Ambari Metrics是否可以使用外部的hbase ... View more

Did you find a solution?  It makes sense to put a protected password field on a single processor (or controller service). The password won't be passed as an attribute along the flow.   It does not make sense to put a (protected) password into a flow attribute, because the password will be exposed.   So just make sure that your nifi/-registry is secure. Make sure that the FTP user has limit capabilities. ... View more

Thanks you! check logs ... View more

@Charles Bernard's suggestion solved it for me. ... View more

Do you have example run Zeppelin REST API ? ... View more

Hi @Dmitro Vasilenko, did u get a chance to look this. @nifi @ExecuteSparkInteractive @spark @Timothy Spann spark.sql("insert into dbname.tablename values('from livy')") is not working spark.sql("insert into dbname.tablename select * from dbname.tablename") is working For more information, please look into my previous posts on same page/tasl/requiement. @nifi @executesparkintactive ... View more

@Dmitro Vasilenko Either consider increasing the lock time out for the backend metastore DB, try using HiveServer2 (it uses embedded metastore). OR just try increasing "hive.lock.sleep.between.retries" to 120s and hive.lock.numretries to a higher value. ... View more

Hi! finded this tables! Thank you! ... View more

@Dmitro Vasilenko Can you try this query without map join . Looks like data size is huge for hybrid grace hash join. set hive.auto.convert.join=false; And then execute this query. ... View more

Why are you using 10g of driver memory? What is the size of your dataset and how many partitions does it have? I would suggest using the config below: --executor-memory 32G \ --num-executors 20 \ --driver-memory 4g \ --executor-cores 3 \ --conf spark.driver.maxResultSize=3g \ ... View more