Hadoop: Datanode process killed - hadoop

Question

I am currently using Hadoop-2.0.3-alpha and after I could work perfectly with HDFS (copying files into HDFS, getting success from an external framework, using the webfrontend), after a new start of my VM, the datanode process is stopping after a while. The namenode process and all yarn processes work without a problem. I installed Hadoop in a folder under an additional user, as I also still have installed Hadoop 0.2, which worked fine too.
Taking a look at the log-file of all datanode processes I got the following information:
2013-04-11 16:23:50,475 WARN org.apache.hadoop.util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
2013-04-11 16:24:17,451 INFO org.apache.hadoop.metrics2.impl.MetricsConfig: loaded properties from hadoop-metrics2.properties
2013-04-11 16:24:23,276 INFO org.apache.hadoop.metrics2.impl.MetricsSystemImpl: Scheduled snapshot period at 10 second(s).
2013-04-11 16:24:23,279 INFO org.apache.hadoop.metrics2.impl.MetricsSystemImpl: DataNode metrics system started
2013-04-11 16:24:23,480 INFO org.apache.hadoop.hdfs.server.datanode.DataNode: Configured hostname is user-VirtualBox
2013-04-11 16:24:28,896 INFO org.apache.hadoop.hdfs.server.datanode.DataNode: Opened streaming server at /0.0.0.0:50010
2013-04-11 16:24:29,239 INFO org.apache.hadoop.hdfs.server.datanode.DataNode: Balancing bandwith is 1048576 bytes/s
2013-04-11 16:24:38,348 INFO org.mortbay.log: Logging to org.slf4j.impl.Log4jLoggerAdapter(org.mortbay.log) via org.mortbay.log.Slf4jLog
2013-04-11 16:24:44,627 INFO org.apache.hadoop.http.HttpServer: Added global filter 'safety' (class=org.apache.hadoop.http.HttpServer$QuotingIn putFilter)
2013-04-11 16:24:45,163 INFO org.apache.hadoop.http.HttpServer: Added filter static_user_filter (class=org.apache.hadoop.http.lib.StaticUserWebFil ter$StaticUserFilter) to context datanode
2013-04-11 16:24:45,164 INFO org.apache.hadoop.http.HttpServer: Added filter static_user_filter (class=org.apache.hadoop.http.lib.StaticUserWebFil ter$StaticUserFilter) to context logs
2013-04-11 16:24:45,164 INFO org.apache.hadoop.http.HttpServer: Added filter static_user_filter (class=org.apache.hadoop.http.lib.StaticUserWebFil ter$StaticUserFilter) to context static
2013-04-11 16:24:45,355 INFO org.apache.hadoop.hdfs.server.datanode.DataNode: Opened info server at 0.0.0.0:50075
2013-04-11 16:24:45,508 INFO org.apache.hadoop.hdfs.server.datanode.DataNode: dfs.webhdfs.enabled = false
2013-04-11 16:24:45,536 INFO org.apache.hadoop.http.HttpServer: Jetty bound to port 50075
2013-04-11 16:24:45,576 INFO org.mortbay.log: jetty-6.1.26
2013-04-11 16:25:18,416 INFO org.mortbay.log: Started SelectChannelConnector#0.0.0.0:50075
2013-04-11 16:25:42,670 INFO org.apache.hadoop.ipc.Server: Starting Socket Reader #1 for port 50020
2013-04-11 16:25:44,955 INFO org.apache.hadoop.hdfs.server.datanode.DataNode: Opened IPC server at /0.0.0.0:50020
2013-04-11 16:25:45,483 INFO org.apache.hadoop.hdfs.server.datanode.DataNode: Refresh request received for nameservices: null
2013-04-11 16:25:47,079 INFO org.apache.hadoop.hdfs.server.datanode.DataNode: Starting BPOfferServices for nameservices: <default>
2013-04-11 16:25:47,660 INFO org.apache.hadoop.hdfs.server.datanode.DataNode: Block pool <registering> (storage id unknown) service to localhost/127.0.0.1:8020 starting to offer service
2013-04-11 16:25:50,515 INFO org.apache.hadoop.ipc.Server: IPC Server Responder: starting
2013-04-11 16:25:50,631 INFO org.apache.hadoop.ipc.Server: IPC Server listener on 50020: starting
2013-04-11 16:26:15,068 INFO org.apache.hadoop.hdfs.server.common.Storage: Lock on /home/hadoop/workspace/hadoop_space/hadoop23/dfs/data/in_use.lock acquired by nodename 3099#user-VirtualBox
2013-04-11 16:26:15,720 FATAL org.apache.hadoop.hdfs.server.datanode.DataNode: Initialization failed for block pool Block pool BP-474150866-127.0.1.1-1365686732002 (storage id DS-317990214-127.0.1.1-50010-1365505141363) service to localhost/127.0.0.1:8020
java.io.IOException: Incompatible clusterIDs in /home/hadoop/workspace/hadoop_space/hadoop23/dfs/data: namenode clusterID = CID-1745a89c-fb08-40f0-a14d-d37d01f199c3; datanode clusterID = CID-bb3547b0-03e4-4588-ac25-f0299ff81e4f
at org.apache.hadoop.hdfs.server.datanode.DataStorage .doTransition(DataStorage.java:391)
at org.apache.hadoop.hdfs.server.datanode.DataStorage .recoverTransitionRead(DataStorage.java:191)
at org.apache.hadoop.hdfs.server.datanode.DataStorage .recoverTransitionRead(DataStorage.java:219)
at org.apache.hadoop.hdfs.server.datanode.DataNode.in itStorage(DataNode.java:850)
at org.apache.hadoop.hdfs.server.datanode.DataNode.in itBlockPool(DataNode.java:821)
at org.apache.hadoop.hdfs.server.datanode.BPOfferServ ice.verifyAndSetNamespaceInfo(BPOfferService.java: 280)
at org.apache.hadoop.hdfs.server.datanode.BPServiceAc tor.connectToNNAndHandshake(BPServiceActor.java:22 2)
at org.apache.hadoop.hdfs.server.datanode.BPServiceAc tor.run(BPServiceActor.java:664)
at java.lang.Thread.run(Thread.java:722)
2013-04-11 16:26:16,212 WARN org.apache.hadoop.hdfs.server.datanode.DataNode: Ending block pool service for: Block pool BP-474150866-127.0.1.1-1365686732002 (storage id DS-317990214-127.0.1.1-50010-1365505141363) service to localhost/127.0.0.1:8020
2013-04-11 16:26:16,276 INFO org.apache.hadoop.hdfs.server.datanode.DataNode: Removed Block pool BP-474150866-127.0.1.1-1365686732002 (storage id DS-317990214-127.0.1.1-50010-1365505141363)
2013-04-11 16:26:18,396 WARN org.apache.hadoop.hdfs.server.datanode.DataNode: Exiting Datanode
2013-04-11 16:26:18,940 INFO org.apache.hadoop.util.ExitUtil: Exiting with status 0
2013-04-11 16:26:19,668 INFO org.apache.hadoop.hdfs.server.datanode.DataNode: SHUTDOWN_MSG:
/************************************************** **********
SHUTDOWN_MSG: Shutting down DataNode at user-VirtualBox/127.0.1.1
************************************************** **********/
Any ideas? May be I made a mistake during the installation process? But it is strange, that it worked once. I also have to say, that if I am logged in as my additional user to execute the commands ./hadoop-daemon.sh start namenode and the same with the datanode, I need to add sudo.
I used this installation guide: http://jugnu-life.blogspot.ie/2012/0...rial-023x.html
By the way, I use the Oracle Java-7 version.

Answer 1

The problem could be that the namenode was formatted after the cluster was set up and the datanodes were not, so the slaves are still referring to the old namenode.
We have to delete and recreate the folder /home/hadoop/dfs/data on the local filesystem for the datanode.
Check your hdfs-site.xml file to see where dfs.data.dir is pointing to
and delete that folder
and then restart the datanode daemon on the machine
The steps above should recreate the folder and resolve the problem.
Please share your config info if the instructions above do not work.

Answer 2

DataNode dies because of incompatible Clusterids. To fix this problem
If you are using hadoop 2.X, then you have to delete everything in the folder that you have specified in hdfs-site.xml - "dfs.datanode.data.dir" (but NOT the folder itself).
The ClusterID will be maintained in that folder. Delete and restart dfs.sh. This should work!!!

Answer 3

You need to delete both
C:\hadoop\data\dfs\datanode and
C:\hadoop\data\dfs\namenode folders.
If you don't have this folders - open your C:\hadoop\etc\hadoop\hdfs-site.xml file and get paths for this folders for next deletion. For me it says:
<property>
<name>dfs.namenode.name.dir</name>
<value>file:/hadoop/data/dfs/namenode</value>
</property>
<property>
<name>dfs.datanode.data.dir</name>
<value>file:/hadoop/data/dfs/datanode</value>
</property>
Run command for Format namenodec:\hadoop\bin>hdfs namenode -format
Now it should work!

Answer 4

I think the recommended way of doing this without deleting the data directory is to simply change the clusterID variable in the datanode's VERSION file.
If you look in your daemons directory, you will see the datanode directory exmaple
data/hadoop/daemons/datanode
The VERSION file should look like this.
cat current/VERSION
#Tue Oct 14 17:31:58 CDT 2014
storageID=DS-23bf7f3a-085c-4531-808f-801ff6d52d14
clusterID=CID-bb3547b0-03e4-4588-ac25-f0299ff81e4f
cTime=0
datanodeUuid=63154929-ae68-4149-9f75-9a6558545041
storageType=DATA_NODE
layoutVersion=-55
You need to change the clusterId to the first value in the output of the message so in your case that would be CID-1745a89c-fb08-40f0-a14d-d37d01f199c3 instead of CID-bb3547b0-03e4-4588-ac25-f0299ff81e4f
The updated version should appear like this with the altered clusterId
cat current/VERSION
#Tue Oct 14 17:31:58 CDT 2014
storageID=DS-23bf7f3a-085c-4531-808f-801ff6d52d14
clusterID=CID-1745a89c-fb08-40f0-a14d-d37d01f199c3
cTime=0
datanodeUuid=63154929-ae68-4149-9f75-9a6558545041
storageType=DATA_NODE
layoutVersion=-55
Restart hadoop and the datanode should start just fine.