Cloudera agent does not start

[#1]

One of my agents failed to report to the cloudera server. I am using the same config.ini as other agents. 

It gives the following error upon start

service cloudera-scm-agent start
Starting cloudera-scm-agent (via systemctl):  Job for cloudera-scm-agent.service failed because the control process exited with error code. See "systemctl status cloudera-scm-agent.service" and "journalctl -xe" for details.
                                                           [FAILED]

The Python version being used by the CM agent does not seem to match with the installed version

/usr/lib64/cmf/agent/build/env/bin/python
Could not find platform dependent libraries <exec_prefix>
Consider setting $PYTHONHOME to <prefix>[:<exec_prefix>]
Python 2.6.6 (r266:84292, Aug 25 2010, 09:58:24)
[GCC 4.5.0] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>>

python --version
Python 2.7.5

Am I looking in the right direction? 

[#2]

Did you try to remove and add node from the cluster?

[#3]

Hi,

I wonder if you have the right agent package installed for your OS.

- What is your OS version?

- What is the rpm -qa |grep cloudera output?

If you are on an OS that ships with python 2.7 natively, then the agent's virtual environment should also support that version.

Ben

[#4]

Correct idea, it turned out that the EL6 packages were installed on this CentOS7 node. Re-installing the cloudera-manager-agent and cloudera-manager-daemons using the EL7 package resolves the issue.

[#5 · Accepted solution]

Installed Cloudera Manager packages were for RHEL6. Uninstall cloudera-manager-agent and cloudera-manager-daemons. Adjust yum repository in /etc/yum.repos.d/cloudera-manager.repo to EL7 Install CM agent packages, now with the correct OS version.