In HUE's Job Browser, I can see jobs, which for us is a bunch of different hive queries submitted by several users. How can I view the SQL for a given job_id? When I see a query that's been running for a long time, I want to know what it is. ... View more

I saw in the link below that the character limit for Actions has been expanded from 50 to 128.   https://issues.apache.org/jira/browse/OOZIE-2168   Does CDH 5.7 have this newer version of Oozie that allows 128 character names for Actions? ... View more

From what I understand, for the lastmodified update method, Sqoop selects records where timestamp_column >= last modified timestamp and timestamp column < current_time. Is there a way to customize that current_time upper bound? Can I do something like current_time - 1 hour?    We have transactions being created on one server, then replicated to another server, then Sqooped from there. I noticed some missing data in our cluster today, and suspect replication delay as the root cause.     ... View more

Is it possible to take advantage of partition pruning without specifying the value of the partition column as a constant in the query? Is there a way to get it from another joined small table?   In this example, table wallet_month has 1 record with column month_of_year, that has value '2016-05'. Table transaction_detail is partitioned on month_of_year, and has 400 partitions.   Looking at the Explain plan, below query still uses all 400 paritions from TD.    SELECT count(*) FROM wallet_month WM JOIN transaction_detail TD AND TD.month_of_year = WM.month_of_year   Adding a where clause (WHERE WM.month_of_year = '2016-05') gives the desired behavior of only using 1 partition from TD, but I'd like to know if there's another way. ... View more

Is dynamic SQL possible in Impala? or Hive? And by that I mean the ability to execute a string like a statement, like:   exec ('select 1') ... View more