sqoop-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Radhe Radhe <radhe.krishna.ra...@live.com>
Subject Sqoop Incremental mode to get updated data in Hive
Date Fri, 19 Sep 2014 13:41:49 GMT
Hello Experts,
I need to extract data from RDBMS tables to Hive Tables on a timely basis(daily/weekly, etc.).
My tables have a 'Primary Key' but does NOT have a 'last-modified' column.
I plan to go as:FIRST RUN: Will use sqoop-import-all-tables command for importing all the
tables at one go to Hive Tables.EACH SUBSEQUENT RUN: Use Sqoop incremental import mode to
retrieve only rows newer than some previously-imported set of rows.
My question is how can I get the updated rows which got updated in between the FIRST RUN and
NEXT SUBSEQUENT RUN.For e.g. Say in FIRST RUN: I fetched all tables. A CUSTOMER table that
has 100 records with CustomerId as Primary Key(1 to 100) is imported to Hive CUSTOMER table.
And now meanwhile some rows in the Source CUSTOMER table got updated.With my NEXT SUBSEQUENT
RUN this will fetch rows > 100, thus skipping the updated rows which are  < 100. 
How can I get the updated rows on each subsequent run.
Referring to Sqoop documentation this strategy works only if we have a last-modified column(which
in my case don't have):An alternate table update strategy supported by Sqoop is called lastmodified
mode. You should use this when rows of the source table may be updated, and each such update
will set the value of a last-modified column to the current timestamp. Rows where the check
column holds a timestamp more recent than the timestamp specified with --last-value are imported.
Any suggestions on how to get the updated data on SUBSEQUENT RUNS using the Sqoop Incremental
View raw message