Hapoxy cluster service setup

Hapoxy colony

1. Hapoxy brief introduction

HAProxy It's a use C Free and open source software written in language [1], It provides high availability 、 Load balancing , And based on TCP and HTTP Application proxy for .

HAProxy Especially suitable for those with heavy load web Site , These sites usually need session persistence or seven layers of processing .HAProxy Running on the current hardware , It can support tens of thousands of concurrent connections . And its running mode makes it easy and safe to integrate into your current architecture , And protect your web The server is not exposed to the network .

HAProxy Implemented an event driven , Single process model , This model supports a very large number of concurrent connections . The multiprocess or multithreaded model is limited by memory 、 System scheduler restrictions and ubiquitous lock restrictions , Rarely handle thousands of concurrent connections . Event driven model because of better resource and time management of user space (User-Space) Achieve all of these tasks , So there are no such problems . The disadvantage of this model is , On a multicore system , These programs usually have poor scalability . That's why they have to optimize to Make each CPU Time slice (Cycle) Do more work .

Include GitHub、Bitbucket[3]、Stack Overflow[4]、Reddit、Tumblr、Twitter5 and Tuenti[7] Famous websites inside , And Amazon Web services HAProxy.

haproxy_ Baidu Encyclopedia (baidu.com)

2. Haproxy The construction of clusters

2.1 Environment configuration

Service configuration	ip
Haproxy	192.168.80.20
web The server 1	192.168.80.30
web The server 2	192.168.80.35

2.2 The configuration process

1. Turn off firewall , And transmit compressed packets

systemctl stop firewalld
setenforce 0

haproxy-1.5.19.tar.gz

Copy

2. Compilation and installation Haproxy

yum install -y pcre-devel bzip2-devel gcc gcc-c++ make

tar zxvf haproxy-1.5.19.tar.gzcd haproxy-1.5.19/
make TARGET=linux2628 ARCH=x86_64
make install

Copy

3.Haproxy Server configuration

mkdir /etc/haproxy
cp examples/haproxy.cfg /etc/haproxy/

cd /etc/haproxy/
vim haproxy.cfg
_______________________________
global
--4~5 That's ok -- modify , Configure logging ,local0 For log devices , It is stored in the system log by default 
		log /dev/log   local0 info		
        log /dev/log   local0 notice
        #log loghost    local0 info
        maxconn 4096					# maximum connection , Need to consider ulimit -n Limit 
--8 That's ok -- notes ,chroot Running path , Self set root directory for the service , Generally, this line needs to be commented out 
        #chroot /usr/share/haproxy
        uid 99							# user UID
        gid 99							# user GID
        daemon							# Daemons mode 

defaults        
		log     global					# Define the log as global Log definition in configuration 
        mode    http					# The model is http
        option  httplog					# use http Log format log 
        option  dontlognull				# Do not record health check log information 
        retries 3						# Check node server failures , Three consecutive failures , The node is considered unavailable 
        redispatch						# When the server load is high , Automatically end connections that have been queued for a long time 
        maxconn 2000					# maximum connection 
        contimeout      5000			# Connection timeout 
        clitimeout      50000			# Client timeout 
        srvtimeout      50000			# Server timeout 

-- Delete all below listen term --, add to 
listen  webcluster 0.0.0.0:80			# Define a name webcluster Application 
        option httpchk GET /index.html	# Check the... Of the server test.html file 
        balance roundrobin				# The load balancing scheduling algorithm uses the polling algorithm roundrobin
        server inst1 192.168.80.30:80 check inter 2000 fall 3		# Define online nodes 
        server inst2 192.168.80.35:80 check inter 2000 fall 3

Copy

4. Add system services

cp /opt/haproxy-1.5.19/examples/haproxy.init /etc/init.d/haproxycd /etc/init.dchmod +x haproxy
chkconfig --add /etc/init.d/haproxyln -s /usr/local/sbin/haproxy /usr/sbin/haproxy
service haproxy start	 or 	/etc/init.d/haproxy start

Copy

2.3 Configuration screenshot

2.3.1haproxy Server configuration

2.3.2 Web Server node configuration

————————————————————————————192.168.80.30 To configure echo "this is WEB SERVER11111" > /var/www/html/index.html
systemctl start httpd

————————————————————————————192.168.80.35 To configure echo "this is WEB SERVER22222" > /var/www/html/index.html
systemctl start httpd

Copy

2.3.3 Client access test

3. Haproxy Cluster log redefinition

haproxy The log of is output to the system by default syslog in , In the previous step, the logs have been defined in /dev/log in , Pictured

Need modification rsyslog To configure , For ease of management . take haproxy The relevant configuration is defined independently to haproxy.conf, And on the /etc/rsyslog.d/ Next ,rsyslog All configuration files in this directory will be loaded automatically at startup .

## This part of the configuration will haproxy Of info The log goes to /var/log/haproxy/haproxy-info.log Next , take notice The log goes to /var/log/haproxy/haproxy-notice.log Next .“&~” Indicates that after the log is written to the log file ,rsyslog Stop processing this message .vim /etc/rsyslog.d/haproxy.conf
___________________________________________
if ($programname == 'haproxy' and $syslogseverity-text == 'info')
then -/var/log/haproxy/haproxy-info.log
&~
if ($programname == 'haproxy' and $syslogseverity-text == 'notice')
then -/var/log/haproxy/haproxy-notice.log

Copy

mkdir /var/log/haproxy/
systemctl restart rsyslog.service

tail -f /var/log/haproxy/haproxy-info.log			# see haproxy Access request log information for 

Copy

3.1 The operation screenshot

be based on 2 The operation of

1. modify rsyslog To configure

2. establish /var/log/haproxy/ Directory and restart the service

3. Use client access web service .

4. Review the log generation again

4. Comparison of several load balancing implementations （Nginx-LVS-Haproxy）

4.1 Nginx

Nginx The advantages of ：

• Working on the Internet 7 Layer above , Can target http Apply some diversion strategies , For example for domain names 、 Directory structure .Nginx Regular rules are better than HAProxy More powerful and flexible .
• Nginx The dependence on network stability is very small , Theoretically, it can ping The load function can be carried out through ,LVS It depends on the stability of the network , Stability is relatively high .
• Nginx install and configure 、 The test is simple 、 convenient , There are clear logs for investigation and management ,LVS Configuration of 、 The test will take a long time .
• Can bear high load pressure and stable , It can support tens of thousands of concurrent times , Load ratio LVS Relatively small .
• Nginx The internal fault of the server can be detected through the port , For example, according to the server processing the status code returned by the web page 、 Timeout and so on .
• Nginx It's not just a good load balancer / Reverse agent software , It's also powerful Web application server .
• Nginx As Web Reverse acceleration caching is becoming more and more mature , Faster than the traditional Squid The server is faster , It is used as a reverse proxy accelerator in many scenarios .
• Nginx As a static web page and image server , The performance in this area is excellent , At the same time, there are many third-party modules .

Nginx The shortcomings of ：

• Nginx Only supported http、https and Email agreement , In this way, the scope of application is smaller .
• Health check for back-end servers , Only through the port to detect , No support for passing url To detect .
• I won't support it Session Keep it straight , Need to pass through ip_hash and cookie Guide to solve .

4.2 LVS

LVS The advantages of ：

• Strong resistance to load 、 It is work in the network 4 Above the layer for distribution only , There is no flow . So the best performance in load balancing software , The memory and cpu Resource consumption is relatively low .
• LVS Stable work , Because of its strong anti load ability , It has a complete hot standby scheme .
• No flow ,LVS Distribute requests only , And traffic doesn't go out of itself , This guarantees the equalizer IO The performance of will not be affected by large traffic .
• It has a wide range of applications , because LVS Working in 4 layer , So it can do load balancing for almost all applications , Include http、 Database etc. .

LVS The shortcomings of ：

• The software itself does not support regular expression processing , We can't separate the static from the dynamic . relatively speaking ,Nginx/HAProxy+Keepalived It has obvious advantages .
• If the website application is relatively large ,LVS/DR+Keepalived It's more complicated to implement . relatively speaking ,Nginx/HAProxy+Keepalived It's much simpler .

4.3 HAProxy

HAProxy The advantages of ：

• HAProxy It also supports virtual hosts .
• HAProxy Support 8 Load balancing strategies .
• HAProxy The advantages of can complement Nginx Some shortcomings of , Such as support Session To keep ,Cookie The boot , At the same time, it supports obtaining the specified url To detect the state of the back-end server .
• HAProxy Follow LVS similar , It's just a load balancing software , In terms of pure efficiency HAProxy than Nginx Better load balancing speed , It is also superior in concurrent processing Nginx Of .
• HAProxy Support TCP Load balanced forwarding of protocol .