My Oracle Support Banner

Mammoth Decommission Fails with "Command 'HdfsDecommission' failed for service 'hdfs'" But Cloudera Manager Decommission Step Succeeds (Doc ID 2528889.1)

Last updated on NOVEMBER 08, 2019

Applies to:

Big Data Appliance Integrated Software - Version 4.12.0 and later
Linux x86-64


Decommissioning a BDA host with Mammoth fails as below, where <HOSTNAMEx> is the host being decommissioned.  However in Cloudera Manager the host decommission continues successfully. 

3. Navigating in Cloudera Manager reports that the DataNode for the host being decommissioned is "BUSY".

See this by navigating in Cloudera Manager: hdfs > Instances > DataNodex > State > Busy
The DataNode is busy because the host is being decommissioned.

4. In Cloudera Manager > All Recent Commands, the "Host Decommission Command" is at the step, "Waiting for Decommission to finish", for potentially a long while (hours). All Recent Commands, shows output like below:

5. The DataNode log shows that the DataNode is moving datablocks. And Cloudera Manager Charts for the node being decommissioned shows I/O operations are running.

6. Once the decommission completes in Cloudera Manger:
a) The host is decommissioned in Cloudera Manager.
b) From Cloudera Manager all roles for the host are stopped.




To view full details, sign in with your My Oracle Support account.

Don't have a My Oracle Support account? Click to get started!

In this Document

My Oracle Support provides customers with access to over a million knowledge articles and a vibrant support community of peers and Oracle experts.