keepalived 安装

Keepalived的下载在官方网站 http://www.keepalived.org/

一、安装

1、首先安装依赖包

查看系统版本
cat /etc/redhat-release
CentOS Linux release 7.2.1511 (Core)

查看内核版本

ls /usr/src/kernels
3.10.0-693.2.2.el7.x86_64

# yum install -y libnl*
# yum install -y libnfnetlink-devel zlib zlib-devel gcc gcc-c++ openssl openssl-devel openssh

# yum install -y libnl-devel libnl3-devel libnfnetlink-devel

pcre: 用来作地址重写的功能。
zlib：nginx 的gzip模块,传输数据打包，省流量（但消耗资源）。
openssl：提供ssl加密协议

2、下载并解压Keepalived
# wget http://www.keepalived.org/software/keepalived-2.2.4.tar.gz
# tar xvf keepalived-2.2.4.tar.gz

3、编译安装
# cd keepalived-2.2.4
# ./configure --prefix=/home/keepalived
# make
# make install

二、配置
keepalived安装完成后默认不会注册为系统服务，所以需要手工添加系统服务脚本。在/etc/init.d目录下新建keepalived文件，并更改权限其即可。

0、创建keepalived的pid目录

　　用于后面手动添加进程号

[root@Nginx-Master ~]# mkdir -p /usr/local/keepalived/var/run/

[root@Nginx-Master ~]# vi /usr/local/keepalived/var/run/keepalived.pid 

　　建立Nginx用户

[root@Nginx-Master nginx-1.9.7]# groupadd -g 1001 nginx
[root@Nginx-Master nginx-1.9.7]# useradd -u 900 nginx -g nginx -s /sbin/nologin
[root@Nginx-Master nginx-1.9.7]# tail -1 /etc/passwd
nginx:x:900:1001::/home/nginx:/sbin/nologin

1、去 keepalived 源码目录下拷贝启动脚本到/etc/init.d/下
[root@bogon keepalived-1.3.5]# pwd
/usr/local/src/keepalived-1.3.5
[root@bogon keepalived-1.3.5]# cp /usr/local/src/keepalived-1.3.5/keepalived/etc/init.d/keepalived    /etc/init.d/keepalived
[root@bogon keepalived-1.3.5]# ll /etc/init.d/ | grep keepalived
-rwxr-xr-x 1 root root 1308 8月 2 14:15 keepalived

2、修改/etc/init.d/keepalived启动脚本
　　方式1、需要修改的数据有以下几项：

# config: /etc/keepalived/keepalived.conf 文件的位置；
# Source configuration file (we set KEEPALIVED_OPTIONS there) 文件的位置。. /etc/sysconfig/keepalived
修改为实际安装的文件路径，我的安装路径如下：

# config: /usr/local/keepalived/etc/keepalived/keepalived.conf
. /usr/local/keepalived/etc/sysconfig/keepalived
另外添加如下变量，并修改文件中相应的部分：

keepalived=/usr/local/keepalived/sbin/keepalived
keepalived_config=/usr/local/keepalived/etc/keepalived/keepalived.conf
keepalived_pid=/usr/local/keepalived/var/run/keepalived.pid
[root@bogon keepalived-1.3.5]# vim /etc/init.d/keepalived
#!/bin/sh
#
# Startup script for the Keepalived daemon
#
# processname: keepalived
# pidfile: /usr/local/keepalived/run/keepalived.pid
# config: /usr/local/keepalived/etc/keepalived/keepalived.conf
# chkconfig: - 21 79
# description: Start and stop Keepalived

# Source function library
. /etc/rc.d/init.d/functions

# Source configuration file (we set KEEPALIVED_OPTIONS there)
. /usr/local/keepalived/etc/sysconfig/keepalived

keepalived=/usr/local/keepalived/sbin/keepalived
keepalived_config=/usr/local/keepalived/etc/keepalived/keepalived.conf
keepalived_pid=/usr/local/keepalived/var/run/keepalived.pid

RETVAL=0

prog="keepalived"

start() {
echo -n $"Starting $prog: "
daemon $keepalived -f ${keepalived_config} ${KEEPALIVED_OPTIONS}
RETVAL=$?
echo
[ $RETVAL -eq 0 ] && touch /var/lock/subsys/$prog
}

2、修改其权限并开机启动

修改权限：chmod 755  /etc/init.d/keepalived
加为系统服务：chkconfig  --add keepalived
开机启动：chkconfig keepalived on  (centos6.0)
查看开机启动的服务：chkconfig  --list

3、备注：keepalived服务控制
systemctl enable keepalived.service #设置开机自动启动  (centos7.0)
systemctl disable keepalived.service #取消开机自动启动
systemctl start keepalived.service #启动服务
systemctl restart keepalived.service #重启服务
systemctl stop keepalived.service #停止服务
systemctl status keepalived.service #查看服务状态

　　方式2、另一种添加为系统服务的方式
1、复制相应文件到相应目录下
# keepalived启动脚本变量引用文件，默认文件路径是/etc/sysconfig/，也可以不做软链接，直接修改启动脚本中文件路径即可（安装目录下）
[root@bogon /]# cp /usr/local/keepalived/etc/sysconfig/keepalived      /etc/sysconfig/keepalived

# 将keepalived主程序加入到环境变量（安装目录下）
[root@bogon /]# cp /usr/local/keepalived/sbin/keepalived      /usr/sbin/keepalived

# keepalived启动脚本（源码目录下），放到/etc/init.d/目录下就可以使用service命令便捷调用
[root@bogon /]# cp /usr/local/src/keepalived-1.3.5/keepalived/etc/init.d/keepalived     /etc/init.d/keepalived

# 将配置文件放到默认路径下
[root@bogon /]# mkdir /etc/keepalived
[root@bogon /]# cp /usr/local/keepalived/etc/keepalived/keepalived.conf     /etc/keepalived/keepalived.conf

[root@bogon /]# cp /etc/keepalived/keepalived.conf /etc/keepalived/keepalived.conf.bak

2、设置开机启动
# 设置开机启动
[root@bogon /]# chkconfig keepalived on   (centos6.0)

[root@bogon /]# service keepalived start|stop|restart  (centos7.0)

附、安装过程中的问题

0、vrrp无法正常识别物理网卡？

vrrp_instance VI_1 {
state MASTER
interface ens160    #需根据不同的系统进行修改，默认是eth0
virtual_router_id 51
mcast_src_ip 192.168.2.146
priority 100
advert_int 1
authentication {
auth_type PASS
auth_pass 1111
}
virtual_ipaddress {
# 192.168.200.16
#192.168.200.17
# 192.168.200.18
192.168.2.147

}

1、修改启动脚本 /etc/init.d/keepalived 后，未生效？
对于此问题一般的解决方法是执行如下命令，重新加载脚本。

[root@bogon sbin]# systemctl daemon-reload

以上方法大多时候会生效，但也有例外。此时，可以试试如下方法：

删除路径 /usr/lib/systemd/system/ 下相应的启动脚本

[root@bogon /]# chkconfig --del keepalived
[root@bogon /]# cd /usr/lib/systemd/system/
[root@bogon system]# ls | grep keepalived
keepalived.service
[root@bogon system]# rm keepalived.service
rm：是否删除普通文件 "keepalived.service"？y

然后重新注册为系统服务：

[root@bogon system]# chkconfig --add keepalived

重新注册后，修改后的配置文件就会生效。

2、a configured resource limit was exceeded？
执行 systemctl start keepalived.service 后出现如下错误：

[root@bogon sbin]# systemctl start keepalived.service

Job for keepalived.service failed because a configured resource limit was exceeded. See "systemctl status keepalived.service" and "journalctl -xe" for details.
这个问题在网上找了半天也没找可以解决的方法，但是把 问题1 解决之后 问题2 就没再出现

3、安装时出现如下问题？

缺少依赖包，安装一下就好：

[root@bogon /]# yum install -y libnfnetlink-devel

4、Failed to start LVS and VRRP High Availability Monitor？

处理方法：需要将进程号手动写入pid文件

处理步骤：查看keepalived的进程

[root@rbtnode1 ~]# ps aux | grep keepalived
root 10960 0.0 0.0 48028 1044 ? Ss 09:01 0:00 /usr/local/keepalived/sbin/keepalived -D
root 10961 0.0 0.1 48028 1920 ? S 09:01 0:00 /usr/local/keepalived/sbin/keepalived -D
root 10962 0.0 0.0 48028 1516 ? S 09:01 0:00 /usr/local/keepalived/sbin/keepalived -D
root 19427 0.0 0.0 112664 972 pts/0 S+ 10:06 0:00 grep --color=auto keepalived

将进程号手动写入pid文件

[root@rbtnode1 ~]# echo 10960 >> /usr/local/keepalived/var/run/keepalived.pid

5、如果vrrp无法切换，需要关闭防火墙？

systemctl stop firewalld.service #停止firewall

systemctl disable firewalld.service #禁止firewall开机启动

6、无法ping通vip地址？

vim /etc/keepalived/keepalived.conf

#vrrp_strict    #在全局配置下面

  

7、centos7.0系统下添加多个虚拟ip地址

 有时你也许想要给一个网卡多个地址。你该怎么做呢？另外买一个网卡来分配地址？在小型网络中其实不用这么做。我们现在可以在CentOS/RHEL 7中给一个网卡分配多个ip地址。想知道怎么做么？好的，跟随我，这并不难。

 首先，让我们找到网卡的IP地址。在我的CentOS 7服务器中，我只使用了一个网卡。

用root特权运行下面的命令：

1. ip addr

示例输出：

1. 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN
2. link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
3. inet 127.0.0.1/8 scope host lo
4. valid_lft forever preferred_lft forever
5. inet6 ::1/128 scope host
6. valid_lft forever preferred_lft forever
7. 2: enp0s3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
8. link/ether 08:00:27:80:63:19 brd ff:ff:ff:ff:ff:ff
9. inet 192.168.1.150/24 brd 192.168.1.255 scope global enp0s3
10. valid_lft forever preferred_lft forever

如上所见，我的网卡名是enp0s3，ip地址是192.168.1.150。

如你所知，网卡的配置文件存储在 /etc/sysconfig/network-scripts/ 目录下。每个网卡的详细内容将会以不同的名字存储，比如ifcfg-enp0s3。

让我们看下ifcfg-enp0s3的细节。

1. cat /etc/sysconfig/network-scripts/ifcfg-enp0s3

示例输出：

1. TYPE="Ethernet"
2. BOOTPROTO="none"
3. DEFROUTE="yes"
4. IPV4_FAILURE_FATAL="no"
5. IPV6INIT="yes"
6. IPV6_AUTOCONF="yes"
7. IPV6_DEFROUTE="yes"
8. IPV6_FAILURE_FATAL="no"
9. NAME="enp0s3"
10. UUID="e9f9caef-cb9e-4a19-aace-767c6ee6f849"
11. ONBOOT="yes"
12. HWADDR="08:00:27:80:63:19"
13. IPADDR0="192.168.1.150"
14. PREFIX0="24"
15. GATEWAY0="192.168.1.1"
16. DNS1="192.168.1.1"
17. IPV6_PEERDNS="yes"
18. IPV6_PEERROUTES="yes"

好的，现在我们将在相同的子网中分配多个地址了。

编辑文件 /etc/sysconfig/network-scripts/ifcfg-enp0s3：

1. vi /etc/sysconfig/network-scripts/ifcfg-enp0s3

像下面那样加入额外的IP地址。

1. TYPE="Ethernet"
2. BOOTPROTO="none"
3. DEFROUTE="yes"
4. IPV4_FAILURE_FATAL="no"
5. IPV6INIT="yes"
6. IPV6_AUTOCONF="yes"
7. IPV6_DEFROUTE="yes"
8. IPV6_FAILURE_FATAL="no"
9. NAME="enp0s3"
10. UUID="933cdc9b-b383-4ddd-b219-5a72c69c9cf0"
11. ONBOOT="yes"
12. HWADDR="08:00:27:3F:AB:68"
13. IPADDR0="192.168.1.150"
14. IPADDR1="192.168.1.151"
15. IPADDR2="192.168.1.152"
16. PREFIX0="24"
17. GATEWAY0="192.168.1.1"
18. DNS1="192.168.1.1"
19. IPV6_PEERDNS="yes"
20. IPV6_PEERROUTES="yes"

如你所见，我已经加了两个IP地址：IPADDR1=”192.168.1.151″ & IPADDR2=”192.168.1.152″

类似地，你可以加入更多的ip地址。

最后，保存并退出文件。重启网络服务来使更改生效。

1. systemctl restart network

现在，让我们检查是否已经加入了ip地址。

1. ip addr

示例输出：

1. : lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN
2. link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
3. inet 127.0.0.1/8 scope host lo
4. valid_lft forever preferred_lft forever
5. inet6 ::1/128 scope host
6. valid_lft forever preferred_lft forever
7. 2: enp0s3: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
8. link/ether 08:00:27:3f:ab:68 brd ff:ff:ff:ff:ff:ff
9. inet 192.168.1.150/24 brd 192.168.1.255 scope global enp0s3
10. valid_lft forever preferred_lft forever
11. inet 192.168.1.151/24 brd 192.168.1.255 scope global secondary enp0s3
12. valid_lft forever preferred_lft forever
13. inet 192.168.1.152/24 brd 192.168.1.255 scope global secondary enp0s3
14. valid_lft forever preferred_lft forever
15. inet6 fe80::a00:27ff:fe3f:ab68/64 scope link
16. valid_lft forever preferred_lft forever

如你所见，单个网卡已经有3个ip地址了。

让我们ping一下新增的IP地址：

1. ping -c 4 192.168.1.151

示例输出：

1. PING 192.168.1.151 (192.168.1.151) 56(84) bytes of data.
2. 64 bytes from 192.168.1.151: icmp_seq=1 ttl=64 time=0.048 ms

如果你想要使用不同的子网，你要改变PREFIX0=24成不同的子网，比如 PREFIX3=16。

比如，我想要添加一个A类地址（*比如10.0.0.1）到我的网卡中。

1. TYPE="Ethernet"
2. BOOTPROTO="none"
3. DEFROUTE="yes"
4. IPV4_FAILURE_FATAL="no"
5. IPV6INIT="yes"
6. IPV6_AUTOCONF="yes"
7. IPV6_DEFROUTE="yes"
8. IPV6_FAILURE_FATAL="no"
9. NAME="enp0s3"
10. UUID="933cdc9b-b383-4ddd-b219-5a72c69c9cf0"
11. ONBOOT="yes"
12. HWADDR="08:00:27:3F:AB:68"
13. IPADDR0="192.168.1.150"
14. IPADDR1="192.168.1.151"
15. IPADDR2="192.168.1.152"
16. IPADDR3="10.0.0.1"
17. PREFIX0="24"
18. PREFIX3=16
19. GATEWAY0="192.168.1.1"
20. DNS1="192.168.1.1"
21. IPV6_PEERDNS="yes"
22. IPV6_PEERROUTES="yes"

你可以看到我已经添加一个A类地址(10.0.0.1)并且前缀是16。

保存并退出文件。重启网络服务，接着，ping新增的地址：

1. ping -c 4 10.0.0.1

 8、开机后keepalived无法正常启动，因为pid号会变化，导致每次需手动更新如下文件？

[root@rbtnode1 ~]# > /usr/local/keepalived/var/run/keepalived.pid

9、关闭nginx进程后，keepalived无法正常切换？

可以改成监控物理接口，监控nginx进程还需修改shell脚本，未完待续！

强制关闭进程建议用如下命令：

ps -ef | grep keepalived

root 3775 1 0 14:05 ? 00:00:00 /usr/local/sbin/keepalived -D
root 3776 3775 0 14:05 ? 00:00:00 /usr/local/sbin/keepalived -D
root 3777 3775 0 14:05 ? 00:00:00 /usr/local/sbin/keepalived -D
root 8257 1269 0 14:41 pts/0 00:00:00 grep --color=auto keepalived

kill -s 9  3775