最近,项目提出需求,日志需要固定输出为JSON
格式,以便后端Flink
程序解析.
项目背景
项目为简单的Maven
项目,日志由Filebeat
采集,因此不需要配置输出至Logstash
.
下面为pom.xml
文件中配置的依赖,此处使用logstash-logback-encoder
完成日志格式转换操作.
<dependency>
<groupId>org.slf4j</groupId>
<artifactId>slf4j-api</artifactId>
<version>1.7.28</version>
</dependency>
<dependency>
<groupId>net.logstash.logback</groupId>
<artifactId>logstash-logback-encoder</artifactId>
<version>6.1</version>
</dependency>
<dependency>
<groupId>ch.qos.logback</groupId>
<artifactId>logback-classic</artifactId>
<version>1.2.3</version>
</dependency>
<dependency>
<groupId>ch.qos.logback</groupId>
<artifactId>logback-access</artifactId>
<version>1.2.3</version>
</dependency>
Logback配置
<?xml version="1.0" encoding="UTF-8"?>
<configuration scan="true" scanPeriod="60 seconds" debug="false">
<contextName>probe</contextName>
<!-- 配置文件存储地址 -->
<property name="LOG_PATH" value="./logs"/>
<!-- 读取应用程序配置文件中内容,获取下文所需要的region属性 --->
<property resource="application.properties"/>
<appender name="FILE" class="ch.qos.logback.core.rolling.RollingFileAppender">
<encoder class="net.logstash.logback.encoder.LoggingEventCompositeJsonEncoder">
<providers class="net.logstash.logback.composite.loggingevent.LoggingEventJsonProviders">
<pattern>
<pattern>
{
"date":"%d{yyyy-MM-dd HH:mm:ss.SSS}",
"level":"%level",
<!-- system属性由程序动态确定,通过Slf4j提供的MDC进行具体设置 -->
"system":"%X{system} ",
<!-- 读取配置文件中的属性,并设置至日志中 -->
"region":"${application.region}",
"filepath":"%class:%line",
"msg":"%msg"
}
</pattern>
</pattern>
</providers>
<charset>UTF-8</charset>
</encoder>
<append>true</append>
<!-- 配置日志文件滚动存储策略 -->
<rollingPolicy class="ch.qos.logback.core.rolling.SizeAndTimeBasedRollingPolicy">
<fileNamePattern>${LOG_PATH}/probe.%d{yyyy-MM-dd}.%i.log</fileNamePattern>
<MaxHistory>10</MaxHistory>
<maxFileSize>10MB</maxFileSize>
<totalSizeCap>300MB</totalSizeCap>
</rollingPolicy>
</appender>
<root level="FILE">
<appender-ref ref="console"/>
</root>
</configuration>
上述即为logback.xml
文件中的配置内容,需要注意日志中的region
从配置文件application.properties
中获取,而system
属性则同msg
字段一样由程序动态设置.
MDC设置system
属性
/**
* 衡量单次http请求的时延
*
* @param url 请求地址
*/
private static void measureHttpTimeDelay(String url, Platform platform) {
// MDC中设置system属性
MDC.put("system", platform.toString());
OkHttpClient client = new OkHttpClient();
client.newBuilder().connectTimeout(5, TimeUnit.SECONDS)
.readTimeout(5, TimeUnit.SECONDS)
.build();
Request request = new Request.Builder().url(url).build();
Instant before = Instant.now();
try (Response response = client.newCall(request).execute()) {
Instant after = Instant.now();
if (response.isSuccessful()) {
long duration = Duration.between(before, after).toMillis();
log.info("请求成功!" + response.message() + "时延 = " + duration + "ms");
} else {
log.info("请求失败!" + response.message());
}
} catch (IOException e) {
log.error("get http response failed, url: {}, ex: {}", url, e.getMessage());
}
// MDC中清除system属性设置
MDC.remove("system");
}
上述程序,模拟一个简单的http
请求,并记录相应日志.
需要注意,MDC
是ThreadLocal
实现,因此在多线程环境下使用需要注意.
最终日志
最终日志如下图所示:
{"date":"2019-09-05 21:16:44.643","level":"INFO","system":"ALI_YUN ","region":"HUABEI","filepath":"com.test.Application:38","msg":"请求成功!OK时延 = 439ms"}
{"date":"2019-09-05 21:16:45.326","level":"INFO","system":"HUAWEI_YUN ","region":"HUABEI","filepath":"com.test.Application:38","msg":"请求成功!OK时延 = 165ms"}
{"date":"2019-09-05 21:16:46.513","level":"INFO","system":"ALI_YUN ","region":"HUABEI","filepath":"com.test.Application:38","msg":"请求成功!OK时延 = 485ms"}
{"date":"2019-09-05 21:16:56.130","level":"INFO","system":"HUAWEI_YUN ","region":"HUABEI","filepath":"com.test.Application:38","msg":"请求成功!OK时延 = 419ms"}