当前位置: 首页 > news >正文

haproxy基本配置详解

一、haproxy基础配置

官方文档:HAProxy version 2.2.22 - Configuration Manual

HAProxy 的配置文件haproxy.cfg由两大部分组成,分别是global和proxies部分。

global:全局配置段

  • 进程及安全配置相关的参数
  • 性能调整相关参数
  • Debug参数

proxies:代理配置段

  • defaults:为frontend, backend, listen提供默认配置
  • frontend:前端,相当于nginx中的server {}
  • backend:后端,相当于nginx中的upstream {}
  • listen:同时拥有前端和后端配置

1、haproxy global 配置

官方文档:HAProxy version 2.2.22 - Configuration Manual

chroot #锁定运行目录
deamon #以守护进程运行
stats socket /var/lib/haproxy/haproxy.sock mode 600 level admin process 1 #socket文件
user, group, uid, gid  #运行haproxy的用户身份
nbproc    n     #开启的haproxy work 进程数,默认进程数是一个
#nbthread  1    #指定每个haproxy进程开启的线程数,默认为每个进程一个线程,和nbproc互斥(版本有关)
#如果同时启用nbproc和nbthread 会出现以下日志的错误,无法启动服务
Apr  7 14:46:23 haproxy haproxy: [ALERT] 097/144623 (1454) : config : cannot enable multiple processes if multiple threads are configured. Please use either nbproc or nbthread but not both.cpu-map 1 0     #绑定haproxy 进程至指定CPU,将第一个work进程绑定至0号CPU
maxconn  n      #每个haproxy进程的最大并发连接数
maxsslconn  n   #每个haproxy进程ssl最大连接数,用于haproxy配置了证书的场景下
maxconnrate n   #每个进程每秒创建的最大连接数量
spread-checks n #后端server状态check随机提前或延迟百分比时间,建议2-5(20%-50%)之间,默认值0
pidfile         #指定pid文件路径
log 127.0.0.1  local2 info #定义全局的syslog服务器;日志服务器需要开启UDP协议,最多可以定义两个

范例:多进程和socket文件

[root@node1 ~]# vim /etc/haproxy/haproxy.cfg
global
maxconn 100000
chroot /apps/haproxy
stats socket /var/lib/haproxy/haproxy.sock1 mode 600 level admin process 1               
stats socket /var/lib/haproxy/haproxy.sock2 mode 600 level admin process 2
uid 99
gid 99
daemon
nbproc 2
[root@node1 ~]# systemctl restart haproxy
[root@node1 ~]# pstree -p |grep haproxy|-haproxy(2281)-+-haproxy(2285)|               `-haproxy(2286)
[root@node1 ~]# ll /var/lib/haproxy/
总用量 4
-rw-r--r-- 1 root root 5 2月  23 10:34 haproxy.pid
srw------- 1 root root 0 2月  23 10:23 haproxy.sock
srw------- 1 root root 0 2月  23 10:34 haproxy.sock1
srw------- 1 root root 0 2月  23 10:34 haproxy.sock2

2、haproxy proxies配置

官方文档:HAProxy version 2.2.22 - Configuration Manual

defaults [<name>] #默认配置项,针对以下的frontend、backend和listen生效,可以多个name也可以没有name
frontend <name>   #前端servername,类似于Nginx的一个虚拟主机 server和LVS服务集群。
backend  <name>   #后端服务器组,等于nginx的upstream和LVS中的RS服务器
listen   <name>   #将frontend和backend合并在一起配置,相对于frontend和backend配置更简洁,生产常用

注意:name字段只能使用大小写字母,数字,‘-’(dash),’_‘(underscore),’.’ (dot)和 ‘:'(colon),并且严格区分大小写

(1)Proxies配置-defaults

defaults 配置参数:

option redispatch       #当server Id对应的服务器挂掉后,强制定向到其他健康的服务器,重新派发
option abortonclose     #当服务器负载很高时,自动结束掉当前队列处理比较久的链接,针对业务情况选择开启
option http-keep-alive  #开启与客户端的会话保持
option  forwardfor      #透传客户端真实IP至后端web服务器
mode http|tcp           #设置默认工作类型,使用TCP服务器性能更好,减少压力
timeout http-keep-alive 120s #session 会话保持超时时间,此时间段内会转发到相同的后端服务器
timeout connect 120s    #客户端请求从haproxy到后端server最长连接等待时间(TCP连接之前),默认单位ms
timeout server  600s    #客户端请求从haproxy到后端服务端的请求处理超时时长(TCP连接之后),默认单位ms,如果超时,会出现502错误,此值建议设置较大些,访止502错误
timeout client  600s    #设置haproxy与客户端的最长非活动时间,默认单位ms,建议和timeout server相同
timeout  check   5s     #对后端服务器的默认检测超时时间
default-server inter 1000 weight 3   #指定后端服务器的默认设置

(2)Proxies配置-frontend

fronted 配置参数:

bind:   #指定HAProxy的监听地址,可以是IPV4或IPV6,可以同时监听多个IP或端口,可同时用于listen字段中#格式:
bind [<address>]:<port_range> [, ...] [param*]#注意:如果需要绑定在非本机的IP,需要开启内核参数:net.ipv4.ip_nonlocal_bind=1

范例:

listen http_proxy                           #监听http的多个IP的多个端口和sock文件bind :80,:443,:8801-8810bind 10.0.0.1:10080,10.0.0.1:10443bind /var/run/ssl-frontend.sock user root mode 600 accept-proxylisten http_https_proxy                     #https监听bind :80bind :443 ssl crt /etc/haproxy/site.pem #公钥和私钥公共文件listen http_https_proxy_explicit            #监听ipv6、ipv4和unix sock文件bind ipv6@:80bind ipv4@public_ssl:443 ssl crt /etc/haproxy/site.pembind unix@ssl-frontend.sock user root mode 600 accept-proxylisten external_bind_app1                   #监听file descriptorbind "fd@${FD_APP1}"

生产示例:

frontend  oepnlab_web_port               #可以采用后面形式命名:业务-服务-端口号bind :80,:8080bind 10.0.0.7:10080,:8801-8810,10.0.0.17:9001-9010mode  http|tcp              #指定负载协议类型use_backend <backend_name>  #调用的后端服务器组名称

(3)backend 配置-backend

定义一组后端服务器,backend服务器将被frontend进行调用:

mode  http|tcp      #指定负载协议类型,和对应的frontend必须一致
option              #配置选项
server              #定义后端real server

注意:option后面加 httpchk,smtpchk,mysql-check,pgsql-check,ssl-hello-chk方法,可用于实现更多应用层检测功能。

option 配置参数:

check               #对指定real进行健康状态检查,如果不加此设置,默认不开启检查addr  <IP>        #可指定的健康状态监测IP,可以是专门的数据网段,减少业务网络的流量port  <num>   #指定的健康状态监测端口inter <num>   #健康状态检查间隔时间,默认2000 msfall  <num>       #后端服务器从线上转为线下的检查的连续失效次数,默认为3rise  <num>       #后端服务器从下线恢复上线的检查的连续有效次数,默认为2
weight  <weight>  #默认为1,最大值为256,0表示不参与负载均衡,但仍接受持久连接
backup              #将后端服务器标记为备份状态,只在所有非备份主机down机时提供服务,类似Sorry Server
disabled            #将后端服务器标记为不可用状态,即维护状态,除了持久模式,将不再接受连接
redirect prefix  http://www.baidu.com/      #将请求临时(302)重定向至其它URL,只适用于http模式
redir http://www.baidu.com                  #将请求临时(302)重定向至其它URL,只适用于http模式
maxconn <maxconn>     #当前后端server的最大并发连接数
backlog <backlog> #当前端服务器的连接数达到上限后的后援队列长度,注意:不支持backend

(4)frontend+backend配置实例

范例1:

frontend openlab-test-httpbind :80,:8080mode tcpuse_backend openlab-test-http-nodesbackend openlab-test-http-nodesmode tcpdefault-server inter 1000 weight 6server web1 192.168.150.14:80 check weight 2 addr 192.168.150.114 port 8080server web2 192.168.150.15:80 check

范例2:

#官网业务访问入口
frontend  WEB_PORT_80bind 192.168.150.11:80mode httpuse_backend  web_prot_http_nodesbackend web_prot_http_nodesmode  httpoption forwardforserver 192.168.150.14 192.168.150.14:80   check inter 3000 fall 3 rise 5server 192.168.150.15 192.168.150.15:80   check inter 3000 fall 3 rise 5

(5)Proxies配置-listen替代frontend+backend

使用listen替换上面的frontend和backend的配置方式,可以简化设置,通常只用于TCP协议的应用:

#官网业务访问入口
listen  WEB_PORT_80bind 192.168.150.11:80mode httpoption  forwardforserver web1   192.168.150.14:80   check inter 3000 fall 3 rise 5server web2   192.168.150.15:80   check inter 3000 fall 3 rise 5

二、使用子配置文件保存配置

当业务众多时,将所有配置都放在一个配置文件中,会造成维护困难。可以考虑按业务分类,将配置信息拆分,放在不同的子配置文件中,从而达到方便维护的目的。

#创建子配置目录
[root@localhost ~]# mkdir /etc/haproxy/conf.d/#创建子配置文件,注意:必须为cfg后缀
[root@localhost ~]# vim   /etc/haproxy/conf.d/test.cfg
listen  WEB_PORT_80bind 192.168.150.11:80mode httpbalance roundrobinserver web1   192.168.150.14:80   check inter 3000 fall 3 rise 5server web2   192.168.150.15:80   check inter 3000 fall 3 rise 5#添加子配置目录到unit文件中
[root@localhost ~]# vim  /lib/systemd/system/haproxy.service
[Unit]
Description=HAProxy Load Balancer
After=syslog.target network.target[Service]
ExecStartPre=/usr/sbin/haproxy -f /etc/haproxy/haproxy.cfg -f /etc/haproxy/conf.d/ -c -q
ExecStart=/usr/sbin/haproxy -Ws -f /etc/haproxy/haproxy.cfg -f /etc/haproxy/conf.d/  -p /var/lib/haproxy/haproxy.pid
ExecReload=/bin/kill -USR2 $MAINPID[Install]
WantedBy=multi-user.target[root@localhost ~]# systemctl daemon-reload 
[root@localhost ~]# systemctl restart haproxy

三、haproxy日志配置

配置HAProxy记录日志到指定日志文件中

1、haproxy配置

#在global配置项定义:
log 127.0.0.1  local{1-7} info #基于syslog记录日志到指定设备,级别有(err、warning、info、debug)listen  web_portbind 127.0.0.1:80mode httplog global                                #开启当前web_port的日志功能,默认不记录日志server web1  127.0.0.1:8080  check inter 3000 fall 2 rise 5# systemctl  restart haproxy

2、rsyslog配置

vim /etc/rsyslog.conf 
$ModLoad imudp
$UDPServerRun 514
......
# Save haproxy messages also to haproxy.log
local2.*                                                /var/log/haproxy.log
......# systemctl  restart rsyslog

3、验证haproxy日志

重启syslog服务并访问app页面,然后验证是否生成日志

[root@localhost ~]# cat /var/log/haproxy.log
Feb 23 16:30:52 localhost haproxy[12324]: Connect from 192.168.121.1:55708 to 192.168.121.51:9999 (stats/HTTP)
Feb 23 16:30:52 localhost haproxy[12324]: Connect from 192.168.121.1:55708 to 192.168.121.51:9999 (stats/HTTP)
Feb 23 16:30:52 localhost haproxy[12324]: Connect from 192.168.121.1:55708 to 192.168.121.51:9999 (stats/HTTP)

四、haproxy调度算法

HAProxy通过固定参数balance指明对后端服务器的调度算法,该参数可以配置在listen或backend选项中。

HAProxy的调度算法分为静态和动态调度算法,但是有些算法可以根据参数在静态和动态算法中相互转换。

官方文档:HAProxy version 2.2.22 - Configuration Manual

1、静态算法

静态算法:按照事先定义好的规则轮询公平调度,不关心后端服务器的当前负载、链接数和响应速度等,且无法实时修改权重,只能靠重启HAProxy生效。

可以利用 socat工具对服务器动态权重和其它状态的调整,Socat 是 Linux 下的一个多功能的网络工具,名字来由是Socket CAT,Socat 的主要特点就是在两个数据流之间建立通道,且支持众多协议和链接方式。如 IP、TCP、 UDP、IPv6、Socket文件等

范例:利用工具socat 对服务器动态权重调整

[root@node1 ~]# yum -y install socat#查看帮助
[root@node1 ~]# socat -h
[root@node1 ~]# echo "help" | socat stdio /var/lib/haproxy/haproxy.sock
The following commands are valid at this level:abort ssl ca-file <cafile>              : abort a transaction for a CA fileabort ssl cert <certfile>               : abort a transaction for a certificate fileabort ssl crl-file <crlfile>            : abort a transaction for a CRL fileadd acl [@<ver>] <acl> <pattern>        : add an acl entryadd map [@<ver>] <map> <key> <val>      : add a map entry (payload supported instead of key/val)add server <bk>/<srv>                   : create a new serveradd ssl ca-file <cafile> <payload>      : add a certificate into the CA fileadd ssl crt-list <list> <cert> [opts]*  : add to crt-list file <list> a line <cert> or a payloadclear acl [@<ver>] <acl>                : clear the contents of this aclclear counters [all]                    : clear max statistics counters (or all counters)clear map [@<ver>] <map>                : clear the contents of this mapclear table <table> [<filter>]*         : remove an entry from a table (filter: data/key)commit acl @<ver> <acl>                 : commit the ACL at this versioncommit map @<ver> <map>                 : commit the map at this versioncommit ssl ca-file <cafile>             : commit a CA filecommit ssl cert <certfile>              : commit a certificate filecommit ssl crl-file <crlfile>           : commit a CRL filedebug counters [?|all|bug|cnt|chk|glt]* : dump/reset rare event countersdebug dev hash   [msg]                  : return msg hashed if anon is setdel acl <acl> [<key>|#<ref>]            : delete acl entries matching <key>del map <map> [<key>|#<ref>]            : delete map entries matching <key>del server <bk>/<srv>                   : remove a dynamically added serverdel ssl ca-file <cafile>                : delete an unused CA filedel ssl cert <certfile>                 : delete an unused certificate filedel ssl crl-file <crlfile>              : delete an unused CRL filedel ssl crt-list <list> <cert[:line]>   : delete a line <cert> from crt-list file <list>disable agent                           : disable agent checksdisable dynamic-cookie backend <bk>     : disable dynamic cookies on a specific backenddisable frontend <frontend>             : temporarily disable specific frontenddisable health                          : disable health checksdisable server (DEPRECATED)             : disable a server for maintenance (use 'set server' instead)dump stats-file                         : dump stats for restoreenable agent                            : enable agent checksenable dynamic-cookie backend <bk>      : enable dynamic cookies on a specific backendenable frontend <frontend>              : re-enable specific frontendenable health                           : enable health checksenable server  (DEPRECATED)             : enable a disabled server (use 'set server' instead)get acl <acl> <value>                   : report the patterns matching a sample for an ACLget map <acl> <value>                   : report the keys and values matching a sample for a mapget var <name>                          : retrieve contents of a process-wide variableget weight <bk>/<srv>                   : report a server's current weightnew ssl ca-file <cafile>                : create a new CA file to be used in a crt-listnew ssl cert <certfile>                 : create a new certificate file to be used in a crt-list or a directorynew ssl crl-file <crlfile>               : create a new CRL file to be used in a crt-listoperator                                : lower the level of the current CLI session to operatorprepare acl <acl>                       : prepare a new version for atomic ACL replacementprepare map <acl>                       : prepare a new version for atomic map replacementset anon global-key <value>             : change the global anonymizing keyset anon off                            : deactivate the anonymized modeset anon on [value]                     : activate the anonymized modeset dynamic-cookie-key backend <bk> <k> : change a backend secret key for dynamic cookiesset map <map> [<key>|#<ref>] <value>    : modify a map entryset maxconn frontend <frontend> <value> : change a frontend's maxconn settingset maxconn global <value>              : change the per-process maxconn settingset maxconn server <bk>/<srv>           : change a server's maxconn settingset profiling <what> {auto|on|off}      : enable/disable resource profiling (tasks,memory)set rate-limit <setting> <value>        : change a rate limiting valueset server <bk>/<srv> [opts]            : change a server's state, weight, address or sslset severity-output [none|number|string]: set presence of severity level in feedback informationset ssl ca-file <cafile> <payload>      : replace a CA fileset ssl cert <certfile> <payload>       : replace a certificate fileset ssl crl-file <crlfile> <payload>    : replace a CRL fileset ssl ocsp-response <resp|payload>       : update a certificate's OCSP Response from a base64-encode DERset ssl tls-key [id|file] <key>         : set the next TLS key for the <id> or <file> listener to <key>set table <table> key <k> [data.* <v>]* : update or create a table entry's dataset timeout [cli] <delay>               : change a timeout settingset weight <bk>/<srv>  (DEPRECATED)     : change a server's weight (use 'set server' instead)show acl [@<ver>] <acl>]                : report available acls or dump an acl's contentsshow activity [-1|0|thread_num]         : show per-thread activity stats (for support/developers)show anon                               : display the current state of anonymized modeshow backend                            : list backends in the current running configshow cache                              : show cache statusshow cli level                          : display the level of the current CLI sessionshow cli sockets                        : dump list of cli socketsshow dev                                : show debug info for developersshow env [var]                          : dump environment variables known to the processshow errors [<px>] [request|response]   : report last request and/or response errors for each proxyshow events [<sink>] [-w] [-n]          : show event sink stateshow fd [-!plcfbsd]* [num]              : dump list of file descriptors in use or a specific oneshow info [desc|json|typed|float]*      : report information about the running processshow libs                               : show loaded object files and librariesshow map [@ver] [map]                   : report available maps or dump a map's contentsshow peers [dict|-] [section]           : dump some information about all the peers or this peers sectionshow pools [by*] [match <pfx>] [nb]     : report information about the memory pools usageshow profiling [<what>|<#lines>|<opts>]*: show profiling state (all,status,tasks,memory)show resolvers [id]                     : dumps counters from all resolvers section and associated name serversshow schema json                        : report schema used for statsshow servers conn [<backend>]           : dump server connections status (all or for a single backend)show servers state [<backend>]          : dump volatile server information (all or for a single backend)show sess [<id>|all|susp|older <age>]   : report the list of current sessions or dump this exact sessionshow ssl ca-file [<cafile>[:<index>]]   : display the SSL CA files used in memory, or the details of a <cafile>, or a single certificate of index <index> of a CA file <cafile>show ssl cert [<certfile>]              : display the SSL certificates used in memory, or the details of a fileshow ssl crl-file [<crlfile[:<index>>]] : display the SSL CRL files used in memory, or the details of a <crlfile>, or a single CRL of index <index> of CRL file <crlfile>show ssl crt-list [-n] [<list>]         : show the list of crt-lists or the content of a crt-list file <list>show ssl ocsp-response [[text|base64] id]  : display the IDs of the OCSP responses used in memory, or the details of a single OCSP response (in text or base64 format)show ssl ocsp-updates                      : display information about the next 'nb' ocsp responses that will be updated automaticallyshow startup-logs                       : report logs emitted during HAProxy startupshow stat [desc|json|no-maint|typed|up]*: report counters for each proxy and servershow table <table> [<filter>]*          : report table usage stats or dump this table's contents (filter: data/key)show tasks                              : show running tasksshow threads                            : show some threads debugging informationshow tls-keys [id|*]                    : show tls keys references or dump tls ticket keys when id specifiedshow trace [<module>]                   : show live tracing stateshow version                            : show version of the current processshutdown frontend <frontend>            : stop a specific frontendshutdown session [id]                   : kill a specific sessionshutdown sessions server <bk>/<srv>     : kill sessions on a servertrace [<module>|0] [cmd [args...]]      : manage live tracing (empty to list, 0 to stop all)update ssl ocsp-response <certfile>     : send ocsp request and update stored ocsp responseuser                                    : lower the level of the current CLI session to userwait {-h|<delay_ms>} cond [args...]     : wait the specified delay or condition (-h to see list)help [<command>]                        : list matching or all commandsprompt [timed]                          : toggle interactive mode with promptquit                                    : disconnect[root@node1 ~]# echo "show info" | socat stdio /var/lib/haproxy/haproxy.sock
Name: HAProxy
Version: 3.0.8-6036c31
Release_date: 2025/01/29
Nbthread: 2
Nbproc: 1
Process_num: 1
Pid: 12965
Uptime: 0d 0h39m50s
Uptime_sec: 2390
Memmax_MB: 0
PoolAlloc_MB: 0
PoolUsed_MB: 0
PoolFailed: 0
Ulimit-n: 200032
Maxsock: 200032
Maxconn: 100000
Hard_maxconn: 100000
CurrConns: 0
CumConns: 2400
CumReq: 113
MaxSslConns: 0
CurrSslConns: 0
CumSslConns: 0
Maxpipes: 0
PipesUsed: 0
PipesFree: 0
ConnRate: 0
ConnRateLimit: 0
MaxConnRate: 15
SessRate: 0
SessRateLimit: 0
MaxSessRate: 15
SslRate: 0
SslRateLimit: 0
MaxSslRate: 0
SslFrontendKeyRate: 0
SslFrontendMaxKeyRate: 0
SslFrontendSessionReuse_pct: 0
SslBackendKeyRate: 0
SslBackendMaxKeyRate: 0
SslCacheLookups: 0
SslCacheMisses: 0
CompressBpsIn: 0
CompressBpsOut: 0
CompressBpsRateLim: 0
ZlibMemUsage: 0
MaxZlibMemUsage: 0
Tasks: 17
Run_queue: 0
Idle_pct: 100
node: localhost.localdomain
Stopping: 0
Jobs: 6
Unstoppable Jobs: 1
Listeners: 4
ActivePeers: 0
ConnectedPeers: 0
DroppedLogs: 0
BusyPolling: 0
FailedResolutions: 0
TotalBytesOut: 45902
TotalSplicedBytesOut: 0
BytesOutRate: 0
DebugCommandsIssued: 0
CumRecvLogs: 0
Build info: 3.0.8-6036c31
Memmax_bytes: 0
PoolAlloc_bytes: 277960
PoolUsed_bytes: 277960
Start_time_sec: 1740300080
Tainted: 0
TotalWarnings: 1
MaxconnReached: 0
BootTime_ms: 16
Niced_tasks: 0
CurrStreams: 1
CumStreams: 113
BlockedTrafficWarnings: 0[root@node1 ~]# cat /etc/haproxy/conf.d/test.cfg 
listen  WEB_PORT_80bind 192.168.150.11:80mode httpbalance roundrobinserver web1   192.168.150.14:80   check inter 3000 fall 3 rise 5server web2   192.168.150.15:80   check inter 3000 fall 3 rise 5 weight 3[root@node1 ~]# echo "show servers state" | socat stdio /var/lib/haproxy/haproxy.sock
# be_id be_name srv_id srv_name srv_addr srv_op_state srv_admin_state srv_uweight srv_iweight srv_time_since_last_change srv_check_status srv_check_result srv_check_health srv_check_state srv_agent_state bk_f_forced_id srv_f_forced_id srv_fqdn srv_port srvrecord
3 WEB_PORT_80 1 web1 192.168.150.14 2 0 1 1 3 6 3 7 6 0 0 0 - 80 -
3 WEB_PORT_80 2 web2 192.168.150.15 2 0 3 3 3 6 3 7 6 0 0 0 - 80 -[root@node1 ~]# echo "get weight WEB_PORT_80/web2" | socat stdio /var/lib/haproxy/haproxy.sock
3 (initial 3)#修改weight,注意只针对单进程有效
[root@node1 ~]# echo "set weight WEB_PORT_80/web2 2" | socat stdio /var/lib/haproxy/haproxy.sock[root@node1 ~]# echo "get weight WEB_PORT_80/web2" | socat stdio /var/lib/haproxy/haproxy.sock
2 (initial 3)#将后端服务器禁用,注意只针对单进程有效
[root@node1 ~]# echo "disable server WEB_PORT_80/web2" | socat stdio /var/lib/haproxy/haproxy.sock#将后端服务器软下线,即weight设为0
[root@node1 ~]# echo "set weight WEB_PORT_80/web1 0" | socat stdio /var/lib/haproxy/haproxy.sock#将后端服务器禁用,针对多进程
[root@node1 ~]# vim /etc/haproxy/haproxy.cfg
......
stats socket /var/lib/haproxy/haproxy1.sock mode 600 level admin process 1
stats socket /var/lib/haproxy/haproxy2.sock mode 600 level admin process 2               nbproc 2
.....

static-rr:基于权重的轮询调度,不支持权重的运行时利用socat进行动态调整及后端服务器慢启动,其后端主机数量没有限制,相当于LVS中的 wrr

listen  web_hostbind 10.0.0.7:80,:8801-8810,10.0.0.7:9001-9010mode httplog globalbalance static-rrserver web1  10.0.0.17:80 weight 1 check inter 3000 fall 2 rise 5server web2  10.0.0.27:80 weight 2 check inter 3000 fall 2 rise 5

first:根据服务器在列表中的位置,自上而下进行调度,但是其只会当第一台服务器的连接数达到上限,新请求才会分配给下一台服务,因此会忽略服务器的权重设置,此方式使用较少

listen  web_hostbind 10.0.0.7:80,:8801-8810,10.0.0.7:9001-9010mode httplog globalbalance firstserver web1  10.0.0.17:80 maxconn 2 weight 1 check inter 3000 fall 2 rise 5server web2  10.0.0.27:80 weight 1 check inter 3000 fall 2 rise 5

测试访问效果:

#同时运行下面命令,观察结果
# while  true;do  curl http://10.0.0.7/index.html ; sleep 0.1;done

2、动态算法

动态算法:基于后端服务器状态进行调度适当调整,优先调度至当前负载较低的服务器,且权重可以在haproxy运行时动态调整无需重启。

roundrobin:基于权重的轮询动态调度算法,支持权重的运行时调整,不同于lvs中的rr轮训模式,HAProxy中的roundrobin支持慢启动(新加的服务器会逐渐增加转发数),其每个后端backend中最多支持4095个real server,支持对real server权重动态调整,roundrobin为默认调度算法

listen  web_hostbind 10.0.0.7:80,:8801-8810,10.0.0.7:9001-9010mode httplog globalbalance roundrobinserver web1  10.0.0.17:80 weight 1  check inter 3000 fall 2 rise 5server web2  10.0.0.27:80 weight 2  check inter 3000 fall 2 rise 5

支持动态调整权重:
 

# echo "get weight web_host/web1" | socat stdio /var/lib/haproxy/haproxy.sock 
1 (initial 1)# echo "set weight web_host/web1 3" | socat stdio /var/lib/haproxy/haproxy.sock # echo "get weight web_host/web1" | socat stdio /var/lib/haproxy/haproxy.sock 
3 (initial 1)

leastconn:加权的最少连接的动态,支持权重的运行时调整和慢启动,即当前后端服务器连接最少的优先调度(新客户端连接),比较适合长连接的场景使用,比如:MySQL等场景。

listen  web_host  bind 10.0.0.7:80,:8801-8810,10.0.0.7:9001-9010  mode http  log global  balance leastconn  server web1  10.0.0.17:80 weight 1  check inter 3000 fall 2 rise 5  server web2  10.0.0.27:80 weight 1  check inter 3000 fall 2 rise 5

random:在1.9版本开始增加一个叫做random的负载平衡算法,其基于随机数作为一致性hash的key,随机负载平衡对于大型服务器场或经常添加或删除服务器非常有用,支持weight的动态调整,weight较大的主机有更大概率获取新请求

listen  web_hostbind 10.0.0.7:80,:8801-8810,10.0.0.7:9001-9010mode httplog globalbalance  randomserver web1  10.0.0.17:80 weight 1  check inter 3000 fall 2 rise 5server web2  10.0.0.27:80 weight 1  check inter 3000 fall 2 rise 5

3、其他算法

其它算法即可作为静态算法,又可以通过选项成为动态算法

(1)source

源地址hash,基于用户源地址hash并将请求转发到后端服务器,后续同一个源地址请求将被转发至同一个后端web服务器。此方式当后端服务器数据量发生变化时,会导致很多用户的请求转发至新的后端服务器,默认为静态方式,但是可以通过hash-type支持的选项更改

这个算法一般是在不插入Cookie的TCP模式下使用,也可给拒绝会话cookie的客户提供最好的会话粘性,适用于session会话保持但不支持cookie和缓存的场景

源地址有两种转发客户端请求到后端服务器的服务器选取计算方式,分别是取模法和一致性hash

(2)map-base取模法:

map-based:取模法,对source地址进行hash计算,再基于服务器总权重的取模,最终结果决定将此请求转发至对应的后端服务器。此方法是静态的,即不支持在线调整权重,不支持慢启动,可实现对后端服务器均衡调度。缺点是当服务器的总权重发生变化时,即有服务器上线或下线,都会因总权重发生变化而导致调度结果整体改变,hash-type 指定的默认值为此算法。

所谓取模运算,就是计算两个数相除之后的余数,10%7=3, 7%4=3
map-based算法:基于权重取模,hash(source_ip)%所有后端服务器相加的总权重

listen  web_hostbind 10.0.0.7:80,:8801-8810,10.0.0.7:9001-9010mode tcplog globalbalance sourcehash-type map-based server web1  10.0.0.17:80 weight 1  check inter 3000 fall 2 rise 3server web2  10.0.0.27:80 weight 1  check inter 3000 fall 2 rise 3[root@haproxy ~]# echo "set weight web_host/10.0.0.27 10" | socat stdio /var/lib/haproxy/haproxy.sock 
Backend is using a static LB algorithm and only accepts weights '0%' and '100%'.[root@haproxy ~]# echo "set weight web_host/10.0.0.27 0" | socat stdio /var/lib/haproxy/haproxy.sock [root@haproxy conf.d]# echo "get weight web_host/10.0.0.27" | socat stdio /var/lib/haproxy/haproxy.sock 
0 (initial 1)

(3)一致性hash

一致性哈希,当服务器的总权重发生变化时,对调度结果影响是局部的,不会引起大的变动,hash(o)mod n ,该hash算法是动态的,支持使用 socat等工具进行在线权重调整,支持慢启动

算法:

1、key1=hash(source_ip)%(2^32)  [0---4294967295]
2、keyA=hash(后端服务器虚拟ip)%(2^32)
3、将key1和keyA都放在hash环上,将用户请求调度到离key1最近的keyA对应的后端服务器

hash环偏斜问题:

增加虚拟服务器IP数量,比如:一个后端服务器根据权重为1生成1000个虚拟IP,再hash。而后端服务器权重为2则生成2000的虚拟IP,再bash,最终在hash环上生成3000个节点,从而解决hash环偏斜问题。

Hash对象到后端服务器的映射关系:

后端服务器在线与离线的调度方式:

一致性hash配置示例:

listen  web_hostbind 10.0.0.7:80,:8801-8810,10.0.0.7:9001-9010mode tcplog globalbalance sourcehash-type consistentserver web1  10.0.0.17:80 weight 1  check inter 3000 fall 2 rise 5server web2  10.0.0.27:80 weight 1  check inter 3000 fall 2 rise 5

(4)uri

基于对用户请求的URI的左半部分或整个uri做hash,再将hash结果对总权重进行取模后,根据最终结果将请求转发到后端指定服务器,适用于后端是缓存服务器场景,默认是静态,也可以通过hash-type指定map-based和consistent,来定义使用取模法还是一致性hash。

注意:此算法是应用层,所有只支持 mode http ,不支持 mode tcp

<scheme>://<user>:<password>@<host>:<port>/<path>;<params>?<query>#<frag>
左半部分:/<path>;<params>
整个uri:/<path>;<params>?<query>#<frag>

uri 取模法配置示例
listen  web_hostbind 10.0.0.7:80,:8801-8810,10.0.0.7:9001-9010mode httplog globalbalance uriserver web1  10.0.0.17:80 weight 1  check inter 3000 fall 2 rise 5server web2  10.0.0.27:80 weight 1  check inter 3000 fall 2 rise 5
uri 一致性hash配置示例
listen  web_hostbind 10.0.0.7:80,:8801-8810,10.0.0.7:9001-9010mode httplog globalbalance urihash-type consistentserver web1  10.0.0.17:80 weight 1  check inter 3000 fall 2 rise 5server web2  10.0.0.27:80 weight 1  check inter 3000 fall 2 rise 5
访问测试

访问不同的uri,确认可以将用户同样的请求转发至相同的服务器

# curl  http://10.0.0.7/test1.html
# curl  http://10.0.0.7/test2..html
url_param

url_param对用户请求的url中的 params 部分中的一个参数key对应的value值作hash计算,并由服务器总权重相除以后派发至某挑出的服务器;通常用于追踪用户,以确保来自同一个用户的请求始终发往同一个real server,如果无没key,将按roundrobin算法

假设:
url = http://www.openlab.com/foo/bar/index.php?key=value则:
host = "www.openlab.com"
url_param = "key=value"
url_param取模法配置示例
listen  web_hostbind 10.0.0.7:80,:8801-8810,10.0.0.7:9001-9010mode httplog globalbalance url_param  userid     #url_param hashserver web1 10.0.0.17:80 weight 1  check inter 3000 fall 2 rise 5server web2 10.0.0.27:80 weight 1  check inter 3000 fall 2 rise 5
url_param一致性hash配置示例
listen  web_hostbind 10.0.0.7:80,:8801-8810,10.0.0.7:9001-9010mode httplog globalbalance url_param  userid             #对url_param的值取hashhash-type consistentserver web1  10.0.0.17:80 weight 1  check inter 3000 fall 2 rise 5server web2  10.0.0.27:80 weight 1  check inter 3000 fall 2 rise 5
测试访问
# curl  http://10.0.0.7/index.html?userid=<NAME_ID> 
# curl  "http://10.0.0.7/index.html?userid=<NAME_ID>&typeid=<TYPE_ID>" 

(5)hdr

针对用户每个http头部(header)请求中的指定信息做hash,此处由 name 指定的http首部将会被取出并做hash计算,然后由服务器总权重取模以后派发至某挑出的服务器,如无有效的值,则会使用默认的轮询调度。

hdr取模法配置示例
listen  web_hostbind 10.0.0.7:80,:8801-8810,10.0.0.7:9001-9010mode httplog globalbalance hdr(User-Agent)#balance hdr(host)server web1  10.0.0.17:80 weight 1  check inter 3000 fall 2 rise 5server web2  10.0.0.27:80 weight 1  check inter 3000 fall 2 rise 5
一致性hash配置示例
listen  web_hostbind 10.0.0.7:80,:8801-8810,10.0.0.7:9001-9010mode httplog globalbalance hdr(User-Agent)hash-type consistentserver web1  10.0.0.17:80 weight 1  check inter 3000 fall 2 rise 5server web2  10.0.0.27:80 weight 1  check inter 3000 fall 2 rise 5
访问测试
[root@centos6 ~]#curl -v http://10.0.0.7/index.html
[root@centos6 ~]#curl -vA 'firefox' http://10.0.0.7/index.html
[root@centos6 ~]#curl -vA 'chrome' http://10.0.0.7/index.html

(6)rdp-cookie

rdp-cookie对远windows远程桌面的负载,使用cookie保持会话,默认是静态,也可以通过hash-type指定map-based和consistent,来定义使用取模法还是一致性hash。

rdp-cookie取模法配置示例
listen RDPbind 10.0.0.7:3389balance rdp-cookiemode tcpserver rdp0 10.0.0.17:3389 check fall 3 rise 5 inter 2000 weight 1
rdp-cookie一致性hash配置示例
[root@haproxy ~]#cat /etc/haproxy/conf.d/windows_rdp.cfg 
listen openlab_RDP_3389bind  172.16.0.100:3389balance rdp-cookiehash-type consistentmode tcpserver rdp0 10.0.0.200:3389 check fall 3 rise 5 inter 2000 weight 1[root@haproxy ~]#hostname -I
10.0.0.100 172.16.0.100 
基于iptables实现RDP协议转发

必须开启ip转发功能:net.ipv4.ip_forward = 1

[root@centos8 ~]#sysctl -w net.ipv4.ip_forward=1
#客户端和Windows在不同网段需要下面命令
[root@centos8 ~]#iptables  -t nat -A PREROUTING -d 172.16.0.100 -p tcp --dport 3389 -j DNAT --to-destination 10.0.0.200:3389#客户端和Windows在同一网段需要再执行下面命令
[root@centos8 ~]#iptables  -t nat -A PREROUTING -d 10.0.0.8 -p tcp --dport 3389 -j DNAT --to-destination 10.0.0.1:3389
[root@centos8 ~]#iptables -t nat -A POSTROUTING -s 10.0.0.0/24 -j SNAT --to-source  10.0.0.8

4、算法总结

static-rr--------->tcp/http  静态
first------------->tcp/http  静态roundrobin-------->tcp/http 动态
leastconn--------->tcp/http 动态
random------------>tcp/http 动态以下静态和动态取决于hash_type是否consistent
source------------>tcp/http
Uri--------------->http
url_param--------->http     
hdr--------------->http
rdp-cookie-------->tcp

各算法使用场景:

first       #使用较少static-rr   #做了session共享的web集群
roundrobin
randomleastconn   #数据库
source      #基于客户端公网IP的会话保持Uri--------------->http  #缓存服务器,CDN服务商,蓝汛、百度、阿里云、腾讯
url_param--------->http hdr         #基于客户端请求报文头部做下一步处理rdp-cookie  #很少使用

http://www.mrgr.cn/news/92377.html

相关文章:

  • 网络基础知识-2
  • go基础语法
  • Java常见设计模式(中):结构型模式
  • 最小化重投影误差求解PnP
  • 【蓝桥杯】第十五届省赛大学真题组真题解析
  • SQL注入(order by,limit),seacms的报错注入以及系统库的绕过
  • 登录逻辑结合redis
  • 第二十:【路由的props配置】
  • java后端开发day19--学生管理系统升级
  • C#使用Semantic Kernel:接入本地deepseek-r1
  • CSS—引入方式、选择器、复合选择器、文字控制属性、CSS特性
  • 深度学习pytorch之19种优化算法(optimizer)解析
  • java后端开发day18--学生管理系统
  • Language Models are Unsupervised Multitask Learners,GPT-2详细讲解
  • 【考研】复试相关上机题目
  • React进阶之前端业务Hooks库(三)
  • 使用elasticdump导出/导入 -- ES数据
  • WEB1~6通杀
  • 2024-2025 学年广东省职业院校技能大赛 “信息安全管理与评估”赛项 技能测试试卷(四)
  • 毕业项目推荐:基于yolov8/yolo11的苹果叶片病害检测识别系统(python+卷积神经网络)