@vsrikanth9  Not exactly now the REALM part was wrong again the rest are okay you substituted  the wrong values here is how it's supposed to be you  see the highlighted part    Configuration snippets may be placed in this directory as well includedir /etc/krb5.conf.d/ [logging] default = FILE:/var/log/krb5libs.log kdc = FILE:/var/log/krb5kdc.log admin_server = FILE:/var/log/kadmind.log [libdefaults] dns_lookup_realm = false ticket_lifetime = 24h renew_lifetime = 7d forwardable = true rdns = false pkinit_anchors = /etc/pki/tls/certs/ca-bundle.crt default_realm = HADOOPSECURITY.COM default_ccache_name = KEYRING:persistent:%{uid} [realms] HADOOPSECURITY.COM = { kdc = p1.bigdata.com admin_server = p1.bigdata.com } [domain_realm] .hadoopsecurity.com = HADOOPSECURITY.COM hadoopsecurity.com = HADOOPSECURITY.COM   Do that and let me know the KDC and Admin server are usually the same 🙂 ... View more

@irfangk1  There are a couple of things to do first check your disk space and inode usage. To rule out permissions, there a are the listings for the relevant directory: $ ls -la /var/lib/mysql Can you share /var/log/mysql/mysql.log What the value of innodb-buffer-pool size in the config file /etc/mysql/my.cnf can you edit the my.cnf by adding   [mysqld] innodb_force_recovery = 1 And then running: sudo systemctl start mysql   Hope that helps   ... View more

@vsrikanth9  1.Your KDC part of the screenshot has an error 🙂 in the domains part just copy and paste the below as is to replace p1.bigdata.com noe the dot(.)  and comma separating the names .hadoopsecurity.com,hadoopsecurity.com   The validation passed because in reality it only test the connectivity ONLY to the KDC server   2. And then the Kadmin part the Admin principal should be the output of your   # kadmin.local Something like admin/admin@hadoopsecurity.com or root/admin@hadoopsecurity.com What ever you chose during the installation of Kerberos after that then launch the recreation of the keytabs and all should be okay. Make sure the KDC server is up and running during this process. Please revert   ... View more

@vsrikanth9    Your krb5.conf entry is wrong  please change it to match the below   [domain_realm] .hadoopsecurity.com = HADOOPSECURITY.COM hadoopsecurity.com = HADOOPSECURITY.COM   The restart the kdc and kadmin # systemctl start krb5kdc.service # systemctl start kadmin.service That should resolve your problem Happy hadooping        ... View more

@irfangk1  There is something wrong I don't see the database entry in your ambari.server.properties  how will it bind ? Something like below server.jdbc.url=jdbc:postgresql://<HOSTNAME>:<PORT>/ambari?ssl=true Can you validate, you are using the embedded postgres right?   ... View more

@Gcima009    When you generate templates in NiFi, they are stripped of all encrypted values. When importing those templates into another NiFi cluster, Check your node that is not starting? has any values in the below parameters? You will have to populate all the processor and controller tasks passwords manually. Backing up flow.xml.gz or flow.tar file will capture the entire flow exactly as it is, encrypted sensitive passwords and all. NiFi will not start if it cannot decrypt these encrypted sensitive properties contained in the flow.xml. When sensitive properties e.g passwords are added they are encrypted using these settings from your nifi.properties file: # security properties # nifi.sensitive.props.key= nifi.sensitive.props.key.protected= nifi.sensitive.props.algorithm=PBEWITHMD5AND256BITAES-CBC-OPENSSL nifi.sensitive.props.provider=BC nifi.sensitive.props.additional.keys= In order to drop your entire flow.xml.gz or flow.tar onto another clean NiFi, these values must all match exactly.   Ref: http://www.contemplatingdata.com/2017/08/28/apache-nifi-sensitive-properties-need-know/ ... View more

@irfangk1    Can you share your ambari.server. properties and your /etc/hosts ... View more

@ThanhP  Good everything is perfect for you now 🙂 You ONLY execute sudo -u hdfs hdfs namenode -format  as a last resort because it's dangerous and not recommended to run that on production cluster as that [re-initializes] formats your Namenode hence deleting all your metadata stored on the NameNode.   Having said that the answer you accepted can't help a member who encounters the same issue "HDFS NameNode won't leave safemode" maybe you should un-accept it and accept your own answer as it's the more realistic answer . Happy hadooping    ... View more