Nodemanager service is failing on new host of HDP cluster after adding to the cluster.

[#1]

Getting the following message:

ERROR nodemanager.NodeStatusUpdaterImpl (NodeStatusUpdaterImpl.java:serviceStart(229)) - Unexpected error starting NodeStatusUpdater org.apache.hadoop.yarn.exceptions.YarnRuntimeException: Recieved SHUTDOWN signal from Resourcemanager, Registration of NodeManager failed, Message from ResourceManager: Disallowed NodeManager from hostxxx.corpintra.net, Sending SHUTDOWN signal to the NodeManager.

[#2 · Accepted solution]

@Narasimha K Have you added this host to exclude list?

1. Check /etc/hadoop/conf/yarn.exclude . If this host exists in this file remove it
2. Execute yarn rmadmin -refreshNodes so YARN re-reads this configuration file

[#3]

The issue was resolved once the steps.

[#4]

Thanks the issue resolved after following the above steps. After re-adding the node, which was decommissioned earlier, it was crashing to start. YARN>Summary>NodeManagers Status was also showing 1 node with decommissioned status but will the cleanup not happen during the install process?

,

Thanks, Issue resolved after completing the below steps. The newly added host/nodemanager was crashing and the YARN>Summary>NodeManagers Status was showing 1 node in decommissioned status

1. Check /etc/hadoop/conf/yarn.exclude . If this host exists in this file remove it
2. Execute yarn rmadmin -refreshNodes so YARN re-reads this configuration file

when you re-add the host again, should not the file get updated automatically?

[#5]

After decommissioning, recommission the nodemanager and then start it again.