The configuration of Nginx is mainly to modify the /usr/local/nginx/conf/nginx, conf file.
# Configure users and user groups
User www www
# number of worker processes, it is recommended to set it to the total number of CPU cores.
Work process 2;
# The global error log defines the type, and the log level from low to high is: debug | info | notice | warn | error | crit.
Error_log log /error.log information;
# File that records the ID of the main process
PID/usr/local/nginx/nginx . PID；
# The maximum file descriptor that a process can open. Theoretically, this value should be the maximum number of files that can be opened divided by the number of processes. But because nginx load is not completely balanced,
# So this value is better equal to the maximum number of files that can be opened. Execute sysctl -a | grep fs.file to view the linux file descriptor.
worker _ rlimit _ nofile 65535
# Working mode and maximum number of connections
Event {
# working mode, epoll is used for linux2.6 version 2.6 and above.
Use epoll
# Maximum number of connections allowed by a single process
Worker _ Connection 65535;
}
# Set p _ level2; # Compression level
Gzip_types text/general application /x-javascript text /css application/XML; # Compression type
#upstream is used for load balancing. Configure the address and port number of the server to be polled here. Max _ failures is the number of allowed request failures, and the default value is 1.
#weight is the polling weight, and the access rate of the server can be balanced according to different weights.
Upstream hostname {
Server192.168.2.149: 8080max _ failures = 0 weight =1;
Server192.168.1.9: 8080max _ failures = 0 weight =1;
}
# Host Configuration
Server {
# Listening port
Listen to 80;
# domain name
Server Name Hostname;
# Character set
Character set utf-8;
# Separate access log file
access _ log logs/ 192. 168.2 . 149 . access . log main；
# Reverse the proxy configuration and forward all requests with http://hostname to the target server defined in the upstream.
Location/{
# The domain name configured here must be the same as the upstream before forwarding.
Proxy _ pass http:// host name;
proxy _ set _ header X-Real-IP $ remote _ addr；
}
# Enable nginx Status Monitoring Page
Location /nginxstatus {
stub _ status on
access _ log on
}
# Error page
error _ page 500 502 503 504/50x . html；
location = /50x.html {
Root html
}
}
}
At this point, nginx's basic load balancing configuration is completed. Two tomcat were deployed in the experiment, and then different results were returned when they were visited. When you enter the address in the browser, you can really see different results. The content of nginx configuration file needs further study.