On BDCS CDH Cluster, HDFS Service is down and CM Diagnostic Bundle Fails to Collect the Logs (Doc ID 2124087.1)

Last updated on APRIL 15, 2016

Applies to:

Big Data Cloud Service - Version 1.0 and later
Linux x86-64

Symptoms

On BDCS, HDFS service went down.

From service-health.json collected as part of Cloudera Manager(CM) diagnostic bundle

 "entityName" : "hdfs:ClusterName-ns",
 "entityType" : "service",
 "entityVersion" : "CDH 5.4.5",
 "summary" : "DISABLED",

Also all the logs collected by diagnostic bundle indicate connection refused error while collecting the logs

From *support-bundle.zip\logs\<BDANode>\hadoop-cmf-hdfs-NAMENODE-<BDANode>.log.out.gz.error and *support-bundle.zip\process-stdout-stderr\<BDANode>-hdfs-NAMENODE-2933.stderr.log.error

Error collecting log /var/log/hadoop-hdfs/hadoop-cmf-hdfs-NAMENODE-<BDANode>.log.out: Connection refused
..
Error accessing http://<BDANode>:9000/process/2933-hdfs-NAMENODE/files/logs/stderr.log java.net.ConnectException: Connection refused

 

Cause

Sign In with your My Oracle Support account

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

My Oracle Support provides customers with access to over a
Million Knowledge Articles and hundreds of Community platforms