OpenLogReplicator

Open Source Oracle database CDC written purely in C++. Reads transactions directly from database redo log files and streams in JSON or Protobuf format to:

• Kafka
• RocketMQ
• flat file
• network stream (plain TCP/IP or ZeroMQ)

Please mind that the code has 2 branches:

1. master - branch with stable code - updated monthly
2. nightly - unstable current branch with daily code updates

Updating Protobuf code:

1. cd proto
2. export PATH=/opt/protobuf/bin:$PATH
3. protoc OraProtoBuf.proto --cpp_out=.
4. mv OraProtoBuf.pb.cc ../src/OraProtoBuf.pb.cpp
5. mv OraProtoBuf.pb.h ../src/OraProtoBuf.pb.h

Compilation for Debug:

1. git clone https://github.com/bersler/OpenLogReplicator
2. cd OpenLogReplicator
3. autoreconf -f -i
4. ./configure CFLAGS='-fsanitize=address' CXXFLAGS='-g -O0 -fsanitize=address' LDFLAGS='-fsanitize=address' --with-rapidjson=/opt/rapidjson --with-rdkafka=/opt/librdkafka --with-instantclient=/opt/instantclient_19_14 --with-protobuf=/opt/protobuf --with-zeromq=/usr --with-rocketmq=/opt/rocketmq-client-cpp-2.2.0
5. make

Compilation for Release:

1. git clone https://github.com/bersler/OpenLogReplicator
2. cd OpenLogReplicator
3. autoreconf -f -i
4. ./configure CXXFLAGS='-O3' --with-rapidjson=/opt/rapidjson --with-rdkafka=/opt/librdkafka --with-instantclient=/opt/instantclient_19_14 --with-protobuf=/opt/protobuf --with-zeromq=/usr --with-rocketmq=/opt/rocketmq-client-cpp-2.2.0
5. make

Step 3 is optional and required if you downloaded the files from GIT and timestamps of files may be changed.

Running:

1. cp config/OpenLogReplicator.example.json config/OpenLogReplicator.json
2. vi config/OpenLogReplicator.json
3. export LD_LIBRARY_PATH=/opt/instantclient_19_14:/opt/protobuf/lib:/opt/librdkafka/lib:/opt/rocketmq-client-cpp-2.2.0/bin
4. ./src/OpenLogReplicator

The documentation for the OpenLogReplicator program can be found on https://www.bersler.com/openlogreplicator/

Please do not create issues claiming that the documentation is missing. It is still being created. This is not speed up the process, but will it will down it down. Instead of writing documenation and finishing the code the time is spent on answering questions.

The currently available chapters are:

1. Getting started: https://www.bersler.com/openlogreplicator/getting-started/
2. Installation: https://www.bersler.com/openlogreplicator/installation/
3. Configuration: https://www.bersler.com/openlogreplicator/configuration/
4. Tutorials: https://www.bersler.com/openlogreplicator/tutorials/
5. FAQ: https://www.bersler.com/openlogreplicator/faq/
6. Support: https://www.bersler.com/openlogreplicator/support/
7. Contribution: https://www.bersler.com/openlogreplicator/contribution/

I have also opened a gitter chat at https://gitter.im/bersler/OpenLogReplicator