Member since
04-28-2016
7
Posts
3
Kudos Received
0
Solutions
12-26-2016
07:54 PM
2 Kudos
@Mon key In HDFS, reads normally go through the DataNode. Thus, when the client asks the DataNode to read a file, the DataNode reads that file off of the disk and sends the data to the client over a TCP socket. So-called “short-circuit” reads bypass the DataNode, allowing the client to read the file directly. Obviously, this is only possible in cases where the client is co-located with the data. Short-circuit reads provide a substantial performance boost to many applications. To configure short-circuit local reads, you must enable libhadoop.so . See Native Libraries for details on enabling this library. Windows is not a supported OS. You need to turn off this feature and re-execute your job.
... View more
05-30-2016
05:15 AM
@Mon key The best way is to find which are corrupted blocks using below command - hdfs fsck /path/to/corrupt/file -locations -blocks -files And then try to manually remove this using "hadoop rm -r </path>" to avoid dataloss. But still fsck does not remove good copied of data blocks.
... View more
02-18-2017
07:44 AM
hi, In my case, the permission on the following folder caused the issue. /hadoop/mapreduce/jhs/mr-jhs-state The following log message at /var/log/hadoop-mapreduce/mapred/mapred-mapred-historyserver-iml6.log helped me solve it. For some reason the files were owned by Yarn, when I given permission to the group I could see the files were rewritten by mapred. IO error: /hadoop/mapreduce/jhs/mr-jhs-state/LOCK: Permission denied ~BASU
... View more
04-28-2016
09:44 AM
@Brandon Wilson thanks
... View more