S1_搭建分布式OpenStack集群_02 虚拟机环境配置

一、配置主机名及hosts和防火墙（只演示一台，其他机器配置方式一样）
vim /etc/hosts  加入    
10.10.11.11 controller
10.10.11.12 compute1
10.10.11.13 neutron
10.10.11.14 cinder

修改主机名
hostnamectl set-hostname controller
hostnamectl set-hostname compute1
hostnamectl set-hostname neutron
hostnamectl set-hostname cinder

关闭防火墙    
systemctl stop firewalld.service
systemctl disable firewalld.service

关闭网络管理NetworkManager
systemctl stop NetworkManager.service
systemctl disable NetworkManager.service

关闭selinux
# vim /etc/selinux/config    
SELINUX=disabled

二、所有主机配置网卡，这里只演示controller
2.1 配置eth0 (仅限controller)
controller ~]# cd /etc/sysconfig/network-scripts/
controller network-scripts]# vim ifcfg-eth0
DEVICE="eth0"
ONBOOT="yes"
IPV6INIT="no"
IPV4_FAILURE_FATAL="no"
NM_CONTROLLED="no"
TYPE="Ethernet"
BOOTPROTO="static"
IPADDR="10.10.10.11"
PREFIX="24"
GATEWAY="10.10.10.1"

controller network-scripts]# systemctl restart network
controller network-scripts]# ifconfig eth0
eth0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 10.10.10.11  netmask 255.255.255.0  broadcast 10.10.10.255
        inet6 fe80::5054:ff:fec3:e842  prefixlen 64  scopeid 0x20<link>
        ether 52:54:00:c3:e8:42  txqueuelen 1000  (Ethernet)
        RX packets 1821  bytes 95276 (93.0 KiB)
        RX errors 0  dropped 1785  overruns 0  frame 0
        TX packets 92  bytes 6164 (6.0 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

2.2 配置eth1
controller network-scripts]# cp ifcfg-eth0 ifcfg-eth1
controller network-scripts]# vim ifcfg-eth1
DEVICE="eth1"
ONBOOT="yes"
IPV6INIT="no"
IPV4_FAILURE_FATAL="no"
NM_CONTROLLED="no"
TYPE="Ethernet"
BOOTPROTO="static"
IPADDR="10.10.11.11"
PREFIX="24"
(GATEWAY="10.10.11.1" 非controller节点都要添加,controller不加)

controller network-scripts]# systemctl restart network
controller network-scripts]# ifconfig eth1
eth1: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 10.10.11.11  netmask 255.255.255.0  broadcast 10.10.11.255
        inet6 fe80::5054:ff:fe21:6a50  prefixlen 64  scopeid 0x20<link>
        ether 52:54:00:21:6a:50  txqueuelen 1000  (Ethernet)
        RX packets 2505  bytes 132478 (129.3 KiB)
        RX errors 0  dropped 2353  overruns 0  frame 0
        TX packets 57  bytes 3674 (3.5 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

2.3 配置eth2
controller network-scripts]# cp ifcfg-eth1 ifcfg-eth2
controller network-scripts]# vim ifcfg-eth2
DEVICE="eth2"
ONBOOT="yes"
IPV6INIT="no"
IPV4_FAILURE_FATAL="no"
NM_CONTROLLED="no"
TYPE="Ethernet"
BOOTPROTO="static"
IPADDR="10.10.12.11"
PREFIX="24"

controller network-scripts]# systemctl restart network
controller network-scripts]# ifconfig eth2
eth2: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 10.10.12.11  netmask 255.255.255.0  broadcast 10.10.12.255
        inet6 fe80::5054:ff:feaa:a8a  prefixlen 64  scopeid 0x20<link>
        ether 52:54:00:aa:0a:8a  txqueuelen 1000  (Ethernet)
        RX packets 2434  bytes 127542 (124.5 KiB)
        RX errors 0  dropped 2372  overruns 0  frame 0
        TX packets 64  bytes 3648 (3.5 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

2.4 查看DNS
controller network-scripts]# cat /etc/resolv.conf
# Generated by NetworkManager
nameserver 10.10.10.1

ping以下百度：证明eth0网卡已启用
controller network-scripts]# ping www.baidu.com
PING www.a.shifen.com (61.135.169.121) 56(84) bytes of data.
64 bytes from 61.135.169.121 (61.135.169.121): icmp_seq=1 ttl=54 time=22.0 ms
64 bytes from 61.135.169.121 (61.135.169.121): icmp_seq=2 ttl=54 time=22.3 ms

--- www.a.shifen.com ping statistics ---
2 packets transmitted, 2 received, 0% packet loss, time 1001ms
rtt min/avg/max/mdev = 22.099/22.201/22.304/0.180 ms

2.5 检查controller网卡：

2.6 所有节点配置完成后，检测节点互通：

三、配置集群yum源
3.1 yum节点网卡配置：
yum ~]# vim /etc/sysconfig/network-scripts/ifcfg-eth0
DEVICE="eth0"
ONBOOT="yes"
IPV6INIT="no"
IPV4_FAILURE_FATAL="no"
NM_CONTROLLED="no"
TYPE="Ethernet"
BOOTPROTO="static"
IPADDR="10.10.11.200"
PREFIX="24"
GATEWAY="10.10.11.1"

yum ~]# systemctl restart network
yum ~]# ifconfig eth0
eth0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST>  mtu 1500
        inet 10.10.11.200  netmask 255.255.255.0  broadcast 10.10.11.255
        inet6 fe80::5054:ff:fefa:2719  prefixlen 64  scopeid 0x20<link>
        ether 52:54:00:fa:27:19  txqueuelen 1000  (Ethernet)
        RX packets 227  bytes 12116 (11.8 KiB)
        RX errors 0  dropped 215  overruns 0  frame 0
        TX packets 66  bytes 3732 (3.6 KiB)
        TX errors 0  dropped 0 overruns 0  carrier 0  collisions 0

ping controller节点：
yum ~]# ping 10.10.11.11
PING 10.10.11.11 (10.10.11.11) 56(84) bytes of data.
64 bytes from 10.10.11.11: icmp_seq=1 ttl=64 time=0.339 ms
64 bytes from 10.10.11.11: icmp_seq=2 ttl=64 time=0.255 ms
--- 10.10.11.11 ping statistics ---
2 packets transmitted, 2 received, 0% packet loss, time 1000ms
rtt min/avg/max/mdev = 0.255/0.297/0.339/0.042 ms

yum ~]# ping 10.10.11.12
PING 10.10.11.12 (10.10.11.12) 56(84) bytes of data.
64 bytes from 10.10.11.12: icmp_seq=1 ttl=64 time=0.249 ms

--- 10.10.11.12 ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 0.249/0.249/0.249/0.000 ms

3.2 搭建FTP本地yum源：
下载163源到本地 http://mirrors.163.com/centos/7.6.1810/
yum ~]# ls /var/ftp/
dvd  every  extrals  mini  pub    queens    updates
这样，除了pub文件夹之外，其余都是本地yum源了

yum ~]# vim /etc/yum.repos.d/all.repo
[dvd]
name=dvd
baseurl=ftp://10.10.11.200/dvd
gpgcheck=1
enable=0
priority=3

[every]
name=every
baseurl=ftp://10.10.11.200/every
gpgcheck=0
enable=0
priority=3

[mini]
name=mini
baseurl=ftp://10.10.11.200/mini
gpgcheck=0
enable=0
priority=3

[queens]
name=queens
baseurl=ftp://10.10.11.200/queens
gpgcheck=0
enable=0
priority=1

[extrals]
name=extrals
baseurl=ftp://10.10.11.200/extrals
gpgcheck=0
enable=0
priority=3

[updates]
name=updates
baseurl=ftp://10.10.11.200/updates
gpgcheck=0
enable=0
priority=2

查看yum：
yum ~]# yum clean all
yum ~]# yum repolist | grep repolist
repolist: 18,246

3.3 其他主机同步yum源
controller ~]# rm -rf /etc/yum.repos.d/*
compute1 ~]#   rm -rf /etc/yum.repos.d/*
neutron ~]#    rm -rf /etc/yum.repos.d/*
cinder ~]#     rm -rf /etc/yum.repos.d/*

yum ~]# cd /etc/yum.repos.d/
yum.repos.d]# ls
all.repo  repo  RPM-GPG-KEY-CentOS-7  RPM-GPG-KEY-CentOS-Testing-7

yum.repos.d]# for i in 1 2 3 4
> do
> scp all.repo RPM-GPG-KEY-CentOS-* root@10.10.11.1$i:/etc/yum.repos.d/
> done

登陆其他任何一个节点查看：
neutron ~]# yum clean all
neutron ~]# yum repolist | grep repolist
repolist: 20,097

由于怕yum源冲突及保持openstack queen yum源的优先权,每个节点安装 yum-plugin-priorities
controller ~]# yum -y install yum-plugin-priorities
其他节点也安装.

四、时间同步配置
在所有节点安装chrony
# yum install chrony –y

4.1 配置controller为时间同步服务器
controller ~]# vim /etc/chrony.conf
更新这几处：
server ntp1.aliyun.com iburst
allow 0/0
local stratum 10
controller ~]# systemctl restart chronyd
controller ~]# chronyc sources -v
^* 120.25.115.20
出现 “^*” 表示同步成功。

4.2 配置其他节点同步controller的时间同步服务（以compute为例)
compute1 ~]# vim /etc/chrony.conf
修改为：
server 10.10.11.11 iburst
 
compute1 ~]# chronyc sources -v
^* controller...
出现 “^*” 表示同步成功。

4.3 将所有节点的chronyd服务设为开机自启（以controller为例)
controller ~]# systemctl enable chronyd

五、安装openstack软件仓库(所有服务器)
yum安装Q版仓库(以controller为例）
controller ~]# yum -y install centos-release-openstack-queens

安装open stack client端:
controller ~]# yum install python-openstackclient -y

安装open stack的selinux（如果关闭了服务器selienx可以不装）
controller ~]# yum install openstack-selinux -y
由于我提前关闭了，所以此处我没有安装。