MRS Hive 内存溢出问题解决方案
Hive 内存溢出问题解决方案
l Mapper/Reducer阶段JVM内存溢出
1. Java.lang.OutOfMemoryError: GC overhead limit exceeded;
2. Error: java heapspace
3. running beyondphysical memory limits.Current usage: 4.3 GB of 4.3 GBphysical memory used; 7.4 GB of 13.2 GB virtual memory used. Killing container
修复方案:运行时
set mapreduce.map.java.opts=-XmxXXXXm(数值调大)
set mapreduce.map.memory.mb = XXXX(数值调大) (Container内存)
set mapreduce.reduce.java.opts=-XmxXXXXm (数值调大)
set mapreduce.reduce.memory.mb = XXXX(数值调大) (Container内存)
set 语句运行时报错: cannot modify XXXXXXXXX at runtime, it is not in list of params that are allowed to be modified at runtime
在此处填写需要加入白名单的参数,重启Hive
l 栈内存溢出
1. Java.lang.StackOverflowError
修复方案:Sql过于复杂,优化sql
l MRAppMaster内存不足
1. Halting due to Out of Memory Error…
修复方案:
修改此二处配置,数值调大,然后重启yarn
- 点赞
- 收藏
- 关注作者
评论(0)