ActiveMQ基于 zookeeper + leveldb 配置 启动失败，一直刷 Warn 日志

[真丶傻]

环境：
jdk：1.7.0_79
activemq: 5.12.3
zookeeper: 3.4.10

zookeeper 集群启动没问题, 三个 activemq 启动后，两个被 lock ，只有一个一直在刷 WARN 日志：

INFO | Opening socket connection to server 10.253.1.88/10.253.1.88:2181
 WARN | unprocessed event state: AuthFailed
 INFO | Socket connection established to 10.253.1.88/10.253.1.88:2181, initiating session
 INFO | Session establishment complete on server 10.253.1.88/10.253.1.88:2181, sessionid = 0x15ea233860e0000, negotiated timeout = 4000
 INFO | Not enough cluster members connected to elect a master.
 INFO | Not enough cluster members connected to elect a master.
 INFO | Not enough cluster members have reported their update positions yet.
 INFO | Promoted to master
 WARN | listeners are taking too long to process the events
 WARN | listeners are taking too long to process the events
 INFO | Using the pure java LevelDB implementation.
 INFO | Master started: tcp://localhost:34707
 WARN | Store update waiting on 1 replica(s) to catch up to log position 0. 
 WARN | Store update waiting on 1 replica(s) to catch up to log position 0. 
 WARN | Store update waiting on 1 replica(s) to catch up to log position 0. 
 WARN | Store update waiting on 1 replica(s) to catch up to log position 0. 
 WARN | Store update waiting on 1 replica(s) to catch up to log position 0. 
 WARN | Store update waiting on 1 replica(s) to catch up to log position 0.

activemq 配置：

 
 <persistenceAdapter>
     <replicatedLevelDB
       directory="/home/tmp/activemq-data"
       replicas="3"
       bind="tcp://0.0.0.0:0"
       zkAddress="10.253.1.88:2181"
       zkPath="/home/tmp/activemq/leveldb-stores"
       hostname="localhost"
       />
  </persistenceAdapter>