I am getting error when i was migrating table from Oracle to SQL Server for CI(case insensitive) database

Violation of PRIMARY KEY constraint 'LXK_INTERFACE_CONFIG_INFOR_PK'. Cannot insert duplicate key in object 'wmdlxk.LXK_INTERFACE_CONFIG_INFO'. The duplicate key value is (CUSTOMER, OUT, TRILLIUM).

But there is only one record as above in oracle Like  (CUSTOMER, OUT, TRILLIUM).

These are the two different records in oracle db

Customer Out Trillium CleanseURL=http://xxxxxxxxxxxx7654/G/Cleanser/CustomerData/PrimaryCustomer/Address;MatchURL=http://trillium_qa.lex.lexmark.com:7654/G/RMatcher/CustomerData/PrimaryCustomer

and

CUSTOMER OUT TRILLIUM selectMaxRetryCountPROP=5;selectMaxBackoffIntervalPROP=2;selectTimeoutPROP=0;updateMaxRetryCountPROP=5;updateMaxBackoffIntervalPROP=2;updateTimeoutPROP=0;PublishMaxRetryCountPROP=5;PublishMaxBackoffIntervalPROP=2;PublishTimeoutPROP=0;URL=http:/xxxxxxxxxx8510/jboss-net/services/TrilliumWebService

But why sql server accepting both as same records and throwing PK voilation error, though the db is CI and last column values are also different.

Please help me in this regard.

• Edited by Vinai Gandla 1 hour 33 minutes ago

use

SET IDENTITY_INSERT <Tablename> ON. then it will allow to insert the records.
More on : http://msdn.microsoft.com/en-IN/library/ms188059.aspx
Thanks,

I was Using SSMA(SQL SERVER MIGRATION ASSISTANT) tool

Sorry is that contents of multiple columns you posted? If yes then first two values seems to be same Customer OutTrillium  vs CUSTOMER OUTTRILLIUM  since collation is CI ie case insensitive.

So if your pk is on these two columns then it will fail

I am getting error when i was migrating table from Oracle to SQL Server for CI(case insensitive) database

Violation of PRIMARY KEY constraint 'LXK_INTERFACE_CONFIG_INFOR_PK'. Cannot insert duplicate key in object 'wmdlxk.LXK_INTERFACE_CONFIG_INFO'. The duplicate key value is (CUSTOMER, OUT, TRILLIUM).

But there is only one record as above in oracle Like  (CUSTOMER, OUT, TRILLIUM).

It looks like you have other records in the Oracle database that differ only by case (e.g. 'customer out' or 'Customer Out').  You'll need to specify a SQL Server case-sensitive collation for the problem column if you need to retain these keys that differ only by case.  Perhaps the root cause is a data quality issue in the source database where only a single "CUSTOMER OUT" row was intended.  In that case, it would be best to combine and/or delete the problem source data.

Thanks 

Thanks