================================================================================

Oracle Endeca Server (2.3.0 Revision A)

  * Documentation Update: The "Oracle Endeca Information Discovery Installation 
    Guide" has been updated to require that secure mode (SSL) must be enabled 
    as the final step in the installation of Oracle Endeca Server.


================================================================================

Oracle Endeca Server

GENERAL (2.3.0)

  - Release Date: April 2012
  
  - This release is not compatible with data indexed using previous releases.

  - The release notes may have been updated since the release date.
    Access the Oracle Documentation page at http://docs.oracle.com for the 
    most recent version of the release notes.


INSTALLATION (2.3.0)
 
  - See the "Oracle Endeca Information Discovery Installation Guide" for
    installation information.


MIGRATION (2.3.0)
 
  - See the "Oracle Endeca Information Discovery Migration Guide" for
    information on migrating to this release.


BUG FIXES (2.3.0)

  * LCORE-80: The Configuration Web Service now includes a new operation 
    updateGroupConfigs that lets you update specified assignments (slices) of 
    the group configuration record, including the display name, for a specified
    group. The group is specified to the operation by listing its
    system-group_Key. Using this operation, you can now change the display name
    of system_navigation_InternalGroup (this group is automatically created to
    contain all attributes that do not belong to any user-created groups).

  * LCORE-107: Previously, the Bulk Load Interface could ingest data sets that
    included non-XML characters (nonprintable control characters U+0000-U+0008,
    U+000B-U+000C, and U+000E-U+001F; the Unicode surrogate blocks; and the 
    Unicode BOM codepoints U+FFFE and U+FFFF), even though such data will cause
    errors at query time. This was fixed so that now the Bulk Load Interface 
    rejects non-XML 1.0 characters upon ingest.

  * LCORE-141: The final step of a bulk load in the Dgraph is to perform a
    merge of all the data and update the spelling dictionary. This step can be 
    time consuming for large data sets. So in the case where the ingest pipeline
    is composed of a number of sequential bulk loads, this work is redundant for
    all but the last bulk load. To allow the user to control the data merges and
    dictionary updates, the Bulk Load Interface is updated with option settings 
    that can disable the final merge as well as the dictionary update. These 
    options can be set in the Integrator Bulk Add/Replace Records connector.

  * LCORE-165: The DGraph process version shown when you issue the 
    admin?op=stats request is now accurate.

  * LCORE-179 and LCORE-187: Under certain circumstance (such as ingesting a 
    very wide record), the Dgraph process may hang or fail. This issue has 
    been resolved.

  * LCORE-214: The WSDL for the Conversation Service incorrectly listed the
    default value of the RefinementConfig/Expose attribute to "true".

  * LCORE-239: Analytics query compilation could take a long time, possibly 
    appearing to hang, consuming 100% CPU. This problem was caused by growth 
    of intermediate state during compilation. This state is now compacted 
    periodically, preventing the apparent hang.

  * LCORE-289: This fix speeds up query checking and optimization time for
    Endeca Query Language (EQL) queries that have a large number of columns 
    in a single statement.
  
  * LCORE-295: WSDL requests previously returned 'http' URLs for all requests,
    regardless of whether the client was using https. WSDL addresses now
    default to the same protocol which was used to request the WSDL itself.

  * LCORE-308: The cluster leader now correctly causes the followers to catch
    up to the latest data version upon transaction commit.
	

KNOWN ISSUES (2.3.0)
 
  * LCORE-317: It is possible to attach multiple data stores in read-write
    mode to the same data files. Doing so without setting up a clustered
    environment can corrupt your data files.


===============================================================================