Cloudera Manager Reports Bad or Concerning Health on Some Nodes

(Doc ID 1550260.1)

Last updated on AUGUST 08, 2017

Applies to:

Big Data Appliance Integrated Software - Version 2.0.1 to 2.0.1 [Release 2.0]
Linux x86-64

Symptoms

A couple of nodes in the BDA cluster are showing "Concerning" health and one node in "Bad" health in Cloudera Manager. The warning message displayed in Cloudera Manager with the "Concerning" health node is the following:
Cloudera Manager agent estimates that the host's clock is offset from the expected clock by 6 second(s). Warning threshold: 3 second(s).

The following similar behavior may be observed showing the times are not synced up on all servers.

Tto check if this is the issue that is being run into log in as root on the bda cluster node01:

# dcli date
or
# dcli -C date
may show similar output:
xxx.xxx.xx.xx: Thu Apr 18 11:05:30 CDT 2013
xxx.xxx.xx.xx: Thu Apr 18 11:05:32 CDT 2013
xxx.xxx.xx.xx: Thu Apr 18 11:05:33 CDT 2013
xxx.xxx.xx.xx: Thu Apr 18 11:05:31 CDT 2013
xxx.xxx.xx.xx: Thu Apr 18 11:05:25 CDT 2013
xxx.xxx.xx.xx: Thu Apr 18 11:05:21 CDT 2013
...
...

 

Note: If running the command using dcli does not work, you may need to run the following command first from the first node as root user then execute the dcli command:

setup-root-ssh -C

 

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