Use the below command instead. Note that to check whether datanode was put into maintenance mode, you will need to go to the Hosts tab in Ambari, navigate to the host running datanode and look for the maintenance icon there (ie the icon won't show up under Ambari > HDFS > Datanode because its a 'slave' component) curl -u admin:OpsAm-iAp1Pass -H "X-Requested-By:ambari" -i -X PUT -d '{"RequestInfo":{"context":"Turn On Maintenance Mode For DataNode"}, "Body":{"HostRoles":{"maintenance_state":"ON"}}}' http://viceroy10:8080/api/v1/clusters/et_cluster/hosts/serf120int.etops.tllsc.net/host_components/DATANODE ... View more

Can you check if those components are in failed install state? If so you can click on the Hosts tab in ambari and navigate to the host where each of those clients/services are installed. Then from the dropdown next to the tez/slider/sqoop, select "Reinstall" This wiki may also be useful ... View more

These should help https://cwiki.apache.org/confluence/display/RANGER/How+to+configure+Solr+Cloud+with+Kerberos+for+Ranger+0.5 https://community.hortonworks.com/articles/15159/securing-solr-collections-with-ranger-kerberos.html ... View more

+ @bbende @Daniel Hendrix couple of things to check 1. Ambari metrics service is up 2, The metrics in question are appearing in AMS. You can query AMS using phoenix. Here is a collection of useful links on building metrics integration (including official apache page and Bryans blog on this and sample code) ... View more

You can remove it from the PostgresDB by runnign below from Ambari node: su postgres psql \c ambari DELETE FROM ambari.hostcomponentstate WHERE service_name IN ('KNOX'); DELETE FROM ambari.hostcomponentdesiredstate WHERE service_name IN ('KNOX'); DELETE FROM ambari.servicecomponentdesiredstate WHERE service_name IN ('KNOX'); DELETE FROM ambari.servicedesiredstate WHERE service_name IN ('KNOX'); DELETE FROM ambari.clusterservices WHERE service_name IN ('KNOX'); ... View more

The "java.net.BindException:Address already in use" exception is just a warning that can be ignored: it will find a different port on its own if the default one (4040) is in use. From your netstat output it seems like it started on 4041 instead. To find process id of which process if running on those ports you can run below netstat -tulpn | grep 404 You can also check status of Spark on YARN jobs using Resource Manager UI e.g. if running on sandbox you can access it on http://VMs_IP_ADDRESS:8088/cluster and click the Spark AM to get to the Spark UI. Check this thread for more details https://community.hortonworks.com/questions/8257/how-can-i-resolve-it.html ... View more