@John It looks like it is complaining about the JDBC driver. So can you please try this: SEVERE: The web application [/oozie] registered the JDBC driver [com.mysql.jdbc.Driver] but failed to unregister it when the web application was stopped. To prevent a memory leak, the JDBC Driver has been forcibly unregistered. It can happen if the Oozie Server was not stopped properly, So please try to Kill the Oozie process properly and then restart it freshly. Also can you please check what is the "mysql" jdbc driver version present here "/usr/hdp/current/oozie-server/oozie-server/webapps/oozie/WEB-INF/lib/" May be you can try tp updated the jdbc driver jar inside "/usr/hdp/current/oozie-server/libext/" and fthen prepare new war file. 1. Stop oozie server 2. Run below command # /usr/hdp/current/oozie-server/bin/oozie-setup.sh prepare-war 3. Start Oozie server. . Still if it fails then can you please share the full log ... View more

@Bala Kolla What is the exact ambari version that you are using? One of the reason may be due to change in the repository. Looks like some of the packages might be installed from a different repo. May be you can try checking if there are some packages installed from multiple HDP repositories? # repoquery -a --installed --qf "%{ui_from_repo} %{name}" | grep -i '^@hdp' Also if the you just changed the REPO url (not the actual HDP version) then may be you can try this. (Not sure what is your Ambari version , and hence can not tell for sure if the following may work or not) 1. On the ambari agent node where the client installation is failing try to search for the file "/usr/lib/ambari-agent/lib/resource_management/libraries/script/script.py" 2. Then try to comment a line in this file which is setting package_version = None Example: # grep 'package_version = None'  /usr/lib/ambari-agent/lib/resource_management/libraries/script/script.py     package_version = None 3. After commenting out that line your file grep output should look like following: # grep 'package_version = None'  /usr/lib/ambari-agent/lib/resource_management/libraries/script/script.py     # package_version = None 4. Remove the "script.pyo and script.pyc" files. (or move it to /tmp or some other directory) # rm -f /usr/lib/ambari-agent/lib/resource_management/libraries/script/script.pyc # rm -f /usr/lib/ambari-agent/lib/resource_management/libraries/script/script.pyo 5. Restart ambari-agent on that node. # ambari-agent restart. 6. Try starting component on this node from ambari UI and then see if it still fails? If it still fails then it will require further investigation. ... View more

@Malleh Ceesay The URL which you are typing is incorrect the substring in the URL "ambari_common/tasks.logrotate.yml" should be actually "ambari_common/tasks/logrotate.yml" Example: # wget -O /etc/logrotate.d/metron-ambari  https://raw.githubusercontent.com/apache/metron/master/metron-deployment/ansible/roles/ambari_common/tasks/logrotate.yml . ... View more

@Monalisa Tripathy Is this what you are looking out for? tez.am.log.level => Root logging level passed to the Tez Application Master. yarn.app.mapreduce.am.log.level => The logging level for the MR ApplicationMaster. The allowed levels are: OFF, FATAL, ERROR, WARN, INFO, DEBUG, TRACE and ALL ... View more