Hi, @Timothy Spann, the Phoenix interpreter has been removed in favor of the generic JDBC interpreter. Please see the 0.6 snapshot docs here for an example of using it to connect to Phoenix. ... View more

bash> cat /path/to/my_script.sql !outputformat csv select count(*) from system.catalog /usr/hdp/current/phoenix-client/bin/sqlline.py zk_host:2181:/hbase-unsecure /path/to/my_script.sql > out.txt bash> cat out.txt 83/83 (100%) Done 'COUNT(1)' '87' ... View more

@Sunile Manjee, you can read and write to Phoenix from NiFi. You'll need to setup a DBConnectionPool ControllerService (click the Wrench/ScrewDriver icon on the top right) pointing to the phoenix-client.jar. Then you can use ExecuteSQL for reads. To write, the following pattern works: ConvertAttributesToJSON->ConvertJSONToSQL in Insert mode->ReplaceText to replace "INSERT" with "UPSERT" -> PutSQL. For details on connecting to Phoenix on a Kerberized cluster, see the instructions at phoenix.apache.org (use your browser's search/find feature, Kerberos details are about 1/3 the way down the page). ... View more