无法启动Hadoop守护进程：内存不足(cannot start Hadoop daemons: Insufficient memory)

 起初我可以启动守护进程并正确运行作业，然后无处可去，我无法启动守护进程（start-dfs，start-yarn）。 运行.sh后终端等待（如图http://imgur.com/Sr5I5aw ）。 停止的唯一方法是ctrl + c。 日志hs_error_pidxxxx.log说明内存不足（ http://imgur.com/3e3VolG ）。 
 
 我在网站上尝试了一些建议，例如添加交换内存，重新启动。 我仍然无法启动守护进程。 
 
 这里有一些结论（如果由于我糟糕的沟通技巧，有人可能会感到困惑）： 
 
 
 
 我的vm有4GB的内存，一开始大约3.5免费。 
 
 
 我可以在同一个虚拟机上正确运行守护进程。 
 
 
 提前感谢您的帮助。 
 
 PS。 我在Ubuntu 14.04上使用Hadoop 2.5.1和HBase 0.98.11 

At first I could be able to start daemons and run jobs properly, then out of nowhere, I cant start the daemons (start-dfs, start-yarn). After running .sh the terminal waits forever (as in the picture http://imgur.com/Sr5I5aw). The only way to stop is ctrl+c. The logs hs_error_pidxxxx.log says something about insufficient memory (http://imgur.com/3e3VolG). 
 
I tried some advises found in sites, such as adding swap memory, rebooting. I still cant start the daemons.
 
Here're some in conclusion (in case someone might be confused due to my bad communication skill):
 
 
 
My vm has 4gb of memory with about 3.5 free at first.
 
 
I could be able to run daemons properly on the very same vm.
 
 
Thank you in advance for every help.
 
PS. I'm using Hadoop 2.5.1 with HBase 0.98.11 on Ubuntu 14.04

原文：https://stackoverflow.com/questions/29872340