Hadoop 配置不当引起集群不稳

配置不当内容

  资源配置不当:内存、文件句柄数量、磁盘空间

需要调整的地方

   Hadoop、OS 配置文件

Oversubscription of MR heap caused swap -> Swap caused RegionSrver to time out and die -> Dead RegionServer caused MR tasks to fial until MR job died.

内存配置不当

TT OOME

JT OOME

Native Threads

Thread 配置不当

Fetch 失败

Replicas

磁盘配置不当

No File

Too Many Files