sqoop-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Nirmal Kumar <nirmal.ku...@impetus.co.in>
Subject RE: Sqoop Incremental Import with lastmodified mode giving Duplicate rows for updated rows
Date Tue, 07 Oct 2014 04:53:29 GMT
Thanks Venkat,

I have posted this in the Hortonworks Forums as well:
http://hortonworks.com/community/forums/topic/sqoop-incremental-import-lastmodified-giving-duplicate-rows-for-updated-rows/

I tried the HDP documentation for incremental import.
But the info given in the HDP documentation is very less.

The below  info is that all given in the bk_HortonworksConnectorForTeradata.pdf for incremental
import.

Incremental Import
Teradata incremental import emulates the check-column and last value options. Here is an
example for a table which has 'hire_date' as the date column to check against and 'name' as
the column that can be used to partition the data.

export USER=dbc
export PASS=dbc
export HOST=<dbhost>
export DB=<dbuser>
export TABLE=<dbtable>
export JDBCURL=jdbc:teradata://$HOST/DATABASE=$DB
export IMPORT_DIR=<hdfs-dir to import>
export VERBOSE=--verbose
export MANAGER=org.apache.sqoop.teradata.TeradataConnManager
export CONN_MANAGER="--connection-manager $MANAGER"
export CONNECT="--connect $JDBCURL"
MAPPERS="--num-mappers 4"
DATE="'1990-12-31'"
FORMAT="'yyyy-mm-dd'"
LASTDATE="cast($DATE as date format $FORMAT)"
SQOOPQUERY="select * from employees where hire_date < $LASTDATE AND \
$CONDITIONS"
$SQOOP_HOME/bin/sqoop import $TDQUERY $TDSPLITBY $INPUTMETHOD $VERBOSE
$CONN_MANAGER $CONNECT -query "$SQOOPQUERY" --username $USER --password $PASS
--target-dir $IMPORT_DIR --split-by name

Values of $TDQUERY $TDSPLITBY $INPUTMETHOD are confusing.

Is there some more info about Incremental Imports in HDP ?

Thanks,
-Nirmal


________________________________
From: Venkat Ranganathan <vranganathan@hortonworks.com>
Sent: Monday, October 6, 2014 10:35 PM
To: user@sqoop.apache.org
Subject: Re: Sqoop Incremental Import with lastmodified mode giving Duplicate rows for updated
rows

Hi Nirmal

hdp connector for TD is a HDP specific work.  Please use the vendor forums for this.
Last modified is not supported as specified in Sqoop.   The HDP documentation has an example
of doing this using queries.  Please look at the documentation

Thanks

Venkat

On Mon, Oct 6, 2014 at 5:29 AM, Nirmal Kumar <nirmal.kumar@impetus.co.in<mailto:nirmal.kumar@impetus.co.in>>
wrote:
Hi All,

I*m trying to do an Incremental Import using Sqoop from Teradata to Hive tables.

I*m using:
-Apache Hadoop 2.4.0
-Apache Hive 0.13.1
-Apache Sqoop 1.4.4
-hdp-connector-for-teradata-1.3.2.2.1.5.0-695-distro
-Teradata 15.0.0.8

From Sqoop documentation:
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 每last-value are imported.

I followed the below steps:

STEP 1: One time activity
I*m doing a full import of the table to hive table.

STEP 2: One time activity
Created a Sqoob Job for incremental import
sqoop job 每create incr1 〞 import 每connection-manager org.apache.sqoop.teradata.TeradataConnManager
每connect jdbc:teradata://192.168.199.137/testdb123<http://192.168.199.137/testdb123>
每username testdb123 每password testdb123 每table Paper_STAGE 每incremental lastmodified 每check-column
last_modified_col 每last-value ※2014-10-03 15:29:48.66∪ 每split-by id 每hive-table paper_stage
每hive-import

STEP 3: This will be done on timely basis from any Scheduler OR Oozie
Executing the Sqoob Job for incremental import everytime I need the updated rows/newly added
rows.
sqoop job 每exec incr1

The source table has a ※unique primary key§ and ※last modified column§ with current timestamp.
The newly added rows though are working fine and getting imported but for the updated rows
I*m getting duplicate rows.
Sqoop is not updating the updated rows but adding a new one with same Id and new current timestamp.

Is this something which is currently not supported in Sqoop as of now ?
This is since I found these:

http://stackoverflow.com/questions/19093417/sqoop-import-lastmodified-gives-duplicate-records-it-doesnt-merger
http://grokbase.com/p/cloudera/cdh-user/13a4n03jrh/sqoop-import-lastmodified-gives-duplicate-records-merger-does-not-happen
https://groups.google.com/a/cloudera.org/forum/#!topic/cdh-user/xAbXEduvahU
https://issues.cloudera.org/browse/DISTRO-464

Is there a way to avoid the duplicate rows for the updated rows and get a merged updated row
for each updated row in the Source table?
Kindly advise me any alternatives to handle this.

Thanks,
-Nirmal

________________________________






NOTE: This message may contain information that is confidential, proprietary, privileged or
otherwise protected by law. The message is intended solely for the named addressee. If received
in error, please destroy and notify the sender. Any use of this email is prohibited when received
in error. Impetus does not represent, warrant and/or guarantee, that the integrity of this
communication has been maintained nor that the communication is free of errors, virus, interception
or interference.


CONFIDENTIALITY NOTICE
NOTICE: This message is intended for the use of the individual or entity to which it is addressed
and may contain information that is confidential, privileged and exempt from disclosure under
applicable law. If the reader of this message is not the intended recipient, you are hereby
notified that any printing, copying, dissemination, distribution, disclosure or forwarding
of this communication is strictly prohibited. If you have received this communication in error,
please contact the sender immediately and delete it from your system. Thank You.

________________________________






NOTE: This message may contain information that is confidential, proprietary, privileged or
otherwise protected by law. The message is intended solely for the named addressee. If received
in error, please destroy and notify the sender. Any use of this email is prohibited when received
in error. Impetus does not represent, warrant and/or guarantee, that the integrity of this
communication has been maintained nor that the communication is free of errors, virus, interception
or interference.
Mime
View raw message