We are doing a major refactoring on replication at trunk now.. and it's almost ready.
When we refactored failover, we had to make a few adjustments to replication and it's almost ready now.. it should be release early next year.
Thanks a lot for the answer.
Is there anything we should be concerned about if we keep using 2.1 replication until then?
This is not so important, but is there any exaplantion about why the replication model was removed from 2.2+?
1 of 1 people found this helpful
There are a few bugs on the 2.1 that you probably want a fix on 2.2.5... Some bugs on compactation.
We have made 2.2+ to have an atomic failover, resilient to temporary failures between client and server. 2.1 would start the backup node right after the first client connects, what wasn't acceptable to some of our users (and a few supported customers)... so we made a major refactoring to make that happen.
we wouldn't be able to deliver 2.2 with replication on the new model as we would also need to refactor replication.
We could start replication back as soon as Francisco Borges joined the team and we are almost done with it now. We will be working on it for a release early next year as I said (probably Feb/March).
Thank you Clebert!