Hadoop中datanode无法启动的解决办法之一

查看datanode的log(在安装目录中)，发现报错：

2015-08-06 13:36:55,619 WARN

java.io.IOException:

org.apache.hadoop.hdfs.server.common.Storage: Incompatible

clusterIDs

in

/usr/local/hadoop-2.7.0/hadoop_tmp/dfs/data:

namenode clusterID = CID-000f8956-eb97-4d68-ad76-af8603681717; datanode clusterID = CID-f1b86113-b4bb-44d5-8851-311881c3f1f5

2.7.0:::::::

root@tl:/usr/local/hadoop-2.7.0/hadoop_tmp/dfs/name/current# ls

edits_inprogress_0000000000000000001 fsimage_0000000000000000000.md5 VERSION

fsimage_0000000000000000000 seen_txid

namenode address dfs.namenode.servicerpc-address or

dfs.namenode.rpc-address is not configured.

name:::VERSION内容如下所示：

#Thu Aug 06 13:43:57 CST 2015

namespaceID=1577383239

clusterID=CID-b614c1a0-4a88-4593-842f-a0b2eac91447

cTime=0

storageType=NAME_NODE

blockpoolID=BP-1608731427-192.168.0.107-1438839837706

layoutVersion=-63

data::::VERSION内容如下所示：

#Thu Aug 06 13:42:02 CST 2015

storageID=DS-ac346161-1af8-4542-9e1d-cd2fb5ffbc93

clusterID=CID-000f8956-eb97-4d68-ad76-af8603681717

cTime=0

datanodeUuid=e51520f3-66e9-4007-982f-18698b1fe248

storageType=DATA_NODE

layoutVersion=-56

2.7.1::::::::::::

root@tl:/usr/local/hadoop-2.7.1/hadoop_tmp/dfs/name/current#

name:::::::::::VERSION内容如下所示：

#Thu Aug 06 13:31:31 CST 2015

namespaceID=315817124

clusterID=CID-6694dba4-7640-4c97-a9e2-df73f8c04ae0

cTime=0

storageType=NAME_NODE

blockpoolID=BP-112055986-192.168.0.107-1438329861534

layoutVersion=-63

data:::::::::::::VERSION内容如下所示：

#Thu Aug 06 13:46:13 CST 2015

storageID=DS-b13913f6-d502-4adb-bc9d-e369347323ba

clusterID=CID-6694dba4-7640-4c97-a9e2-df73f8c04ae0

cTime=0

datanodeUuid=c3ea8e52-89e3-4f68-875e-9929870ec4ee

storageType=DATA_NODE

layoutVersion=-56

讲2.7.0系列中VERSION中的clusterID修改为一致的,问题就解决了！

root@tl:/# jps

Picked up _JAVA_OPTIONS: -Dawt.useSystemAAFontSettings=gasp

15371 SecondaryNameNode

15946 Jps

15644 NodeManager

15053 NameNode

15195 DataNode

15532 ResourceManager

root@tl:/#

当然，还有其他可能导致Datanode不能启动的原因，再次仅把自己遇到的情况写出来！