v1.3.0 - Single instance physical standby for Oracle RAC

When running against single instance physical standby for Oracle RAC connector automatically detects opened redo threads and starts required number of connector tasks (max.tasks parameter must be equal to or greater than the number of redo threads)

OCT-22 fixes and enhancements

OCT-22 fixes and enhancements

LOB_TRIM/LOB_ERASE output to log & Jackson Databind version change (fix for CVE-2022-42004)

v1.2.1 - replace log4j with reload4j

replace log4j with reload4j (CVE-2022-23305, CVE-2019-17571, CVE-2022-23302, CVE-2022-23307, CVE-2020-9488)

v1.2.0 - support for Oracle RAC

For more information please see What about Oracle RAC?

Connect to Oracle Database: standardization and simplification

Deprecation of parameters a2.tns.admin, a2.tns.alias, a2.standby.tns.admin, a2.standby.tns.alias, a2.distributed.tns.admin, and a2.distributed.tns.alias. Please use a2.jdbc.url, a2.standby.jdbc.url, and a2.distributed.jdbc.url respectively. Please see KAFKA-CONNECT.md for parameter description and Oracle® Database JDBC Java API Reference, Release 21c for more information about JDBC URL format.

v1.0.0

Min Java version -> Java11, Java 17 LTS - recommended
Package name change: eu.solutions.a2 -> solutions.a2
a2.resiliency.type = fault-tolerant - now default

a2solutions-oracdc-kafka-1.0.0.zip - Confluent CC package
oracdc-kafka-1.0.0-standalone.jar - fat jar (with all dependencies) for any distribution

0.9.9.2: a2.resiliency.type = fault-tolerant

a2.resiliency.type = fault-tolerant to ensure 100% compatibility with Kafka Connect distributed mode

0.9.9.1: ORA-1291 fixes and new non-static connection pool for LogMiner connector

v0.9.9.1

fix parameter check

v0.9.9: DDL & Schema evolution for LogMiner source

The following Data Definition Language (DDL) clauses of ALTER TABLE command are currently supported:

1. add column(s)
2. modify column(s)
3. drop column(s)
4. rename column
5. set unused column(s)

To ensure compatibility with Schema Evolution the following algorithm is used:

1. When oracdc first encounters an operation on a table in the redo logs, information about the table is read from the data dictionary or from the JSON file specified by the a2.dictionary.file parameter. Two schemas are created: immutable key schema with unique identifier [PDB_NAME:]OWNER.TABLE_NAME.Key, version=1 and mutable value schema with unique identifier [PDB_NAME:]OWNER.TABLE_NAME.Value, version=1.
2. After successful parsing of DDL and comparison of columns definition before and after, value schema gets an incremented version number.

0.9.8.4 : support connection to CDB$ROOT or to an individual PDB

In CDB Architecture oracdc must connected to CDB$ROOT, but starting from Oracle Database 19c RU 10 and Oracle Database 21c you can chose to connect either to the CDB$ROOT, or to an individual PDB.