Why don't you set SPARK_DAEMON_MEMORY=2g in spark-env.sh ? ... View more

Why don't you try with string argument like this? hive>select qp("hello"); ... View more

Writing flowfiles to disk will be costly operation as the size of the flow files is huge and there are lots of such flow files. Any alternative solution apart from writing contents of flowfiles to disk? I think you should select the better solution(as I know, there is no best) because both of them have the trade-off. Your questioned solution has no file operation and no writing contents of flowfiles to disk. But, very slo~~~w ! My solution has costly file operation and writing contents of flowfiles to disk. But, Fast ! ... View more

Yes, You did ! ^^ ... View more

You can use ExcecuteProcess for SCP or GetSFTP... ... View more

I mean ... ListDataBaseTables -> GenerateTableFecth -> Execute SQL -> ConvertRecord(Avro to CSV) -> CSV to MySQL server local file -> ExcecuteProcess(MySQL LOAD DATA... command) ... View more

If possible, I recommend MySQL 'LOAD DATA' command instead of putSQL and PutDatabaseRecord. NiFi save csv --> MySQL local CSV --> MySQL LOAD DATA ... MySQL LOAD DATA is dramatically fast ! ... View more

SolrCloud already enabled. Sometimes rager_audit collection created, but not for me. ... View more

A few days ago, I have suffered from exactly same issue. ZK, Solr, Ranger all restarted but there is no collection. Anybody knows the solution? ... View more

I'm sorry about my missunderstand. Zeppelin does not support HA, so you need to manually setup like knox. https://community.hortonworks.com/questions/24961/how-to-configure-ha-for-knox-gateway-using-any-loa.html ... View more