Dear Tariq,
My current setup is as below:
Node-1 Node-2
/ 40GB/ 40GB
/opt120GB /opt120GB
/u0195GB/u0195GB
/dev/shm19GB/dev/shm19GB
Common Storage
/DATA1 100GB
/DATA2 100GB
Hi Yuvraj,
Once the disk is mounted we just need to give the names of all
the directories as a comma separated value for 'dfs.data.dir'
property, and the data will be sent to all these places. And we don't
have to worry as far as writing the data to a new storage is
concerned. As soon as add
Ok that's fine. Now can I create files directly on new partition ? - My
requirement is, If I write file from one node, It should be visible to other
node which running under cluster at the same time.
Also If I add new partition, shall I need to restart DFS and mapred ?
Regards,
Yuvrajsinh Chauh
Hi,
i have lately been running into problems since i started running hadoop
on a cluster:
The setup is the following:
1 Computer is NameNode and Jobtracker
1 Computer is SecondaryNameNode
2 Computers are TaskTracker and DataNode
I ran into problems with running the wordcount example: NameNode
Hello,
As far as I understand Bulk Import functionality will not take into account
the Data Locality question. MR job will create number of reducer tasks same
as regions to write into, but it will not "advice" on which nodes to run
these tasks. In that case Reducer task which writes HFiles of some
Can you share information on the java version that you are using.
- Is it as obvious as some previous processes still running and new
processes cannot bind to the port?
- Another pointer -
http://stackoverflow.com/questions/8360913/weird-java-net-socketexception-permission-denied-connect-error-when
Cross posting this to the hdfs-user group.
See below for the context, but basically I'm wondering if it is safe to lower
the dfs.socket.timeout to something like 5-10 seconds in my hbase-site.xml.
I'm thinking this would only affect the hdfs client calls that come from HBase,
so it wouldn't af
Thank you a lot for the replies.
To me it is clear when data locality gets broken though (and it is not only
the failure of the RS, there are other cases). I was hoping more for
suggestions around this particular use-case: assuming that nodes/RSs are
stable, how to make sure to achieve the data lo
