Identifying Duplicate Transactions

David MeegoFrom the Useful SQL Scripts Series.

Because Microsoft Dynamics GP is designed to have separate tables for WORK, OPEN and HISTORY transactions, there are times where it is possible for a transaction record to exist in more than one table.  This is usually the result of an error or interruption.

These duplicate records become an issue when it comes time to move a transaction from one table to another. For example: when it is posted, fully applied, or paid transaction removal is used.  At this time, the duplicate will generate a duplicate key error and the process will be aborted.

Another time when duplicate records can cause problems is when using an Inquiry window which can show data from more than one of the WORK, OPEN and HISTORY tables at the same time.  To achieve this, the Inquiry windows transfer data into a single temporary table and use this table for their display.  If duplicate records exist, you will get an error when the data is being copied into the temporary table.

An error message that may be generated by several inquiry windows, when there are duplicates, refers to a createSQLTmpTable stored procedure. This stored procedure does not actually exist, but is in fact referring to pass through SQL script called from the Dexterity code.  For example: The error message from the Payables Transaction Inquiry window is 

The stored procedure createSQLTmpTable returned the following results: DBMS: 2627, Microsoft Dynamics GP: 0.

Note: Error 2627 is a SQL Cannot insert duplicate key error.

To make it easier to find duplicates, I am providing the following SQL queries to look for duplicate headers in the core Distribution and Financial modules.  While these will not find every possible duplicate in every table, they are a great basis for checking for duplicate transactions.

SQL Script to look for duplicate headers in SOP, IVC, POP, RM, PM, IV, & GL modules

-- SOP Duplicates
select SOPTYPE, SOPNUMBE, COUNT(*) as [COUNT] from
(
select SOPTYPE, SOPNUMBE from SOP10100 W
UNION ALL
select SOPTYPE, SOPNUMBE from SOP30200 H
) C
group by SOPTYPE, SOPNUMBE
having COUNT(*) > 1

-- IVC Duplicates
select DOCTYPE, INVCNMBR, COUNT(*) as [COUNT] from
(
select DOCTYPE, INVCNMBR from IVC10100 W
UNION ALL
select DOCTYPE, INVCNMBR from IVC30101 H
) C
group by DOCTYPE, INVCNMBR
having COUNT(*) > 1

-- POP PO Duplicates
select PONUMBER, COUNT(*) as [COUNT] from
(
select PONUMBER from POP10100 W
UNION ALL
select PONUMBER from POP30100 H
) C
group by PONUMBER
having COUNT(*) > 1

-- POP Receivingss Duplicates
select POPRCTNM, COUNT(*) as [COUNT] from
(
select POPRCTNM from POP10300 W
UNION ALL
select POPRCTNM from POP30300 H
) C
group by POPRCTNM
having COUNT(*) > 1

-- RM Duplicates
select RMDTYPAL, DOCNUMBR, COUNT(*) as [COUNT] from
(
select RMDTYPAL, RMDNUMWK as DOCNUMBR from RM10301 W
UNION ALL
select RMDTYPAL, DOCNUMBR from RM10201 W
UNION ALL
select RMDTYPAL, DOCNUMBR from RM20101 O
UNION ALL
select RMDTYPAL, DOCNUMBR from RM30101 H
) C
group by RMDTYPAL, DOCNUMBR
having COUNT(*) > 1

-- PM Duplicates
select DOCTYPE, VCHRNMBR, COUNT(*) as [COUNT] from
(
select DOCTYPE, VCHNUMWK as VCHRNMBR from PM10000 W
UNION ALL
select DOCTYPE, VCHRNMBR from PM10300 P
UNION ALL
select DOCTYPE, VCHRNMBR from PM10400 M
UNION ALL
select DOCTYPE, VCHRNMBR from PM20000 O
UNION ALL
select DOCTYPE, VCHRNMBR from PM30200 H
) C
group by DOCTYPE, VCHRNMBR
having COUNT(*) > 1

-- IV Duplicates
select IVDOCTYP, DOCNUMBR, COUNT(*) as [COUNT] from
(
select IVDOCTYP, IVDOCNBR as DOCNUMBR from IV10000 W
UNION ALL
select IVDOCTYP, DOCNUMBR from IV30200 H
) C
group by IVDOCTYP, DOCNUMBR
having COUNT(*) > 1

-- GL Duplicates
select JRNENTRY, RCTRXSEQ, SEQNUMBR, ACTINDX, TRXDATE, CRDTAMNT, DEBITAMT, [YEAR], COUNT(*) as [COUNT] from
(
select WH.JRNENTRY, WH.RCTRXSEQ, WL.SQNCLINE as SEQNUMBR, WL.ACTINDX, TRXDATE, CRDTAMNT, DEBITAMT, WH.OPENYEAR as [YEAR] from GL10000 WH JOIN GL10001 WL ON WL.JRNENTRY = WH.JRNENTRY
UNION ALL
select JRNENTRY, RCTRXSEQ, SEQNUMBR, ACTINDX, TRXDATE, CRDTAMNT, DEBITAMT, OPENYEAR as [YEAR] from GL20000 O
UNION ALL
select JRNENTRY, RCTRXSEQ, SEQNUMBR, ACTINDX, TRXDATE, CRDTAMNT, DEBITAMT, HSTYEAR as [YEAR] from GL30000 H
) C
group by JRNENTRY, RCTRXSEQ, SEQNUMBR, ACTINDX, TRXDATE, CRDTAMNT, DEBITAMT, [YEAR]
having COUNT(*) > 1

Once the duplicate records have been identified, you will need to use SQL queries to check which of the transactions are the correct ones. Sometimes a duplicate transaction only has the key fields entered and the rest of the fields are blank and/or there are no line records associated with the header. 

Once you have identified what is the incorrect data and have made a backup, you can remove the duplicate data using Transact-SQL commands.

The script is also available as an attachment at the bottom of this post.

You might also want to look at the Automated Solutions, the links are at the bottom of the General Articles & Links page.

Let me know if you find this useful.

David

SQL Duplicates Check.zip