hadoop之 node manager起不來， 執行mapreduce 程式hang住

現象： node manager起不來， 執行mapreduce 程式hang住

namenode 程序狀态查詢

[root@hadp-master sbin]# jps

8608 ResourceManager

8257 NameNode

9268 Jps

8453 SecondaryNameNode

datanode 程序狀态查詢 ， 發現 nodemanager 沒有起來

[root@hadp-node1 hadoop]# jps

4179 DataNode

4940 Jps

namenode 單獨起 nodemanager

[root@hadp-master sbin]# yarn nodemanager

18/01/26 03:53:54 INFO webapp.WebApps: Web app node started at 8042

18/01/26 03:53:54 INFO client.RMProxy: Connecting to ResourceManager at hadp-master/10.100.25.30:8031

18/01/26 03:53:54 INFO nodemanager.NodeStatusUpdaterImpl: Sending out 0 NM container statuses: []

18/01/26 03:53:54 INFO nodemanager.NodeStatusUpdaterImpl: Registering with RM using containers :[]

18/01/26 03:53:54 ERROR nodemanager.NodeStatusUpdaterImpl: Unexpected error starting NodeStatusUpdater

org.apache.hadoop.yarn.exceptions.YarnRuntimeException: Recieved SHUTDOWN signal from Resourcemanager ,Registration of NodeManager failed, Message from ResourceManager: NodeManager from hadp-master doesn't satisfy minimum allocations, Sending SHUTDOWN signal to the NodeManager. ## 提示 NodeManager 記憶體不足

解決過程：

1、原虛拟機755m記憶體，調大虛拟機記憶體至2G

2、修改 yarn-site.xml ,添加如下參數

<property>

<name>yarn.scheduler.minimum-allocation-mb</name>

<value>200</value>

</property>

<name>yarn.scheduler.maximum-allocation-mb</name>

<value>1600</value>

<name>yarn.nodemanager.resource.memory-mb</name>

<name>yarn.nodemanager.resource.cpu-vcores</name>

<value>1</value>

參數 : yarn.scheduler.minimum-allocation-mb 

預設 : 1024 m 

官方解釋: The minimum allocation for every container request at the RM, in MBs. Memory requests lower than this will throw a InvalidResourceRequestException.

參數 : yarn.scheduler.maximum-allocation-mb

預設 : 8192 m

官方解釋: The maximum allocation for every container request at the RM, in MBs. Memory requests higher than this will throw a InvalidResourceRequestException.

參數 : yarn.nodemanager.resource.memory-mb

官方解釋: Amount of physical memory, in MB, that can be allocated for containers.

參數 : yarn.nodemanager.resource.cpu-vcores

預設 : 8

官方解釋: Number of vcores that can be allocated for containers. This is used by the RM scheduler when allocating resources for containers. This is not used to limit the number of physical cores used by YARN containers.

文章可以轉載，必須以連結形式标明出處。

本文轉自 張沖andy 部落格園部落格，如需轉載請自行聯系 原作者原文連結：http://www.cnblogs.com/andy6/p/8372184.html