Jellyfish是一款轻量级的, 用Java写的, 分布式微服务实时监控系统，可无缝对接Spring Boot或Spring Cloud工程
Jellyfish提供的监控功能主要分为两个部分：
- 应用程序日志统一收集和查询
- 统计和监控SpringBoot应用程序HTTP接口3个指标：请求耗时，错误率，并发度
Jellyfish部署方式
Jellyfish分为服务端和agent端
服务端通常是一个独立的SpringBoot应用或集群，集群模式是通过另一个微服务分布式协作框架tridenter实现的，Jellyfish服务端还需要部署Redis和ElasticSearch
Agent端 通常是加入了jellyfish相关jar包的另一组SpringBoot应用或集群，它们向服务端实时发送数据包。
Jellyfish的特性：
- Jellyfish服务端基于流式计算框架vortex（底层是Netty4），低延迟，高并发，且支持TCP和HTTP两种传输协议
- Jellyfish服务端的网络通信层依赖vortex框架，支持动态水平扩展，支持数据最终一致性
- Jellyfish的应用接口统计是准实时的，纯内存计算，默认的统计时间窗口为1分钟（可自定义），但统计的接口数量变多可能会影响实时性，经测试，最大延迟保持在1分钟内，平均延迟在秒级。
Jellyfish系列一共有4个子工程：
- jellyfish-console
Jellyfish Web控制台，可独立运行，通常作为服务端，当然你也可以自定义服务 端，通过使用注解@EnableJellyfishServer - jellyfish-http-spring-boot-starter
jar包，Jellyfish监控应用程序接口进行指标统计的Agent程序包 - jellyfish-slf4j
jar包，Jellyfish对接slf4j（目前只实现了logback对接），统一应用收集日志的
agent程序包 - jellyfish-spring-boot-starter Jellyfish核心jar包, 实现了全部的核心功能
Install
<dependency>
<artifactId>jellyfish-spring-boot-starter</artifactId>
<groupId>com.github.paganini2008.atlantis</groupId>
<version>1.0-RC3</version>
</dependency>
Jellyfish-console使用说明
如何对接slf4j, 统一收集日志：
你的应用程序需要：
- pom.xml 添加依赖
<dependency>
<artifactId>jellyfish-slf4j</artifactId>
<groupId>com.github.paganini2008.atlantis</groupId>
<version>1.0-RC3</version>
</dependency>
- 修改logback.xml
<?xml version="1.0" encoding="utf-8" ?>
<configuration>
<appender name="logTracker" class="indi.atlantis.framework.jellyfish.slf4j.logback.HttpTransportClientAppender">
<applicationName>tester5</applicationName> <!-- 修改成你的applicationName -->
<brokerUrl>http://192.168.159.1:10010</brokerUrl> <!-- 修改Jellyfish console地址 -->
</appender>
<!-- omit other configuration -->
<root level="info">
<appender-ref ref="STDOUT" />
<appender-ref ref="INFO" />
<appender-ref ref="ERROR" />
<appender-ref ref="logTracker" /> <!-- 引用你定义的appender -->
</root>
</configuration>
目前只支持logback, 未来会支持更多
-
先启动Jellyfish-console
启动命令：java -jar jellyfish-console-1.0-RC1.jar
默认端口：6100 地址：http://localhost:6100/jellyfish/log/
image.png -
再启动你的应用程序，此时日志会源源不断的流入Jellyfish Console：
image.png
说明：
jellyfish服务端从agent端接受来的每行日志数据会保存在es里。
如何对接Spring Boot或Spring Cloud项目，统计HTTP接口指标？
你的应用需要：
- 修改你的pom.xml, 添加：
<dependency>
<artifactId>jellyfish-http-spring-boot-starter</artifactId>
<groupId>com.github.paganini2008.atlantis</groupId>
<version>1.0-RC3</version>
</dependency>
- 在application.properties加入：
atlantis.framework.jellyfish.brokerUrl=http://localhost:6100 # Jellyfish console地址
-
使用压测工具测试某一个接口，如图：
接口列表：
image.png
点击【View】，会打开一直统计详情页，页面比较长
首先在在顶部，会看到接口概览和一些统计
image.png
接着会看到一些仪表盘：
image.png
上图中仪表盘分别显示当前的并发度，QPS, 平均响应时间，接着是两个饼图，分别监控http状态码和接口执行结果
然后往下翻，会看到几个大图，都是按分钟统计的
响应时间统计和QPS统计
image.png
最后显示的是接口调用次数统计和Http状态码统计
image.png
按F12，查看返回的json数据：
{
"msg": "ok",
"data": {
"20:05:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623585900737
}
},
"20:06:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623585960737
}
},
"20:07:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623586020737
}
},
"20:08:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623586080737
}
},
"20:09:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623586140737
}
},
"20:10:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623586200737
}
},
"20:11:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623586260737
}
},
"20:12:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623586320737
}
},
"20:13:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623586380737
}
},
"20:14:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623586440737
}
},
"20:15:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623586500737
}
},
"20:16:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623586560737
}
},
"20:17:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623586620737
}
},
"20:18:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623586680737
}
},
"20:19:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623586740737
}
},
"20:20:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623586800737
}
},
"20:21:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623586860737
}
},
"20:22:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623586920737
}
},
"20:23:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623586980737
}
},
"20:24:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623587040737
}
},
"20:25:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623587100737
}
},
"20:26:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623587160737
}
},
"20:27:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623587220737
}
},
"20:28:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623587280737
}
},
"20:29:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623587340737
}
},
"20:30:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623587400737
}
},
"20:31:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623587460737
}
},
"20:32:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623587520737
}
},
"20:33:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623587580737
}
},
"20:34:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623587640737
}
},
"20:35:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623587700737
}
},
"20:36:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623587760737
}
},
"20:37:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623587820737
}
},
"20:38:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623587880737
}
},
"20:39:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623587940737
}
},
"20:40:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623588000737
}
},
"20:41:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623588060737
}
},
"20:42:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623588120737
}
},
"20:43:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623588180737
}
},
"20:44:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623588240737
}
},
"20:45:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623588300737
}
},
"20:46:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623588360737
}
},
"20:47:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623588420737
}
},
"20:48:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623588480737
}
},
"20:49:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623588540737
}
},
"20:50:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623588600737
}
},
"20:51:00": {
"rt": {
"middleValue": 0,
"count": 0,
"highestValue": 0,
"lowestValue": 0,
"timestamp": 1623588660737
}
},
"20:52:00": {
"rt": {
"middleValue": 2158,
"count": 68,
"highestValue": 4941,
"lowestValue": 1,
"timestamp": 1623588778333
}
},
"20:53:00": {
"rt": {
"middleValue": 2252,
"count": 1815,
"highestValue": 4996,
"lowestValue": 0,
"timestamp": 1623588839757
}
},
"20:54:00": {
"rt": {
"middleValue": 2233,
"count": 2680,
"highestValue": 5000,
"lowestValue": 0,
"timestamp": 1623588899641
}
},
"20:55:00": {
"rt": {
"middleValue": 2306,
"count": 2605,
"highestValue": 4994,
"lowestValue": 0,
"timestamp": 1623588959700
}
},
"20:56:00": {
"rt": {
"middleValue": 2224,
"count": 2669,
"highestValue": 5422,
"lowestValue": 0,
"timestamp": 1623589019949
}
},
"20:57:00": {
"rt": {
"middleValue": 2239,
"count": 2678,
"highestValue": 4997,
"lowestValue": 0,
"timestamp": 1623589079640
}
},
"20:58:00": {
"rt": {
"middleValue": 2233,
"count": 2678,
"highestValue": 4998,
"lowestValue": 0,
"timestamp": 1623589139980
}
},
"20:59:00": {
"rt": {
"middleValue": 2305,
"count": 2601,
"highestValue": 4998,
"lowestValue": 0,
"timestamp": 1623589199745
}
},
"21:00:00": {
"rt": {
"middleValue": 2227,
"count": 2678,
"highestValue": 5000,
"lowestValue": 0,
"timestamp": 1623589259579
}
},
"21:01:00": {
"rt": {
"middleValue": 2243,
"count": 2677,
"highestValue": 4998,
"lowestValue": 0,
"timestamp": 1623589319569
}
},
"21:02:00": {
"rt": {
"middleValue": 2255,
"count": 2656,
"highestValue": 5000,
"lowestValue": 0,
"timestamp": 1623589378556
}
},
"21:03:00": {
"rt": {
"middleValue": 2288,
"count": 2616,
"highestValue": 5000,
"lowestValue": 0,
"timestamp": 1623589439666
}
},
"21:04:00": {
"rt": {
"middleValue": 1272,
"count": 152,
"highestValue": 4271,
"lowestValue": 0,
"timestamp": 1623589441456
}
}
},
"success": true,
"statusCode": 200
}
Jellyfish接口统计部分调用了vortex的metric模块的sdk,，所以是基于时间序列的，默认统计时间窗口是1分钟，滚动保存前60条数据，历史数据默认不落地，但可以通过实现MetricEvictionHandler和MetricSequencerFactory接口进行扩展，参考vortex metrics
最后，Jellyfish服务端的参考配置：
spring.application.name=jellyfish-console
spring.application.cluster.name=jellyfish-console-cluster
#management.endpoints.web.exposure.include=*
#management.endpoint.health.show-details=always
#management.security.enabled=false
#spring.boot.admin.client.url=http://localhost:10081
atlantis.framework.vortex.bufferzone.collectionName=jellyfish
atlantis.framework.vortex.bufferzone.pullSize=100
#Elasticsearch Configuration
spring.data.elasticsearch.cluster-name=es
spring.data.elasticsearch.cluster-nodes=localhost:9300
spring.data.elasticsearch.repositories.enabled=true
spring.data.elasticsearch.properties.transport.tcp.connect_timeout=60s
#Redis Configuration
atlantis.framework.redis.host=localhost
atlantis.framework.redis.port=6379
atlantis.framework.redis.password=123456
atlantis.framework.redis.database=0
#Freemarker Configuration
spring.freemarker.enabled=true
spring.freemarker.suffix=.ftl
spring.freemarker.cache=false
spring.freemarker.charset=UTF-8
spring.freemarker.template-loader-path=classpath:/META-INF/templates/
spring.freemarker.expose-request-attributes=true
spring.freemarker.expose-session-attributes=true
spring.freemarker.setting.number_format=#
spring.freemarker.setting.url_escaping_charset=UTF-8
具体更详细的参数设置请参考：https://github.com/paganini2008/jellyfish
