The Cloudera Agent Fails to Start with "Address already in use"
Last updated on JANUARY 20, 2018
Applies to:Big Data Appliance Integrated Software - Version 4.9.0 and later
On BDCS 4.9.1/CDH 5.11 system the Cloudera agent will not start because address already in use. However further investigation shows that the address is held by a Cloudera agent process.
1. On the Cloudera Manager (CM) host, Node 3 by default. cloudera-scm-agent.log shows the error:
Traceback (most recent call last):
File "/usr/lib64/cmf/agent/build/env/lib/python2.6/site-packages/CherryPy-3.2.2-py2.6.egg/cherrypy/process/servers.py", line 187, in _start_http_thread
File "/usr/lib64/cmf/agent/build/env/lib/python2.6/site-packages/CherryPy-3.2.2-py2.6.egg/cherrypy/wsgiserver/wsgiserver2.py", line 1825, in start
error: No socket could be created on ('bdanode03.example.com', 9000) -- [Errno 98] Address already in use
2. However checking who is holding port 9000, finds it to be the Cloudera agent process:
a) Use netstat to find the process holding port 9000: