11211 auth_type = password project_domain_name = default user_domain_name = default project_name = service username = nova password = che001 [vnc] #下面的为管理ip vncserver_listen = 172.16.209.115 #下面的为管理ip vncserver_proxyclient_address = 172.16.209.115 [oslo_concurrency] lock_path = /var/lib/nova/tmp
5.同步数据库：（此处会报一些关于future的问题，自行忽略）
su -s /bin/sh -c "nova-manage api_db sync" nova su -s /bin/sh -c "nova-manage db sync" nova
6.启动服务
systemctl enable openstack-nova-api.service openstack-nova-consoleauth.service openstack-nova-scheduler.service openstack-nova-conductor.service openstack-nova-novncproxy.service systemctl start openstack-nova-api.service openstack-nova-consoleauth.service openstack-nova-scheduler.service openstack-nova-conductor.service openstack-nova-novncproxy.service
二：计算节点配置
1.安装软件包：
yum install openstack-nova-compute libvirt-daemon-lxc -y
2.修改配置：
编辑/etc/nova/nova.conf
[DEFAULT] rpc_backend = rabbit auth_strategy = keystone #计算节点管理网络ip my_ip = 172.16.209.117 use_neutron = True firewall_driver = nova.virt.firewall.NoopFirewallDriver [oslo_messaging_rabbit] rabbit_host = controller01 rabbit_userid = openstack rabbit_password = che001 [vnc] enabled = True vncserver_listen = 0.0.0.0 #计算节点管理网络ip vncserver_proxyclient_address = 172.16.209.117 #控制节点管理网络ip novncproxy_base_url = http://172.16.209.115:6080/vnc_auto.html [glance] api_servers = http://controller01:9292 [oslo_concurrency] lock_path = /var/lib/nova/tmp
3.如果在不支持虚拟化的机器上部署nova，请确认
egrep -c ‘(vmx|svm)‘ /proc/cpuinfo结果为0
则编辑/etc/nova/nova.conf
[libvirt] virt_type = qemu
4.启动服务
systemctl enable libvirtd.service openstack-nova-compute.service systemctl start libvirtd.service openstack-nova-compute.service
三：验证
控制节点
[[email protected] ~]# source admin-openrc [[email protected] ~]# openstack compute service list +----+------------------+--------------+----------+---------+-------+----------------------------+ | Id | Binary | Host | Zone | Status | State | Updated At | +----+------------------+--------------+----------+---------+-------+----------------------------+ | 1 | nova-consoleauth | controller01 | internal | enabled | up | 2016-08-17T08:51:37.000000 | | 2 | nova-conductor | controller01 | internal | enabled | up | 2016-08-17T08:51:29.000000 | | 8 | nova-scheduler | controller01 | internal | enabled | up | 2016-08-17T08:51:38.000000 | | 12 | nova-compute | compute01 | nova | enabled | up | 2016-08-17T08:51:30.000000 |
part5：部署网络服务
一：控制节点配置
1.建库建用户
CREATE DATABASE neutron; GRANT ALL PRIVILEGES ON neutron.* TO ‘neutron‘@‘localhost‘ IDENTIFIED BY ‘che001‘; GRANT ALL PRIVILEGES ON neutron.* TO ‘neutron‘@‘%‘ IDENTIFIED BY ‘che001‘; flush privileges;
2.keystone相关
. admin-openrc
openstack user create --domain default --password-prompt neutron openstack role add --project service --user neutron admin openstack service create --name neutron --description "OpenStack Networking" network openstack endpoint create --region RegionOne network public http://controller01:9696 openstack endpoint create --region RegionOne network internal http://controller01:9696 openstack endpoint create --region RegionOne network admin http://controller01:9696
3.安装软件包
yum install openstack-neutron openstack-neutron-ml2 python-neutronclient which -y
4.配置服务器组件
编辑 /etc/neutron/neutron.conf文件,并完成以下动作:
在[数据库]节中,配置数据库访问: [DEFAULT] core_plugin = ml2 service_plugins = router #下面配置:启用重叠IP地址功能 allow_overlapping_ips = True rpc_backend = rabbit #auth_strategy = keystone notify_nova_on_port_status_changes = True notify_nova_on_port_data_changes = True [oslo_messaging_rabbit] rabbit_host = controller01 rabbit_userid = openstack rabbit_password = che001 [database] connection = mysql+pymysql://neutron:[email protected]/neutron [keystone_authtoken] auth_url = http://controller01:5000 memcached_servers = controller01:11211 auth_type = password project_domain_name = default user_domain_name = default project_name = service username = neutron password = che001 [nova] auth_url = http://controller01:5000 auth_type = password project_domain_name = default user_domain_name = default region_name = RegionOne project_name = service username = nova password = che001 [oslo_concurrency] lock_path = /var/lib/neutron/tmp
编辑/etc/neutron/plugins/ml2/ml2_conf.ini文件
[ml2] type_drivers = flat,vlan,vxlan,gre tenant_network_types = vxlan mechanism_drivers = openvswitch,l2population extension_drivers = port_security [ml2_type_flat] flat_networks = provider [ml2_type_vxlan] vni_ranges = 1:1000 [securitygroup] enable_ipset = True
编辑/etc/nova/nova.conf文件：
[neutron] url = http://controller01:9696 auth_url = http://controller01:5000 auth_type = password project_domain_name = default user_domain_name = default region_name = RegionOne project_name = service username = neutron password = che001 service_metadata_proxy = True
5.创建连接
ln -s /etc/neutron/plugins/ml2/ml2_conf.ini /etc/neutron/plugin.ini
6.同步数据库：（此处会报一些关于future的问题，自行忽略）
su -s /bin/sh -c "neutron-db-manage --config-file /etc/neutron/neutron.conf --config-file /etc/neutron/plugins/ml2/ml2_conf.ini upgrade head" neutron
7.重启nova服务
systemctl restart openstack-nova-api.service
8.启动neutron服务
systemctl enable neutron-server.service systemctl start neutron-server.service
二：网络节点配置
1. 编辑 /etc/sysctl.conf
net.ipv4.ip_forward=1 net.ipv4.conf.all.rp_filter=0 net.ipv4.conf.default.rp_filter=0
2.执行下列命令，立即生效
sysctl -p
3.安装软件包
yum install openstack-neutron openstack-neutron-ml2 openstack-neutron-openvswitch -
4．配置组件
编辑/etc/neutron/neutron.conf文件
[DEFAULT] core_plugin = ml2 service_plugins = router allow_overlapping_ips = True rpc_backend = rabbit auth_strategy = keystone [database] connection = mysql+pymysql://neutron:[email protected]/neutron [oslo_messaging_rabbit] rabbit_host = controller01 rabbit_userid = openstack rabbit_password = che001 [oslo_concurrency] lock_path = /var/lib/neutron/tmp
6、编辑 /etc/neutron/plugins/ml2/openvswitch_agent.ini文件:
[ovs] #下面ip为网络节点数据网络ip [agent] tunnel_types=gre,vxlan #l2_population=True prevent_arp_spoofing=True
7.配置L3代理。编辑 /etc/neutron/l3_agent.ini文件:
[DEFAULT] interface_driver=neutron.agent.linux.interface.OVSInterfaceDriver external_network_bridge=br-ex
8.配置DHCP代理。编辑 /etc/neutron/dhcp_agent.ini文件:
[DEFAULT] interface_driver=neutron.agent.linux.interface.OVSInterfaceDriver dhcp_driver=neutron.agent.linux.dhcp.Dnsmasq enable_isolated_metadata=True
9.配置元数据代理。编辑 /etc/neutron/metadata_agent.ini文件:
[DEFAULT] nova_metadata_ip=controller01 metadata_proxy_shared_secret=che001
10.启动服务(先启动服务再建网桥br-ex)
网路节点：
systemctl start neutron-openvswitch-agent.service neutron-l3-agent.service neutron-dhcp-agent.service neutron-metadata-agent.service systemctl enable neutron-openvswitch-agent.service neutron-l3-agent.service neutron-dhcp-agent.service neutron-metadata-agent.service
11.建网桥
注意，如果网卡数量有限，想用网路节点的管理网络网卡作为br-ex绑定的物理网卡
#那么需要将网络节点管理网络网卡ip去掉，建立br-ex的配置文件，ip使用原管理网ip
[[email protected] ~]# cat /etc/sysconfig/network-scripts/ifcfg-eth0 DEVICE=eth0 TYPE=Ethernet ONBOOT="yes" BOOTPROTO="none" NM_CONTROLLED=no [[email protected] ~]# cat /etc/sysconfig/network-scripts/ifcfg-br-ex DEVICE=br-ex TYPE=Ethernet ONBOOT="yes" BOOTPROTO="none" #HWADDR=bc:ee:7b:78:7b:a7 IPADDR=172.16.209.10 GATEWAY=172.16.209.1 NETMASK=255.255.255.0 DNS1=202.106.0.20 DNS1=8.8.8.8 NM_CONTROLLED=no #注意加上这一句否则网卡可能启动不成功 ovs-vsctl add-br br-ex ovs-vsctl add-port br-ex eth2 #要在network服务重启前将物理端口eth0加入网桥br-ex
systemctl restart network #重启网络时，务必保证eth2网卡没有ip或者干脆是down掉的状态，并且一定要NM_CONTROLLED=no，否则会无法启动服务
三：计算节点配置
1. 编辑 /etc/sysctl.conf
net.ipv4.conf.all.rp_filter=0 net.ipv4.conf.default.rp_filter=0
2.sysctl -p
3.yum install openstack-neutron openstack-neutron-ml2 openstack-neutron-openvswitch -y
4.编辑 /etc/neutron/neutron.conf文件
[DEFAULT] rpc_backend = rabbit #auth_strategy = keystone [oslo_messaging_rabbit] rabbit_host = controller01 rabbit_userid = openstack rabbit_password = che001 [oslo_concurrency] lock_path = /var/lib/neutron/tmp
5．编辑 /etc/neutron/plugins/ml2/openvswitch_agent.ini
[ovs] #下面ip为计算节点数据网络ip local_ip = 1.1.1.117 #bridge_mappings = vlan:br-vlan [agent] tunnel_types = gre,vxlan l2_population = True #开启l2_population功能用于接收sdn控制器（一般放在控制节点）发来的（新建的vm）arp信息，这样就把arp信息推送到了每个中断设备（计算节点），减少了一大波初识arp广播流量（说初始是因为如果没有l2pop机制，一个vm对另外一个vm的arp广播一次后就缓存到本地了），好强大，详见https://assafmuller.com/2014/05/21/ovs-arp-responder-theory-and-practice/ arp_responder = True #开启br-tun的arp响应功能，这样br-tun就成了一个arp proxy，来自本节点对其他虚拟机而非物理主机的arp请求可以基于本地的br-tun轻松搞定，不能再牛逼了 prevent_arp_spoofing = True [securitygroup] firewall_driver = neutron.agent.linux.iptables_firewall.OVSHybridIptablesFirewallDriver enable_security_group = True 7.编辑 /etc/nova/nova.conf [neutron] url = http://controller01:9696 auth_url = http://controller01:5000 auth_type = password project_domain_name = default user_domain_name = default region_name = RegionOne project_name = service username = neutron password = che001
8．启动服务
systemctl enable neutron-openvswitch-agent.service systemctl start neutron-openvswitch-agent.service systemctl restart openstack-nova-compute.service
part6:部署控制面板dashboard
在控制节点
1.安装软件包
yum install openstack-dashboard -y
2.配置/etc/openstack-dashboard/local_settings
OPENSTACK_HOST = "controller01"
ALLOWED_HOSTS = [‘*‘, ]
SESSION_ENGINE = ‘django.contrib.sessions.backends.cache‘
CACHES = {
‘default‘: {
‘BACKEND‘: ‘django.core.cache.backends.memcached.MemcachedCache‘,
‘LOCATION‘: ‘controller01:11211‘,
}
}
#注意：必须是v3而不是v3.0
OPENSTACK_KEYSTONE_URL = "http://%s:5000/v3" % OPENSTACK_HOST
OPENSTACK_KEYSTONE_MULTIDOMAIN_SUPPORT = True
OPENSTACK_API_VERSIONS = {
"identity": 3,
"image": 2,
"volume": 2,
}
OPENSTACK_KEYSTONE_DEFAULT_DOMAIN = "default"
OPENSTACK_KEYSTONE_DEFAULT_ROLE = "user"
TIME_ZONE = "UTC"
3.启动服务
systemctl enable httpd.service memcached.service systemctl restart httpd.service memcached.service
4.验证；
http://172.16.209.115/dashboard
总结：
- 与keystone打交道的只有api层，所以不要到处乱配
- 建主机的时候由nova-compute负责调用各个api，所以不要再控制节点配置啥调用
- ml2是neutron的core plugin，只需要在控制节点配置
- 网络节点只需要配置相关的agent
- 各组件的api除了接收请求外还有很多其他功能，比方说验证请求的合理性，控制节点nova.conf需要配neutron的api、认证，因为nova boot时需要去验证用户提交网络的合理性，控制节点neutron.conf需要配nova的api、认证，因为你删除网络端口时需要通过nova-api去查是否有主机正在使用端口。计算几点nova.conf需要配neutron，因为nova-compute发送请求给neutron-server来创建端口。这里的端口值得是‘交换机上的端口‘
- 不明白为啥？或者不懂我在说什么，请好好研究openstack各组件通信机制和主机创建流程。
网路故障排查：
网络节点：
[[email protected] ~]# ip netns show qdhcp-e63ab886-0835-450f-9d88-7ea781636eb8 qdhcp-b25baebb-0a54-4f59-82f3-88374387b1ec qrouter-ff2ddb48-86f7-4b49-8bf4-0335e8dbaa83 [[email protected] ~]# ip netns exec qrouter-ff2ddb48-86f7-4b49-8bf4-0335e8dbaa83 bash [[email protected] ~]# ping -c2 www.baidu.com PING www.a.shifen.com (61.135.169.125) 56(84) bytes of data. 64 bytes from 61.135.169.125: icmp_seq=1 ttl=52 time=33.5 ms 64 bytes from 61.135.169.125: icmp_seq=2 ttl=52 time=25.9 ms
如果无法ping通，那么退出namespace
ovs-vsctl del-br br-ex
ovs-vsctl del-br br-int
ovs-vsctl del-br br-tun
ovs-vsctl add-br br-int
ovs-vsctl add-br br-ex
ovs-vsctl add-port br-ex eth0
systemctl restart neutron-openvswitch-agent.service neutron-l3-agent.service \
neutron-dhcp-agent.service neutron-metadata-agent.service