Hi,
Scenario: computations on large number of data ( around 3 million
per file) on a cluster setup.Files being moved from local to hdfs.
What we encounter is that the computations run fine and file
movement is quick for the initial runs ( say 50) but after that the time
taken to copy the file from local took a long time.
Below is the snippet from the datanode logs (not sure if this is
related to our problem) :
* ERROR org.apache.hadoop.hdfs.server.datanode.DataNode:
DatanodeRegistration(10.58.116.222:50010,
storageID=DS-1630779127-10.58.116.222-50010-1324885312250, infoPort=50075,
ipcPort=50020):DataXceiver*
*java.net.SocketTimeoutException: 60000 millis timeout while waiting for
channel to be ready for connect. ch :
java.nio.channels.SocketChannel[connection-pending remote=/
10.58.116.223:50010]*
* at
org.apache.hadoop.net.SocketIOWithTimeout.connect(SocketIOWithTimeout.java:213)
*
* at org.apache.hadoop.net.NetUtils.connect(NetUtils.java:404)*
* at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.writeBlock(DataXceiver.java:282)
*
* at
org.apache.hadoop.hdfs.server.datanode.DataXceiver.run(DataXceiver.java:103)
*
* at java.lang.Thread.run(Thread.java:619)*
*
*
* *
Any idea or hints on the above error and why this could occur?
Also, it would be great if you could point out a few reasons why copy of
files after a certain time could be slow.
Thanks,
Sahana
*
*
*
*
*
*
* *
