Oracle | DBMS

Oracle DBMS_KAFKA Version 26ai

General Information

Library Note

Morgan's Library Page Header

The best Oracle News for FY2026

Oracle Database 26ai will be available on generic Linux platforms in January and soon on AIX and Windows

Purpose Oracle SQL Access to Kafka (OSAK) is a feature of Oracle Database that allows data in Kafka topics to be queried from Oracle SQL. OSAK allows Kafka to be queried in one of three ways, each abstracted as a type of application created using one of three PL/SQL calls: CREATE_LOAD_APP, CREATE_STREAMING_APP, or CREATE_SEEKABLE_APP.

AUTHID CURRENT_USER

Constants

Name	Data Type	Value
Application Types
APPLICATION_TYPE_LOAD	VARCHAR2(30)	'LOAD'
APPLICATION_TYPE_SEEKABLE	VARCHAR2(30)	'SEEKABLE'
APPLICATION_TYPE_STREAMING	VARCHAR2(30)	'STREAMING'
AVRO Options
OPT_AVRODECIMALTYPE	VARCHAR2(30)	'avrodecimaltype'
DSV Options
OPT_BLANKASNULL	VARCHAR2(30)	'blankasnull'
Debugging/Logging Options
OPT_DEBUG	VARCHAR2(30)	'debug'
OPT_LOGEXEC	VARCHAR2(30)	'logexec'
OPT_LOGOPT	VARCHAR2(30)	'logopt'
OPT_LOGQC	VARCHAR2(30)	'logqc'
General Options
OPT_BUFSIZE	VARCHAR2(30)	'bufsize'
OPT_SEPARATOR	VARCHAR2(30)	'separator'
OPT_STRICT	VARCHAR2(30)	'strict'
OPT_TERMINATOR	VARCHAR2(30)	'terminator'
Kafka Value Options
OPT_AVROSCHEMA	VARCHAR2(30)	'avroschema'
OPT_FMT	VARCHAR2(30)	'fmt'
OPT_JSONDT	VARCHAR2(30)	'jsondt'
OPT_REFTABLE	VARCHAR2(30)	'reftable'
Kafka Key Options
OPT_KEYFMT	VARCHAR2(30)	'keyfmt'
Not Implemented Option Character Sets
OPT_CONVERSIONERRS	VARCHAR2(30)	'conversionerrs'
OPT_DATEFMT	VARCHAR2(30)	'datefmt'
OPT_ESCAPEDBY	VARCHAR2(30)	'escapedby'
OPT_NULLDEFINEDAS	VARCHAR2(30)	'nulldefinedas'
Not Yet Implemented Option Columns
OPT_QUOTE	VARCHAR2(30)	'quote'
OPT_REJECTLMT	VARCHAR2(30)	'rejectlmt'
OPT_REMOVEQUOTES	VARCHAR2(30)	'removequotes'
OPT_TSFMT	VARCHAR2(30)	'tsfmt'
OPT_TSLTZFMT	VARCHAR2(30)	'tsltzfmt'
OPT_TSTZFMT	VARCHAR2(30)	'tstzfmt'
OPT_TRIMSPACES	VARCHAR2(30)	'trimspaces'
OPT_TRUNCATECOL	VARCHAR2(30)	'truncatecol'
NYI Option Key Columns
OPT_KEYAVROSCHEMA	VARCHAR2(30)	'keyavroschema'
OPT_KEYREFTABLE	VARCHAR2(30)	'keyreftable'
Topic Formats
TOPIC_FORMAT_AVRO	VARCHAR2(30)	'AVRO'
TOPIC_FORMAT_DSV	VARCHAR2(30)	'DSV'
TOPIC_FORMAT_JSON	VARCHAR2(30)	'JSON'
Watermark
WATER_MARK_HIGH	VARCHAR2(30)	'HIGH'
WATER_MARK_LOW	VARCHAR2(30)	'LOW'

Dependencies

DBMS_ASSERT	DBMS_STANDARD	USER_KAFKA_OPS
DBMS_KAFKA_MD	DBMS_SYS_ERROR	USER_KAFKA_OPS_RESULTS
DBMS_KAFKA_UTL	DUAL	USER_KAFKA_PARTITIONS
DBMS_LOB	JSON_OBJECT_T	USER_VIEWS
DBMS_OUTPUT	USER_KAFKA_APPLICATIONS

Documented No

First Available 23ai

Pragmas SUPPLEMENTAL_LOG_DATA(default, NONE);

Security Model Owned by SYS with EXECUTE granted to PUBLIC

Source {ORACLE_HOME}/rdbms/admin/dbmskafka.sql
{ORACLE_HOME}/rdbms/admin/prvtkafka.plb

Subprograms

ADD_PARTITIONS	DROP_ALL_APPS	INIT_OFFSET
CONVERT_MS_TO_TIMESTAMP	DROP_LOAD_APP	INIT_OFFSET_TS
CONVERT_MS_TO_TIMESTAMP_TZ	DROP_SEEKABLE_APP	LOAD_TEMP_TABLE
CONVERT_TIMESTAMP_TO_MS	DROP_STREAMING_APP	SEEK_OFFSET_TS
CREATE_LOAD_APP	ENABLE_VIEW_QUERY	SET_TRACING
CREATE_SEEKABLE_APP	EXECUTE_LOAD_APP	UPDATE_OFFSET
CREATE_STREAMING_APP	GET_GRANULE_COUNT

ADD_PARTITIONS (new 23ai)

Adds additional Kafka partitions to an existing set of OSAK views dbms_kafka.add_partitions( cluster_name IN VARCHAR2, application_name IN VARCHAR2);

TBD

CONVERT_MS_TO_TIMESTAMP (new 23ai)

Convert the input number of milliseconds since the epoch time to a TIMESTAMP WITH TIME ZONE and return this value dbms_kafka.convert_ms_to_timestamp( milliseconds IN INTEGER, timezone IN VARCHAR2 DEFAULT NULL) RETURN TIMESTAMP WITH TIME ZONE;

TBD

CONVERT_MS_TO_TIMESTAMP_TZ (new 23ai)

TBD

CONVERT_TIMESTAMP_TO_MS (new 23ai)

Calculate and return the number of milliseconds since epoch time from the input TIMESTAMP WITH TIME ZONE

Overload 1 dbms_kafka.timestamp_to_ms(datetime_tz IN TIMESTAMP WITH TIME ZONE) RETURN INTEGER;

TBD

Overload 2 dbms_kafka.timestamp_to_ms( datetime IN TIMESTAMP, timezone IN VARCHAR2 DEFAULT NULL) RETURN INTEGER;

TBD

CREATE_LOAD_APP (new 23ai)

Creates an Oracle SQL Access to Kafka LOAD application that will retrieve data from all partitions in a Kafka topic for the purpose of loading Kafka data into an Oracle Database table dbms_kafka.create_load_app( cluster_name IN VARCHAR2, application_name IN VARCHAR2, topic_name IN VARCHAR2, options IN CLOB);

TBD

CREATE_SEEKABLE_APP (new 23ai)

Creates one OSAK view and an associated global temporary table to seek and load Kafka records between a user specified window of time dbms_kafka.create_seekable_app( cluster_name IN VARCHAR2, application_name IN VARCHAR2, topic_name IN VARCHAR2, options IN CLOB);

TBD

CREATE_STREAMING_APP (new 23ai)

Creates an OSAK streaming application which includes a set of dedicated OSAK global temporary tables and OSAK views used for retrieving new, unread records from partitions in a Kafka topic. dbms_kafka.create_streaming_app( cluster_name IN VARCHAR2, application_name IN VARCHAR2, topic_name IN VARCHAR2, options IN CLOB, view_count IN INTEGER DEFAULT 1);

TBD

DROP_ALL_APPS (new 23ai)

Drop all applications for the Kafka cluster dbms_kafka.drop_all_apps(cluster_name IN VARCHAR2);

TBD

DROP_LOAD_APP (new 23ai)

Drop LOAD application and remove the related metadata dbms_kafka.drop_load_app( cluster_name IN VARCHAR2, application_name IN VARCHAR2);

TBD

DROP_SEEKABLE_APP (new 23ai)

Dropa seekable application and remove the related metadata dbms_kafka.drop_seekable_app( cluster_name IN VARCHAR2, application_name IN VARCHAR2);

TBD

DROP_STREAMING_APP (new 23ai)

Dropa streaming application and remove the related metadata dbms_kafka.drop_streaming_app( cluster_name IN VARCHAR2, application_name IN VARCHAR2);

TBD

ENABLE_VIEW_QUERY (new 23ai)

Sets session context to allow an OSAK application to query an OSAK view directly dbms_kafka.enableview_query(view_name IN VARCHAR2);

TBD

EXECUTE_LOAD_APP (new 23ai)

Loads a user defined target table with content from a Kafka topic.

For subsequent calls, only new unread Kafka records are inserted into the target table. dbms_kafka.execute_load_app( cluster_name IN VARCHAR2, application_name IN VARCHAR2, target_table IN VARCHAR2, records_loaded OUT INTEGER, parallel_hint IN INTEGER DEFAULT 0);

TBD

GET_GRANULE_COUNT (new 23ai)

Returnss the number of granules associated with an OSAK view dbms_kafka.get_granule_count(view_name IN VARCHAR2) RETURN INTEGER;

TBD

INIT_OFFSET (new 23ai)

Specifies a starting offset by specifying a delta number of records from either the high or low watermark of every partition dbms_kafka.init_offset( view_name IN VARCHAR2, record_count IN INTEGER, water_mark IN VARCHAR2 DEFAULT WATER_MARK_HIGH);

TBD

INIT_OFFSET_TS (new 23ai)

Positions the processing of Kafka records to a point that is relatively current, potentially skipping unprocessed older records in the Kafka partitions

Overload 1 dbms_kafka.init_offset_ts( view_name IN VARCHAR2, start_timestamp_ms IN INTEGER);

TBD

Overload 2 dbms_kafka.init_offset_ts( view_name IN VARCHAR2, start_timestamp IN TIMESTAMP WITH TIME ZONE);

TBD

Overload 3 dbms_kafka.init_offset_ts( view_name IN VARCHAR2, start_timestamp IN TIMESTAMP, timezone IN VARCHAR2 DEFAULT NULL);

TBD

LOAD_TEMP_TABLE (new 23ai)

Loads a global temporary table from a SELECT * FROM <view_name>, where view name is an OSAK view dbms_kafka.load_temp_table( temp_table_name IN VARCHAR2, parallel_hint IN INTEGER DEFAULT 0);

TBD

SEEK_OFFSET_TS (new 23ai)

Positions an OSAK view to start reading Kafka records between two user specified epoch timestamps

Overload 1 dbms_kafka.seek_offset_ts( view_name IN VARCHAR2, start_timestamp_ms IN INTEGER, end_timestamp_ms IN INTEGER);

TBD

Overload 2 dbms_kafka.seek_offset_ts( view_name IN VARCHAR2, start_timestamp IN TIMESTAMP WITH TIME ZONE, end_timestamp IN TIMESTAMP WITH TIME ZONE);

TBD

Overload 3 dbms_kafka.seek_offset_ts( view_name IN VARCHAR2, start_timestamp IN TIMESTAMP, end_timestamp IN TIMESTAMP, timezone IN VARCHAR2 DEFAULT NULL);

TBD

SET_TRACING (new 23ai)

Enables or disables debug level tracing for the external table driver code dbms_kafka.set_tracing( cluster_name IN VARCHAR2, application_name IN VARCHAR2, enable IN BOOLEAN);

TBD

UPDATE_OFFSET (new 23ai)

Updates the last Kafka partition offsets read after a STREAMING application instance has has finished querying a set of Kafka records captured in an OSAK temporary table dbms_kafka.update_offset(view_name IN VARCHAR2);

TBD