If you make a SIF Put call on a child record during a match batch job, and there is a match path from a parent to the child base object, the match job on the parent base object fails.

If you search for and update a record, you can still search for and find the record through the values from the previous version of the record.

When you add multiple business entities that have searchable columns, performance becomes slower.

If you have hard delete detection enabled, the MDM Hub generates an error when you try to delete all records from one source system.

In WebLogic environments, a Business Entity Service Post call fails to tokenize records.

In the Microsoft SQL Server environment, a match job fails with the following error:

[ERROR] com.siperian.mrm.match.cmxma.GetMatch: Could not open connection to DB: com.microsoft.sqlserver.jdbc.SQLServerException: The instance of the SQL Server Database Engine cannot obtain a LOCK resource at this time.

When the MDM Hub updates STRP tables during tokenization and automerge, the performance of queries is poor.

If you use a user exit to handle ActiveVOS tasks, the

Task Actions

drop-down menu disappears occasionally.

When you run a load job, null values might override values that are not null in columns that do not have trust enabled and in columns that are not selected in the staging table.

The REST POST business entity service request does not add all the child records specified in the request to the business entity.

When you validate an Operational Reference Store by using the Repository Manager, the validation generates errors if dependent lookups are defined on the Part Of child relationships.

If a non-administrative user that does not have read access to searchable fields performs a smart search, the search fails.

Numeric data type does not work for the CleansePut SIF API requests and generates the following error:

SIP-10131: Data access error in running verb CLEANSE_PUT.
SIP-23038: Internal error. The SQL statement generated the SQL error.
Error converting data type nvarchar to numeric. The SQL statement is incorrect.

In the Merge Manager tool, manual trust override might not work.

If you use default values for integer attributes in lookup columns, the load job fails with the following error:

ORA-00932: inconsistent datatypes: expected CHAR got NUMBER

In a Microsoft SQL Server environment, load jobs fail if records contain empty column values.

The records that the SearchMatch and the GetSearchResults SIF APIs fetch are not sorted based on the sort criteria.

When you update records in a subject area and send for approval, a null pointer exception is generated.

If a numeric column is included in the mapping between the landing and the staging tables, the stage job might fail.

In an IBM DB2 environment, when you enable the case insensitive index on base object columns, the Hub Console generates an exception.

In a database that contains data in the Japanese character set, some characters do not display correctly in search results.

In a WebSphere environment, when two sessions close at the same time, the environment becomes unresponsive.

If a mapping has a number column, a stage job fails.

When you create a match rule set with multiple match columns which take multibyte characters, the Generate Match Tokens batch job fails.

SIF Put and CleansePut calls do not update cross-reference records correctly.

A delete API call fails with a null pointer exception.

You cannot view the most recent data change to an ActiveVOS task if you have a different role from the user who updated the task.

If you unmerge a base object with a large volume of cross-reference records with multiple child records, the batch unmerge job fails with an out of memory error message.

A Search Match API calls fails to return any records when the exact match column has a date data type.

When you enable trust and validation rules on a base object column, the MDM Hub drops any custom index that you created on the XREF table for the column.

If you have an Operational Reference Store (ORS) with SSL encryption enabled, the ORS cannot synchronize with the Model repository of the Informatica platform and the MDM Hub generates an error.

cmx.server.encryptionMethod=SSL 
cmx.server.validateServerCertificate=false

In an IBM DB2 environment, if you run a match job with an exact match rule for a base object with high volume data, the transaction log becomes unmanageably large.

When you try to log in to the Hub Console, the following errors might be generated:

SIP-10313: SIP-10314: Couldn't load user console settings due to data access error.
SIP-10324: Unable to close down JDBC objects opened for query.
DSRA9350E: Operation Connection.commit is not allowed during a global transaction.

If an update to a lookup column tries to replace a value with a null value, the update fails. The change is not written to the cross-reference record.

In the legacy Matches view, the list of Potential Matches can contain multiple XREF and Relationship entries for the same item.

When you delete a hierarchy record from the Hub Console, the hierarchy record is not deleted. If the RBO hierarchy table is also enabled for lookups, the value continues to appear in lists in IDD.

When a

removeMatchedRecords

request is called without the required

recordKeyList

parameter, the request deletes all records from the _MTCH table for the base object.

When you run a match job on records with consolidation_ind=9 and consolidation_ind=4, duplicate entries are inserted into the match table.

When you have smart search enabled, if you make an update in the database and then run a load job, the job fails with the following error:

java.rmi.RemoteException: com.siperian.common.SipException: SIP-05001: Database error.
Caused by SQLException: sqlState=99999	errorCode=17006	Invalid column name 

Sometimes an updated cross-reference record does not cause the MDM Hub to fire a message trigger.

When you run a load job on a base object with a staging table that is enabled for cell updates, an error occurs.

In a Microsoft SQL Server and IBM WebSphere environment, if you enable distinct mapping load jobs fail with the following error:

DataAccessException: SIP-10322: Unable to execute the SQL
Invalid column name ROWNUM

If you use a Korean population, and run a match job for Korean data, the Process Server fails.

Geocode matching fails in a match job run on a database that is configured with UTF-8 encoding.

The staging process truncates the values of some records after the decimal point.

When staging tables are dropped and re-created, the best version of the truth (BVT) calculation does not consider the Allow Null Update property settings for the staging table columns. When a staging table is recreated, the cross-reference records loaded by the original staging table during BVT calculation are affected.

When you run a match and merge job, some records remain with consolidation_ind=4 in the base object table.

An endpoint was vulnerable to a potential attack.

After you enable the

cmx.server.kerberos.keytab

property in

cmxserver.properties

file, the following error appears in the

cmxserver.log

:

Trying to get credentials for null
usage: initiate only, use no subject, useAllCred : false
acquiring default creds

When you create a Data Quality cleanse function that uses the StrikeIron service for Dun & Bradstreet

http://ws.strikeiron.com/DCPPremium3?WSDl

, the cleanse call fails with a user ID error.

When a foreign key column is trusted and has a default value, and you create a validation rule to downgrade trust, the best version of the truth calculation might not result in the correct value for this column.

After you run an external match, the proxy role loses all assigned privileges. The EMO table that contained the privileges is empty.

If an update to a lookup column tries to replace a value with a null value, the update fails. The change is not written to the cross-reference record.

The PostMerge user exit fails when it makes a SIF Put call.

When there are a high number of tasks, ActiveVOS times out while retrieving them and no tasks are returned.

When a cleanse function uses a Data Quality mapplet with a decimal data type for the input and output ports, the decimal is considered a string and cannot be mapped to the float data type.

When a cleanse function uses a Data Quality mapplet with a date data type port, the function fails.

When you use a Data Quality mapplet in a cleanse function and run a stage job, the stage job processes and inserts only one record per minibatch size.

The installer installs the SSA-NAME3 library files in the directory

<MDM installation directory>/cleanse/lib/<operating system>

instead of in

<MDM installation directory>/cleanse/lib

.

If you merge parent records and then unmerge them, the last_update_date field in a child record is incorrect.

If you use Kerberos encryption, the Import Schema wizard prompts you for a password instead of a principal.

If you have a message queue configured, a merge task publishes a message but the following error occurs:

SIP-09143: Error invoking addMergedRowid(String)

If you use the automation tool in the Resource Kit to synchronize users through LDAP servers, all the created user names are based on the common name (CN) component instead of the Security Access Manager account name.

The proxy user role is not available.

In the IBM DB2 environment, you cannot run the

create_db

script on any port other than the default port.

When you use WebSphere or JBoss, after you install or upgrade the Hub Server, you need to delete temporary files from the application server

Temp

directory and restart the application server.

When you choose to install the Informatica platform as part of the Hub Server installation process, the prompts for the Informatica platform installation file locations do not appear.