ELKstack

downmovies

前言
　　ELKstack抓取分析应用日志的前提是开发要将日志的输出格式改成json格式，否则产出的日志很难进行分析。
　　主要流程： tomcat通过log4j输出日志-->  logstash接收数据并发送到redis -->  logstash从redis获取数据并写入es
tomcat怎样输出应用日志？
tomcat7.0配置log4j日志
　　详细文档http://tomcat.apache.org/tomcat-7.0-doc/logging.html
一、创建log4j.properties文件
　　在$CATALINA_BASE/lib文件夹下，新建log4j.properties文件，内容如下
　　log4j.rootLogger = INFO, CATALINA
# Define all the appenders
log4j.appender.CATALINA = org.apache.log4j.DailyRollingFileAppender
log4j.appender.CATALINA.File = ${catalina.base}/logs/catalina
log4j.appender.CATALINA.Append = true
log4j.appender.CATALINA.Encoding = UTF-8
# Roll-over the log once per day
log4j.appender.CATALINA.DatePattern = '.'yyyy-MM-dd'.log'
log4j.appender.CATALINA.layout = org.apache.log4j.PatternLayout
log4j.appender.CATALINA.layout.ConversionPattern = %d [%t] %-5p %c- %m%n
log4j.appender.LOCALHOST = org.apache.log4j.DailyRollingFileAppender
log4j.appender.LOCALHOST.File = ${catalina.base}/logs/localhost
log4j.appender.LOCALHOST.Append = true
log4j.appender.LOCALHOST.Encoding = UTF-8
log4j.appender.LOCALHOST.DatePattern = '.'yyyy-MM-dd'.log'
log4j.appender.LOCALHOST.layout = org.apache.log4j.PatternLayout
log4j.appender.LOCALHOST.layout.ConversionPattern = %d [%t] %-5p %c- %m%n
log4j.appender.MANAGER = org.apache.log4j.DailyRollingFileAppender
log4j.appender.MANAGER.File = ${catalina.base}/logs/manager
log4j.appender.MANAGER.Append = true
log4j.appender.MANAGER.Encoding = UTF-8
log4j.appender.MANAGER.DatePattern = '.'yyyy-MM-dd'.log'
log4j.appender.MANAGER.layout = org.apache.log4j.PatternLayout
log4j.appender.MANAGER.layout.ConversionPattern = %d [%t] %-5p %c- %m%n
log4j.appender.HOST-MANAGER = org.apache.log4j.DailyRollingFileAppender
log4j.appender.HOST-MANAGER.File = ${catalina.base}/logs/host-manager
log4j.appender.HOST-MANAGER.Append = true
log4j.appender.HOST-MANAGER.Encoding = UTF-8
log4j.appender.HOST-MANAGER.DatePattern = '.'yyyy-MM-dd'.log'
log4j.appender.HOST-MANAGER.layout = org.apache.log4j.PatternLayout
log4j.appender.HOST-MANAGER.layout.ConversionPattern = %d [%t] %-5p %c- %m%n
log4j.appender.CONSOLE = org.apache.log4j.ConsoleAppender
log4j.appender.CONSOLE.Encoding = UTF-8
log4j.appender.CONSOLE.layout = org.apache.log4j.PatternLayout
log4j.appender.CONSOLE.layout.ConversionPattern = %d [%t] %-5p %c- %m%n
# Configure which loggers log to which appenders
log4j.logger.org.apache.catalina.core.ContainerBase.[Catalina].[localhost] = INFO, LOCALHOST
log4j.logger.org.apache.catalina.core.ContainerBase.[Catalina].[localhost].[/manager] =\
INFO, MANAGER
log4j.logger.org.apache.catalina.core.ContainerBase.[Catalina].[localhost].[/host-manager] =\
INFO, HOST-MANAGER二、下载jar包
（一）log4j.jar（1.2.x版本）
　　http://mvnrepository.com/artifact/log4j/log4j/1.2.17
（二）tomcat-juli.jar、tomcat-juli-adapters.jar
　　http://tomcat.apache.org/download-70.cgi
　　注意：下载时一定要下载"extras"的包，它与自带的包有区别，官方说明：
　　This tomcat-juli.jar differs from the default one. It contains the full Apache Commons Logging implementation and thus is able to discover the presence of log4j and configure itself.
三、替换jar包
　　将log4j.jar、tomcat-juli-adapters.jar放入$CATALINA_BASE/lib目录
　　将下载的tomcat-juli.jar替换$CATALINA_HOME/bin目录下的tomcat-juli.jar
　　注意：两个包放置的位置是不同的
四、删除logging.properties
　　删除tomcat默认的日志文件$CATALINA_BASE/conf/logging.properties文件
五、启动tomcat
　　启动后在$CATALINA_BASE/logs目录下就能看到log4j输出的日志
　　

开发自定义应用日志输出
　　一般开发java程序，运行tomcat在应用上，日志格式定义普遍放在工程下：WEB-INF/class 目录下，一般都叫做log4j.properties
　　作者使用的logstash版本为2.3 ，属于最新版，如果小伙伴借鉴时出现异常，可以看下版本，对比官方文档。
　　让开发在log4j.properties中做以下更改就可:
#log4j.rootLogger = [ level ] , appenderName, appenderName, ...
log4j.rootLogger = info,console,infoR,errorR,logstash
#branch logger
log4j.logger.org.springframework=WARN
#log4j.logPath
log4j.logPath=/opt/tomcat_xxxxxweb/log
#console
log4j.appender.console = org.apache.log4j.ConsoleAppender
log4j.appender.console.layout = org.apache.log4j.PatternLayout
log4j.appender.console.layout.ConversionPattern = [%p] %d{yyyy-MM-dd HH:mm:ss} [%l] %m%n
#infoR
log4j.appender.infoR = org.apache.log4j.DailyRollingFileAppender
log4j.appender.infoR.File =${log4j.logPath}/XXXXXX.log
log4j.appender.infoR.layout = org.apache.log4j.PatternLayout
log4j.appender.infoR.layout.ConversionPattern=[%p] %d{yyyy-MM-dd HH:mm:ss} [%l] %m%n
log4j.appender.infoR.datePattern='.'yyyyMMdd'.log'
log4j.appender.infoR.Threshold = INFO
#errorR
log4j.appender.errorR = org.apache.log4j.DailyRollingFileAppender
log4j.appender.errorR.File =${log4j.logPath}/XXXXXError.log
log4j.appender.errorR.layout = org.apache.log4j.PatternLayout
log4j.appender.errorR.layout.ConversionPattern=[%p] %d{yyyy-MM-dd HH:mm:ss} [%l] %m%n
log4j.appender.errorR.datePattern='.'yyyyMMdd'.log'
log4j.appender.errorR.Threshold = ERROR
#logstash
log4j.appender.logstash=org.apache.log4j.net.SocketAppender
log4j.appender.logstash.Port=12201
log4j.appender.logstash.RemoteHost=sz-a-xxxxxlogstash01-logstash-xen.xxxxx.com
log4j.appender.logstash.ReconnectionDelay=60000
log4j.appender.logstash.LocationInfo=true
log4j.appender.logstash.Threshold = INFOlogstash如何接受log4j日志并写入redis？
　　此处将不再介绍logstash的安装，建议使用yum安装管理，使用起来比较方便。
　　可以将logstash配置文件放入/etc/logstash/conf.d 内
logstash配置文件展示：
input {
tcp {
port => "11011"
type => "syslog"
}
log4j {
type => "log4j-json-web"
port=>12202
}
log4j {
type => "log4j-json-tomcat"
port=>12201
}
}
filter {
if [type] == "syslog" {
grok {
match => { "message" => "%{SYSLOGTIMESTAMP:syslog_timestamp} %{SYSLOGHOST:syslog_hostname} %{DATA:syslog_program}(?:\[%{POSINT:syslog_pid}\])?: %{GREEDYDATA:syslog_message}" }
}
}
if [type] == "log4j-json-web" {
json {
source => "message"
}
}
}
output {
if [type] == "syslog" {
redis {
host => "sz-a-xxxxxredis01-redis-xen.xxxxx.com"
data_type => "list"
key => "logstash:syslog-log"
}
}
if [type] == "log4j-json-web" {
redis {
host => "sz-a-xxxxxredis01-redis-xen.xxxxx.com"
data_type => "list"
key => "logstash:xxxxxweb-web-log"
}
}
if [type] == "log4j-json-tomcat" {
redis {
host => "sz-a-xxxxxredis01-redis-xen.xxxxx.com"
data_type => "list"
key => "logstash:xxxxxweb-tomcat-log"
}
}
}　　

　　一个logstash进程可以有多个input输入，然后为每个输入设置type属性，此时需要注意！！日志中不要出现type字段，不然是大坑！！！！
　　ouput可以通过if语句判断input来源并输出到不同的地方
　　redis使用list模式来存储数据，这样就起到缓存的作用，一入一出，不会撑爆redis