Implement a simple Master/Slave replication

[alexandrestein]

For a database to be secured it needs to replicate the data.
The first steep would be to implement:

• a very basic replication system where, it synchronise the all database at the first shot and then forward all writes to the replicated services
• in case of data corruption pull a valid record from a replicated service
• and run a periodic or a constant data integrity check to repair corruption before requesting the document

In the first place the system needs to be fairly simple to be ready for production.

It should be easy to use log shipping approach.
Just keep a bucket for events that describes any operation on the db.
Then sync them across as needed as chunks.
Pull or push ?
I would use push because the slave is likely to have a public static IP.

[alexandrestein]

Sorry for the late answer.
Yes it can be a good way of doing this.

I would prefer to have slaves maintaining a valid bidirectional connection to the master and than the master pushing the records to every active connexions.
A timestamp or something similar would probably also be needed.