[HBASE-10295] Refactor the replication implementation to eliminate permanent zk node

• Type: Bug
• Status: Closed
• Priority: Critical
• Resolution: Duplicate
• Component/s: Replication
• Assignee: Unassigned
• Reporter: Honghua Feng
• Created: 08/Jan/14 01:53
• Updated: 15/Jun/22 17:31
• Resolved: 12/Jun/22 13:01

Description

Though this is a broader and bigger change, it original motivation derives from HBASE-8751: the newly introduced per-peer tableCFs attribute should be treated the same way as the peer-state, which is a permanent sub-node under peer node but using permanent zk node is deemed as an incorrect practice. So let's refactor to eliminate the permanent zk node. And the HBASE-8751 can then align its newly introduced per-peer tableCFs attribute with this correct implementation theme.

Issue Links

• relates to: HBASE-11629 Operational concerns for Replication should call out ZooKeeper
• relates to: HBASE-15867 Move HBase replication tracking from ZooKeeper to HBase
• relates to: HBASE-9864 Notifications bus for use by cluster members keeping up-to-date on changes
• relates to: HBASE-10296 Replace ZK with a consensus lib(paxos,zab or raft) running within master processes to provide better master failover performance and state consistency