zookeeper无法启动"Unable to load database on disk

quorumpeermain,resourcemanager都没有起来
resourcemanager.log如下

2018-09-28 23:17:02,787 fatal org.apache.hadoop.yarn.server.resourcemanager.resourcemanager: error starting resourcemanager
org.apache.hadoop.service.servicestateexception: org.apache.zookeeper.keeperexception$connectionlossexception: keepererrorcode = connectionloss
        at org.apache.hadoop.service.servicestateexception.convert(servicestateexception.java:59)
        at org.apache.hadoop.service.abstractservice.init(abstractservice.java:172)
        at org.apache.hadoop.service.compositeservice.serviceinit(compositeservice.java:107)
        at org.apache.hadoop.yarn.server.resourcemanager.adminservice.serviceinit(adminservice.java:142)
        at org.apache.hadoop.service.abstractservice.init(abstractservice.java:163)
        at org.apache.hadoop.service.compositeservice.serviceinit(compositeservice.java:107)
        at org.apache.hadoop.yarn.server.resourcemanager.resourcemanager.serviceinit(resourcemanager.java:267)
        at org.apache.hadoop.service.abstractservice.init(abstractservice.java:163)
        at org.apache.hadoop.yarn.server.resourcemanager.resourcemanager.main(resourcemanager.java:1185)
caused by: org.apache.zookeeper.keeperexception$connectionlossexception: keepererrorcode = connectionloss
        at org.apache.zookeeper.keeperexception.create(keeperexception.java:99)
        at org.apache.hadoop.ha.activestandbyelector$watcherwithclientref.waitforzkconnectionevent(activestandbyelector.java:1061)
        at org.apache.hadoop.ha.activestandbyelector$watcherwithclientref.access$400(activestandbyelector.java:1032)
        at org.apache.hadoop.ha.activestandbyelector.getnewzookeeper(activestandbyelector.java:636)
        at org.apache.hadoop.ha.activestandbyelector.createconnection(activestandbyelector.java:775)
        at org.apache.hadoop.ha.activestandbyelector.<init>(activestandbyelector.java:229)
        at org.apache.hadoop.yarn.server.resourcemanager.embeddedelectorservice.serviceinit(embeddedelectorservice.java:92)
        at org.apache.hadoop.service.abstractservice.init(abstractservice.java:163)
        ... 7 more

发现连接不上zookeeer,从xcall.sh jps中得知s103,s104的quorumpeermain没启动，去s103的~/zookeeper/zookeeper.out查看日志，发现

2018-09-29 03:40:56,375 [myid:103] - error [main:quorumpeer@692] - unable to load database on disk
java.io.ioexception: found ^@^@^@ in /home/centos/zookeeper/version-2/acceptedepoch
        at org.apache.zookeeper.server.quorum.quorumpeer.readlongfromfile(quorumpeer.java:1362)
        at org.apache.zookeeper.server.quorum.quorumpeer.loaddatabase(quorumpeer.java:677)
        at org.apache.zookeeper.server.quorum.quorumpeer.start(quorumpeer.java:635)
        at org.apache.zookeeper.server.quorum.quorumpeermain.runfromconfig(quorumpeermain.java:170)
        at org.apache.zookeeper.server.quorum.quorumpeermain.initializeandrun(quorumpeermain.java:114)
        at org.apache.zookeeper.server.quorum.quorumpeermain.main(quorumpeermain.java:81)

datadir=/home/centos/zookeeper //快照日志,在conf/zoo.cfg下,事务日志和快照日志位于相同目录
zookeeper.log.dir=. //log4j日志：记录zokeeper的健康状态,在conf/log4j.properties下
参考

解决方法：zookeeper无法启动"unable to load database on disk"