I have a Hive server in my cluster running a MySQL instance.  The logfile /var/run/mysqld/mysqld.log is getting really large and I need to figure out the proper way to prune it.   The server it's running on is CentOS 7.  Can I just 'cat /dev/null > mysqld.log' to clear it out, or are there other considerations?  Can I get away with simply stopping the MySQL instance briefly to delete or 'zero out' the file, or do I need to stop all the services on that Hive master node in Ambari?   Please advise if you can. Thanks, Mike ... View more

I have an Ambari cluster that has 12 members, and I want to add another one.  All O/S are CentOS 7.9, Ambari 2.7.5, HDP 3.1.5.   The host that I want to add to the cluster can't get to the internet, so I added /etc/yum.repos.d/ambari-hdp-1.repo to it.  Contents of that file have entries for HDP, HDP-GPL, and HDP-UTILS on the Ambari master.  The directories exist on the Ambari master, and I can browse to them.  "yum repolist" on that host lists the repos as expected.   When I invoke the Add New Host wizard, it successfully finds the host and says the SSH key is valid.  When I get to the final page of the wizard (right before deploy), it lists the repositories it's going to pull files from.  The repo for HDP-GPL shows the HortonWorks (internet) repo instead of the local repo I put in the host's repo file listed above.  At this point, I cancel out of the wizard.   What step(s) do I need to take to get that repo changed?  Please advise when you can.   Thanks in advance, Mike ... View more

I'm getting the error " message:java.lang.UnsupportedOperationException: Storage schema reading not supported"   I've found this online and similar articles to it: https://knowledge.informatica.com/s/article/579038?language=en_US   That says to "To resolve this issue, add the following property to the hive_site.xml file in the cluster configuration and restart the Data Integration Service: metastore.storage.schema.reader.impl=org.apache.hadoop.hive.metastore.SerDeStorageSchemaReader​" I'm running Hive 3.1.5.  I have hive-site.xml files on all of the hive clients in the cluster as well as on the server.  I don't see an option to add that value to the Hive configs in the Ambari GUI.  Also, when I look at the Ambari GUI, there's no Data Integration Service for Hive.     So, to fix that issue, do I add that line to all of the client hive-site.xml files then stop/start certain Hive servers on the Ambari GUI?  Or is this all command line and outside of that GUI?     Please advise when you can.  TIA       ... View more