日志收集 demo 基础镜像

Requirements

1. Java - 1.8.x

2. Maven - 3.x.x

Steps to setup

1. Clone the application

https://github.com/liyongjian5179/log-collection-demo.git

2. Build and run the app using maven

cd log-collection-demo
mvn package
java -jar target/log-collection-demo-0.0.1-SNAPSHOT.jar

You can also run the app without packaging it using -

mvn spring-boot:run

编译命令

mvn -U clean package -Dsonar.skip=true -Dmaven.test.skip=true

启动命令

java -jar target/log-collection-demo-0.0.1-SNAPSHOT.jar

接口访问

curl http://localhost:8080/

返回："Success! Master"

curl http://localhost:8080/hello

返回："Hello, Master"

Prometheus 监控项

curl http://localhost:8081/actuator/prometheus

返回：

# HELP jvm_gc_live_data_size_bytes Size of old generation memory pool after a full GC
# TYPE jvm_gc_live_data_size_bytes gauge
jvm_gc_live_data_size_bytes{application="log-collection-demo",} 0.0
# HELP jvm_gc_memory_allocated_bytes_total Incremented for an increase in the size of the young generation memory pool after one GC to before the next
# TYPE jvm_gc_memory_allocated_bytes_total counter
jvm_gc_memory_allocated_bytes_total{application="log-collection-demo",} 5.4525952E7
# HELP system_cpu_count The number of processors available to the Java virtual machine
# TYPE system_cpu_count gauge
system_cpu_count{application="log-collection-demo",} 4.0
# HELP tomcat_sessions_alive_max_seconds  
# TYPE tomcat_sessions_alive_max_seconds gauge
tomcat_sessions_alive_max_seconds{application="log-collection-demo",} 0.0
# HELP jvm_classes_unloaded_classes_total The total number of classes unloaded since the Java virtual machine has started execution
# TYPE jvm_classes_unloaded_classes_total counter
jvm_classes_unloaded_classes_total{application="log-collection-demo",} 1.0
# HELP jvm_classes_loaded_classes The number of classes that are currently loaded in the Java virtual machine
# TYPE jvm_classes_loaded_classes gauge
jvm_classes_loaded_classes{application="log-collection-demo",} 7899.0
# HELP jvm_threads_peak_threads The peak live thread count since the Java virtual machine started or peak was reset
# TYPE jvm_threads_peak_threads gauge
jvm_threads_peak_threads{application="log-collection-demo",} 22.0
# HELP system_load_average_1m The sum of the number of runnable entities queued to available processors and the number of runnable entities running on the available processors averaged over a period of time
# TYPE system_load_average_1m gauge
system_load_average_1m{application="log-collection-demo",} 0.11
...

Grafana 监控 JMX 模板

Spring Boot 2.1 Statistics

https://grafana.com/grafana/dashboards/10280

健康检查

curl http://localhost:8081/actuator/health

返回：{"status":"UP"}

构建镜像

docker build -t log-collection-demo .

启动容器

docker run -d  --name=log-collection-demo -p8080:8080 log-collection-demo

远程镜像

registry.cn-beijing.aliyuncs.com/liyongjian5179/log-collection-demo:latest

本地测试

直接执行 ./build_start.sh 脚本即可

日志输出路径

/usr/local/logs

业务日志收集方案

1. filebeat + sidecar 的方式

cd log-sidecar
kubectl apply -f deployment.yaml

2. filebeat + daemonset 的方式(默认不收集 k8s 系统日志)

cd log-daemonset
kubectl apply -f daemonset.yaml
kubectl apply -f java-demo.yaml

应用通过 annotations 的方式开启日志收集， filebeat.harvest: "true" 为开启， filebeat.index: "xxxxxxxx" 写成对应的服务名即可

annotations:
  filebeat.harvest: "true"
  filebeat.index: "xxxxxxxx"

3. fluentd 收集所以容器日志

参照 https://github.com/fluent/fluentd-kubernetes-daemonset 即可

$ git clone https://github.com/fluent/fluentd-kubernetes-daemonset