aws常用的集群管理服務有：
在emr 2.x,3.x中所有服務守護程式都通過 init.d 指令碼作為 sysv init 系統的一部分執行，查詢服務名稱：
ls /etc/init.d/sudo /etc/init.d/hadoop-yarn-resourcemanager stopsudo /etc/init.d/hadoop-yarn-resourcemanager startsudo /etc/init.d/hadoop-yarn-resourcemanager status在emr4.x,5.x中服務已新增到upstart服務了，查詢服務：
sudo initctl listsudo stop hadoop-yarn-resourcemanagersudo start hadoop-yarn-resourcemanagersudo status hadoop-yarn-resourcemanagersudo initctl list
//tomcat
sudo service tomcat8 stop
sudo service tomcat8 start
sudo service mysqld status
sudo stop zeppelin
sudo stop hue
sudo stop hive-server2
sudo stop hive-hcatalog-server
sudo start hive-hcatalog-server
sudo start hive-server2
sudo stop presto-server
sudo start presto-server
sudo stop livy-server
sudo stop zookeeper-server
sudo initctl list | grep -i hadoop
對於hbase:
#批量stop
sudo service tomcat8 stop
sudo service mysqld stop
sudo stop zeppelin
sudo stop hue
sudo stop hive-server2
sudo stop hive-hcatalog-server
sudo stop presto-server
sudo stop hadoop-yarn-resourcemanager
sudo stop oozie
sudo stop hadoop-hdfs-namenode
sudo stop hadoop-mapreduce-historyserver
sudo stop hadoop-yarn-proxyserver
sudo stop hadoop-yarn-timelineserver
sudo stop hadoop-yarn-nodemanager
sudo stop livy-server
sudo stop zookeeper-server
sudo stop spark-history-server
sudo stop gmetad
Go Web服務重啟
通過go 1.8 提供的net http server.go中的func srv server shutdown ctx context.context error方法可以實現執行中的web服務關閉。如下 監聽埠 啟動服務 關閉後再次啟動 監聽埠，啟動服務 再次啟動服務 通過使用curl，驗證 cu...
服務重啟指令碼
每次都是查詢程序，然後關閉，再重啟，甚是麻煩。於是就寫了個自動重啟服務。思路 1.找到原有的程序的pid，並關閉 2.在後台重啟服務 bin bash process ps ef grep python u service.py grep v grep grep v ppid awk for i i...
elk服務重啟 elk重啟記錄
重啟前 檢查elk程序 ps ef grep elk netstat antp grep listen curl 172.18.233.71 9200 cat nodes?v curl 172.18.233.71 9200 cat health?v 以上檢查elk程序號與埠匹配入主節點在哪個位置，還...