How much does it cost?
OpenLogReplicator is released under GPL v3. license. Running the program both in development and production environment do not require paying any fee to the author. See license for details.
Is it legal?
OpenLogReplicator is basically a translator which reads redo log files and produces JSON as output. It does not perform any other activity.
For EU users – please refer to Article 6 of EU directive 91/250/EEC for more details about reading proprietary Oracle format for interoperability cases.
Please consult your legal department if using third party products or reading your own database’s redo log files (which you have created) does not violate any agreement you have signed with the vendor.
I would like xxx feature to be implemented. Why is it not done yet?
This is a very good question. Some of the features are very important and are planned to be implemented in the future, some not. It really depends on the popularity of the program and the feedback the author receives from users.
Will older versions than 11.2 be supported?
Can I use it in production environment?
There are actually first companies who have actually started using it in production environment. The code of the is currently being reviewed and soon an official beta version is to be released. Before using it in production – please test it in your test environment and make sure that the impact on the production database is acceptable for you. Please verify the output with your test scenarios to make sure that the JSON output you have is correct. Please track github open issues list and make sure your code is up to date.
How can I be sure that all transactions are captured?
The only way to ensure is to use a vendor-supported replication tool – Oracle GoldenGate. Oracle redo log format is a proprietary format. There is no publicly available documentation which describes it. All third party replication tools are based on their own interpretation of redo log files. Still many customers worldwide use in production software written by third parties.
First – prepare a good testing environment, test the software with all kind of possible scenarios. Secondly – run a second tool to verify the results. If there is a case when you find that the results are inappropriate – please report it to author – so that the code can be fixed.
OpenLogReplicator is written in a way that if it encounters a redo log record which it can’t decode (but it is expecting this redo log record should contain some valid information) – then redo log processing would stop immediately. There are many safe checks in the code that verify the integrity of the redo log and every anomaly is reported. The user can also force the program to skip invalid log records and continue to work (but this is not done by default).
Stay tuned and keep track on open issues – to be up to date with errors reported by other users. If any issue appears in your environment – please report it to the author.
Can I use OpenLogReplicator together with xxx?
OpenLogReplicator has no impact on the database and is not dependent on any other tool. As long as you keep your archived redo logs available it can operate without any problem. In offline mode it can even operate when the database is shut down.
Why is the documentation incomplete?
The documentation is currently in the process of being written. Now we also have vacation and everybody wants also some time to rest.
Where can I find information about redo log format?
This format is proprietary and you can’t. You can however read some analysis of other people. Here is a brief summary what is available.
What about Redis?
As you can remember, the very early code version of OpenLogReplicator had support for Redis. It was removed because it was way to complicated to keep it up to date and include all changes both in Kafka and Redis support. Currently there are no plans to re-implement other target than Kafka.
I have another question which is not listed here!
Please use the contact form and send it to the author.