[HBASE-2692] Master rewrite and cleanup for 0.90

• Type: Improvement
• Status: Closed
• Priority: Blocker
• Resolution: Fixed
• Fix Version/s: 0.90.0
• Hadoop Flags: Reviewed
• Assignee: Jonathan Gray
• Reporter: Jonathan Gray
• Created: 08/Jun/10 17:58
• Updated: 20/Nov/15 12:44
• Resolved: 01/Sep/10 05:14

Description

This is the parent issue for master changes targeted at 0.90 release.

Changes done as part of this issue grew out of work done over in HBASE-2485 to move region transitions into ZK.

In addition to that work, this issue will include general HMaster and ZooKeeper refactorings and cleanups.

Issue Links

• breaks: HBASE-14079 improve error message when Master fails to connect to Hadoop-auth
• incorporates: HBASE-1851 Broken master failover
• incorporates: HBASE-2485 Persist Master in-memory state so on restart or failover, new instance can pick up where the old left off
• is depended upon by: HBASE-2896 Retain assignment information between cluster shutdown/startup
• is related to: HBASE-2701 Move user-triggered region admin methods (compact/split/flush) to go directly to RS

Sub-Tasks

• Move RS to Master region open/close messaging into ZooKeeper: Closed
• HMaster cleanup and refactor: Closed
• ZooKeeper cleanup and refactor: Closed
• Implement new open/close logic in handlers and stop using heartbeats for open/close messages: Closed
• Reimplement cluster startup to use new load balancer: Closed
• Reimplement load balancing to be a background process and to not use heartbeats: Closed
• Handle master failover for regions in transition: Closed
• Move user-triggered region admin methods (compact/split/flush) to go directly to RS: Closed