2. 程式人生 > >Keepalived+Nginx實現高可用負載均衡集群

Keepalived+Nginx實現高可用負載均衡集群

阿新 • • 發佈：2017-05-15
連接 靜態 adf -1 rip mail fff hostname dex

一 環境介紹

1.操作系統
CentOS Linux release 7.2.1511 (Core)

2.服務
keepalived+lvs雙主高可用負載均衡集群及LAMP應用
keepalived-1.2.13-7.el7.x86_64
nginx-1.10.2-1.el7.x86_64
httpd-2.4.6-45.el7.centos.x86_64


二 原理及拓撲圖

1.vrrp協議
在現實的網絡環境中，兩臺需要通信的主機大多數情況下並沒有直接的物理連接。對於這樣的情況，它們之間路由怎樣選擇？主機如何選定到達目的主機的下一跳路由，這個問題通常的解決方法有二種：
在主機上使用動態路由協議(RIP、OSPF等)
在主機上配置靜態路由
很明顯，在主機上配置動態路由是非常不切實際的，因為管理、維護成本以及是否支持等諸多問題。配置靜態路由就變得十分流行，但路由器(或者說默認網關default gateway)卻經常成為單點故障。VRRP的目的就是為了解決靜態路由單點故障問題，VRRP通過一競選(election)協議來動態的將路由任務交給LAN中虛擬路由器中的某臺VRRP路由器。

2.nginx反代
nginx是以反向代理的方式進行負載均衡的。反向代理（Reverse Proxy）方式是指以代理服務器來接受Internet上的連接請求，然後將請求轉發給內部網絡上的服務器，並將從服務器上得到的結果返回給Internet上請求連接的客戶端，此時代理服務器對外就表現為一個服務器。(為了理解反向代理，這裏插播一條什麽是正向代理：正向代理指的是，一個位於客戶端和原始服務器之間的服務器，為了從原始服務器取得內容，客戶端向代理發送一個請求並指定目標(原始服務器)，然後代理向原始服務器轉交請求並將獲得的內容返回給客戶端。)
3.拓撲圖

技術分享


三 配置

1.後端RS配置

[[email protected]
/* */ ~]# yum install httpd -y [[email protected] ~]# yum install httpd -y

2.Nginx反代配置
MASTER:

upstream websrvs {
 server 172.18.67.11:80;
 server 172.18.67.12:80;
 server 127.0.0.1:80 backup;
}
server {
    listen       80 ;
    location / {
    proxy_pass http://websrvs;
    }

BACKUP:

upstream websrvs {
 server 172.18.67.11:80;
 server 172.18.67.12:80;
 server 127.0.0.1:80 backup;
}
server {
    listen       80 ;
    location / {
    proxy_pass http://websrvs;
    }

3.keepalived高可用配置
MASTER:

! Configuration File for keepalived
global_defs {
    notification_email {
 [email protected]
    }
    notification_email_from [email protected]
    smtp_server 127.0.0.1
    smtp_connect_timeout 30
    router_id node1
    vrrp_mcast_group4 224.0.67.67
}
vrrp_script chk_down {
    script "[[ -f /etc/keepalived/down ]] && exit 1 || exit 0"
    interval 1
    weight -5
}
vrrp_script chk_nginx {
    script "killall -0 nginx && exit 0 || exit 1"
    interval 1
    weight -5
    fall 2
    rise 1
}
vrrp_instance myr {
    state MASTER
    interface eno16777736
    virtual_router_id 167
    priority 100
    advert_int 1
    authentication {
 auth_type PASS
 auth_pass 571f97b2
    }
    virtual_ipaddress {
 172.18.67.33/16 dev eno16777736
    }
    track_script {
 chk_down
 chk_nginx
    }
    notify_master "/etc/keepalived/notify.sh master"
    notify_backup "/etc/keepalived/notify.sh backup"
    notify_fault "/etc/keepalived/notify.sh fault"
}

BACKUP:

! Configuration File for keepalived
global_defs {
    notification_email {
 [email protected]
    }
    notification_email_from [email protected]
    smtp_server 127.0.0.1
    smtp_connect_timeout 30
    router_id node1
    vrrp_mcast_group4 224.0.67.67
}
vrrp_script chk_down {
    script "[[ -f /etc/keepalived/down ]] && exit 1 || exit 0"
    interval 1
    weight -5
}
vrrp_script chk_nginx {
    script "killall -0 nginx && exit 0 || exit 1"
    interval 1
    weight -5
    fall 2
    rise 1
}
vrrp_instance myr {
    state BACKUP
    interface eno16777736
    virtual_router_id 167
    priority 95
    advert_int 1
    authentication {
 auth_type PASS
 auth_pass 571f97b2
    }
    virtual_ipaddress {
 172.18.67.33/16 dev eno16777736
    }
    track_script {
 chk_down
 chk_nginx
    }
    notify_master "/etc/keepalived/notify.sh master"
    notify_backup "/etc/keepalived/notify.sh backup"
    notify_fault "/etc/keepalived/notify.sh fault"
}

4.通知腳本示例

[[email protected] nginx]# vim notify.sh
#!/bin/bash
#
[email protected]
notify() {
 mailsubject="$(hostname) to be $1, vip floating"
 mailbody="$(date +‘%F %T‘): vrrp transition, $(hostname) changed to be $1"
 echo "$mailbody" | mail -s "$mailsubject" $contact
}
case $1 in
master)
 notify master
 ;;
backup)
 notify backup
 ;;
fault)
 notify fault
 ;;
*)
 echo "Usage: $(basename $0) {master|backup|fault}"
 exit 1
 ;;
esac


節點二同樣配置


四 啟動服務並測試

1.啟動後端web服務器

[[email protected] ~]# systemctl start httpd
[[email protected] ~]# systemctl start httpd


為了測試顯示效果明顯一點，自定義一個訪問頁面

[[email protected] ~]# echo "RS1:172.18.67.11" > /var/www/html/index.html
[[email protected] ~]# echo "RS2:172.18.67.12" > /var/www/html/index.html

2.測試
MASTER:

[[email protected] ~]# systemctl start  keepalived
[[email protected] ~]# systemctl status -l  keepalived
● keepalived.service - LVS and VRRP High Availability Monitor
   Loaded: loaded (/usr/lib/systemd/system/keepalived.service; disabled; vendor preset: disabled)
   Active: active (running) since Mon 2017-05-15 15:45:20 CST; 3s ago
  Process: 20971 ExecStart=/usr/sbin/keepalived $KEEPALIVED_OPTIONS (code=exited, status=0/SUCCESS)
 Main PID: 20972 (keepalived)
   CGroup: /system.slice/keepalived.service
           ├─20972 /usr/sbin/keepalived -D
           ├─20973 /usr/sbin/keepalived -D
           └─20974 /usr/sbin/keepalived -D
May 15 15:45:20 inode2 Keepalived_healthcheckers[20973]: Opening file ‘/etc/keepalived/keepalived.conf‘.
May 15 15:45:20 inode2 Keepalived_healthcheckers[20973]: Configuration is using : 7521 Bytes
May 15 15:45:20 inode2 Keepalived_healthcheckers[20973]: Using LinkWatch kernel netlink reflector...
May 15 15:45:20 inode2 Keepalived_vrrp[20974]: VRRP_Script(chk_nginx) succeeded
May 15 15:45:21 inode2 Keepalived_vrrp[20974]: VRRP_Instance(myr) Transition to MASTER STATE
May 15 15:45:22 inode2 Keepalived_vrrp[20974]: VRRP_Instance(myr) Entering MASTER STATE
May 15 15:45:22 inode2 Keepalived_vrrp[20974]: VRRP_Instance(myr) setting protocol VIPs.
May 15 15:45:22 inode2 Keepalived_vrrp[20974]: VRRP_Instance(myr) Sending gratuitous ARPs on eno16777736 for 172.18.67.33
May 15 15:45:22 inode2 Keepalived_vrrp[20974]: Opening script file /etc/keepalived/notify.sh
May 15 15:45:22 inode2 Keepalived_healthcheckers[20973]: Netlink reflector reports IP 172.18.67.33 added
[[email protected] ~]# ip a l
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
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: eno16777736: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 00:0c:29:8b:08:6f brd ff:ff:ff:ff:ff:ff
    inet 172.18.67.13/16 brd 172.18.255.255 scope global eno16777736
       valid_lft forever preferred_lft forever
    inet 172.18.67.33/16 scope global secondary eno16777736
       valid_lft forever preferred_lft forever
    inet6 fe80::20c:29ff:fe8b:86f/64 scope link tentative dadfailed 
       valid_lft forever preferred_lft forever


主節點啟動，Entering MASTER STATE，此時我們在客戶端進行測試訪問

[[email protected] ~]# for i in {1..4};do curl http://172.18.67.33;done
RS1:172.18.67.11
RS2:172.18.67.12
RS1:172.18.67.11
RS2:172.18.67.12

訪問正常，接下來我們啟動備用節點的服務器

BACKUP:

[[email protected] keepalived]# systemctl start keepalived
[[email protected] keepalived]# systemctl status -l keepalived
● keepalived.service - LVS and VRRP High Availability Monitor
   Loaded: loaded (/usr/lib/systemd/system/keepalived.service; disabled; vendor preset: disabled)
   Active: active (running) since Mon 2017-05-15 15:46:51 CST; 3s ago
  Process: 24329 ExecStart=/usr/sbin/keepalived $KEEPALIVED_OPTIONS (code=exited, status=0/SUCCESS)
 Main PID: 24330 (keepalived)
   CGroup: /system.slice/keepalived.service
           ├─24330 /usr/sbin/keepalived -D
           ├─24331 /usr/sbin/keepalived -D
           └─24332 /usr/sbin/keepalived -D
May 15 15:46:51 inode3 Keepalived_vrrp[24332]: Registering Kernel netlink command channel
May 15 15:46:51 inode3 Keepalived_vrrp[24332]: Registering gratuitous ARP shared channel
May 15 15:46:51 inode3 Keepalived_vrrp[24332]: Opening file ‘/etc/keepalived/keepalived.conf‘.
May 15 15:46:51 inode3 Keepalived_vrrp[24332]: Configuration is using : 66427 Bytes
May 15 15:46:51 inode3 Keepalived_vrrp[24332]: Using LinkWatch kernel netlink reflector...
May 15 15:46:51 inode3 Keepalived_vrrp[24332]: VRRP_Instance(myr) Entering BACKUP STATE
May 15 15:46:51 inode3 Keepalived_vrrp[24332]: Opening script file /etc/keepalived/notify.sh
May 15 15:46:51 inode3 Keepalived_vrrp[24332]: VRRP sockpool: [ifindex(2), proto(112), unicast(0), fd(10,11)]
May 15 15:46:51 inode3 Keepalived_vrrp[24332]: VRRP_Script(chk_down) succeeded
May 15 15:46:51 inode3 Keepalived_vrrp[24332]: VRRP_Script(chk_nginx) succeeded
[[email protected] keepalived]# ip a l
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
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: eno16777736: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 00:0c:29:78:24:c3 brd ff:ff:ff:ff:ff:ff
    inet 172.18.67.14/16 brd 172.18.255.255 scope global eno16777736
       valid_lft forever preferred_lft forever
    inet6 fe80::20c:29ff:fe78:24c3/64 scope link tentative dadfailed 
       valid_lft forever preferred_lft forever

此時，我們可以看到備用節點服務器啟動後進入了BACKUP狀態，Entering BACKUP STATE。接下來我們測試主節點宕機的情形下，我們的服務是否還可用

[[email protected] ~]# systemctl stop keepalived


主節點宕機後我們查看備用節點的狀態

[[email protected] keepalived]# systemctl status -l keepalived
● keepalived.service - LVS and VRRP High Availability Monitor
   Loaded: loaded (/usr/lib/systemd/system/keepalived.service; disabled; vendor preset: disabled)
   Active: active (running) since Mon 2017-05-15 15:46:51 CST; 2min 19s ago
  Process: 24329 ExecStart=/usr/sbin/keepalived $KEEPALIVED_OPTIONS (code=exited, status=0/SUCCESS)
 Main PID: 24330 (keepalived)
   CGroup: /system.slice/keepalived.service
           ├─24330 /usr/sbin/keepalived -D
           ├─24331 /usr/sbin/keepalived -D
           └─24332 /usr/sbin/keepalived -D
May 15 15:46:51 inode3 Keepalived_vrrp[24332]: VRRP sockpool: [ifindex(2), proto(112), unicast(0), fd(10,11)]
May 15 15:46:51 inode3 Keepalived_vrrp[24332]: VRRP_Script(chk_down) succeeded
May 15 15:46:51 inode3 Keepalived_vrrp[24332]: VRRP_Script(chk_nginx) succeeded
May 15 15:48:35 inode3 Keepalived_vrrp[24332]: VRRP_Instance(myr) Transition to MASTER STATE
May 15 15:48:36 inode3 Keepalived_vrrp[24332]: VRRP_Instance(myr) Entering MASTER STATE
May 15 15:48:36 inode3 Keepalived_vrrp[24332]: VRRP_Instance(myr) setting protocol VIPs.
May 15 15:48:36 inode3 Keepalived_vrrp[24332]: VRRP_Instance(myr) Sending gratuitous ARPs on eno16777736 for 172.18.67.33
May 15 15:48:36 inode3 Keepalived_vrrp[24332]: Opening script file /etc/keepalived/notify.sh
May 15 15:48:36 inode3 Keepalived_healthcheckers[24331]: Netlink reflector reports IP 172.18.67.33 added
May 15 15:48:41 inode3 Keepalived_vrrp[24332]: VRRP_Instance(myr) Sending gratuitous ARPs on eno16777736 for 172.18.67.33
[[email protected] keepalived]# ip a l
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
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: eno16777736: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 00:0c:29:78:24:c3 brd ff:ff:ff:ff:ff:ff
    inet 172.18.67.14/16 brd 172.18.255.255 scope global eno16777736
       valid_lft forever preferred_lft forever
    inet 172.18.67.33/16 scope global secondary eno16777736
       valid_lft forever preferred_lft forever
    inet6 fe80::20c:29ff:fe78:24c3/64 scope link tentative dadfailed 
       valid_lft forever preferred_lft forever

我們發現備用節點由備用狀態進入了主狀態，並且IP地址也成功綁定至備用節點下。再次進行測試訪問

[[email protected] ~]# for i in {1..4};do curl http://172.18.67.33;done
RS1:172.18.67.11
RS2:172.18.67.12
RS1:172.18.67.11
RS2:172.18.67.12

測試一臺web服務器宕機

[[email protected] ~]# systemctl stop httpd
[[email protected] ~]# for i in {1..4};do curl http://172.18.67.33;done
RS2:172.18.67.12
RS2:172.18.67.12
RS2:172.18.67.12
RS2:172.18.67.12


在實際生產環境中後端兩臺web服務器的內容應該一樣的，在這裏我們可認為客戶端已成功訪問到服務器，因此我們可認為這樣的架構體現了高可用負載均衡。

Keepalived+Nginx實現高可用負載均衡集群

相關推薦

Keepalived+Nginx實現高可用負載均衡集群

連接 靜態 adf -1 rip mail fff hostname dex 一 環境介紹 1.操作系統CentOS Linux release 7.2.1511 (Core) 2.服務keepalived+lvs雙主高可用負載均衡集群及LAMP應用keepalived-1

keepalived+lvs實現高可用負載均衡集群

keepalived+lvs實現高可用LVS實戰篇第1章 環境準備1.1 系統環境1.1.1 系統版本[root@lb01 ~]# cat /etc/redhat-release CentOS release 6.9 (Final) [root@lb01 ~]# uname -r 2.6.32-696.el6

LVS+keepalived高可用負載均衡集群部署（二）---LAMP網站服務器與LVS服務器

虛擬機l 系統環境： RHEL7 l 硬件環境：虛擬機l 項目描述：為解決網站訪問壓力大的問題，需要搭建高可用、負載均衡的 web集群。 l 架構說明：整個服務架構采用功能分離的方式部署。後端采用2臺mysql 數據庫，實現主從結構以及讀寫分離。中間LAMP網站服務器共有2臺， 前端 2臺LVS服務

Nginx高可用負載均衡集群實例架設

notify net 服務 反向代理 interface nss 完成 ansi tran Nginx高可用負載均衡集群實例架設一、 Nginx高可用集群拓撲架構 node1：17216.100.67node2：170.16.100.68VIP：172.16.100.89

docker下用keepalived+Haproxy實現高可用負載均衡叢集

先記錄下遇到的坑，避免之後忘了； 花時間最多去解決的一個題是：在docker下啟動haproxy服務後在這個docker服務內安裝keepalived無法ping通的問題，雖然最後也是莫名其妙就好了，但是加強了不少對docker的理解和還需深入學習的地方。 為什麼要用

RHCS套件+Nginx實現高可用負載均衡

紅帽叢集套件（RedHat Cluter Suite, RHCS）是一套綜合的軟體元件，可以通過在部署時採用不同的配置，以滿足你對高可用性，負載均衡，可擴充套件性，檔案共享和節約成本的需要。 它提供有如下兩種不同型別的叢集：  1、高可用性：應用/服務故障切換－通過建立n

搭建MySQL高可用負載均衡集群

分別是 $? 啟動腳本 hang 常見 ase 說明 配置步驟 分享 閱讀目錄1、簡介2、基本環境3、配置MySQL主主復制4、中間件簡述　　4.1、Haproxy介紹　　4.2、keepalived介紹5、中間件的安裝與配置（haproxy、keepalived）　　5.

跨園區容災，升級不停服：高可用負載均衡集群實踐

vrrp 優化 數量 一個 域名 live 內容 tro 靈活 歡迎大家前往騰訊雲+社區，獲取更多騰訊海量技術實踐幹貨哦~ 本文由騰訊雲中間件團隊發表於雲+技術周刊特別版 作者：方坤丁 對於雲計算行業來說，雲服務的可用性和可擴展性是的檢測其服務質量的重要標準，也是最

nginx+keepalived實現高可用負載均衡

其中 centos7.3 9.png IT 配置文件 bsp 是我 add nginx 環境： centos7.3虛擬機A 10.0.3.46 centos7.3虛擬機B 10.0.3.110 虛擬機A和B都需要安裝nginx和keepalived（過程省略，其中keepa

分散式架構學習之：Keepalived+Nginx實現高可用Web負載均衡

一、場景需求 二、Keepalived 簡要介紹 Keepalived 是一種高效能的伺服器高可用或熱備解決方案，Keepalived 可以用來防止伺服器單點故障的發生，通過配合 Nginx 可以實現 web 前端服務的高可用。 Keepalived 以 VRRP

Keepalived + Nginx 實現高可用(雙機熱備) Web 負載均衡 安裝測試筆記

system ati 根據 properly 節點配置 sys .gz ive error: keepalived是什麽 keepalived是集群管理中保證集群高可用的一個服務軟件，其功能類似於heartbeat，用來防止單點故障。 keepalived工作原理 k

LVS+Keepalived實現高可用負載均衡

lvs+keepalived 高可用 負載均衡 用LVS+Keepalived實現高可用負載均衡，簡單來說就是由LVS提供負載均衡，keepalived通過對rs進行健康檢查、對主備機(director)進行故障自動切換，實現高可用。1. LVS NAT模式配置準備三臺服務器，一臺director, 兩

HAProxy+Varnish+LNMP實現高可用負載均衡動靜分離集群部署

else 應用服務器 bash == 開機啟動 多少 heal 啟用 4.0 轉自http://bbs.hfteams.com/forum.php?mod=viewthread&tid=11&extra=page%3D1 HAProxy+Varnish+LN

keepalived+Nginx實現高可用集群

auth timeout cati -c stat -h track ast 顯示 keepalived安裝Nginx安裝 配置keepalived#修改配置vim /etc/keepalived/keepalived.conf #配置文件 ! Config

nginx+keepalive實現高可用負載均衡

keepalived+nginx高可實驗一：實驗環境 主nginx負載均衡器：192.168.10.63 （通過keepalived配置了VIP：192.168.10.188供外使用）副nginx負載均衡器：192.168.10.200（通過keepalived配置了VIP：192.168.10.188供外

LVS+Keepalived 實現高可用負載均衡叢集

LVS+Keepalived  實現高可用負載均衡叢集     隨著網站業務量的增長，網站的伺服器壓力越來越大？需要負載均衡方案！商業的硬體如 F5 ，Array又太貴，你們又是創業型互聯公司如何有效節約成本，節省不必要的浪費？同時還需要實現商業硬體一樣的高效能高可

Keepalived + LVS + LAMP 高可用負載均衡叢集實現

一、Keepalived 簡介           Keepalived 是一個用 C 語言編寫的路由軟體。它最初是專門為 LVS 負載均衡軟體設計的，用來管理並監控 LVS 集群系統中各個服務節點的狀態，後來又加

saltstack實現一鍵部署keepalived+haproxy的高可用負載均衡叢集

配置環境 主機名 ip 服務 server1 172.25.1.1 salt-master、salt-minion haproxy、keepalived server2 172.25.1.2 salt-minion

【Linux運維-叢集技術進階】Nginx+Keepalived+Tomcat搭建高可用/負載均衡/動靜分離的Web伺服器叢集

額，部落格名字有點長。。。 前言 終於到這篇文章了，心情是有點激動的。因為這篇文章會集中以前部落格講到的所有Nginx功能點，包括基本的負載均衡，還有動靜分離技術再加上這篇文章的重點，通過Keepalived實現的HA(High Available)，為什麼

【Linux】LVS+Keepalived實現高可用負載均衡（Web群集）

一、Keepalived概述 keepalived是一個類似於layer3,4,5交換機制的軟體，也就是我們平時說的第3層、第4層和第5層交換。Keepalived的作用是檢測web伺服器的狀態，