@Suresh Bonam Not out of box. You can build custom. CSV is still an option. If your source is streaming data in real-time then Flume is a reasonable option. An alternative is Apache NiFi. Assuming the streaming in real-time and willingness for Flume, the target files to be stored to HDFS will have a similar structure (no transformation in flight). Apache NiFi could help you to perform some transformation in-flight as such the file at the target is easier to consume, e.g. Hive external tables. You could achieve something like that with Flume but with coding and pain involved. If your Excel is static then you should use something else like a MapReduce or Spark job. ... View more

I completed this task by dowloading hwi.*.war file from 0.12 version of hive as i didn't find it in 0.13 and 0.14 ... View more

Check your connection details and it should be like below: Host <Give IP of HiveServer2>, Port: 10000, Database: optional( you can specify your target as well, if you don't specify it will show default db). Hive Server Type: Hive Server 2, Mechanism: User Name and Password, User Name : give your name and ensure that user has access in that edge node. Paswd: XXXXX ... View more

You can do something like this: hdfs dfs -put *filename*[0-9].txt /tmp For example: $ touch ${RANDOM}filename-$(date +"%F").txt ${RANDOM}filename.txt $ ls *filename*.txt 17558filename-2016-01-05.txt 27880filename.txt $ hdfs dfs -put *filename*[0-9].txt /tmp $ hdfs dfs -ls /tmp -rw-r--r-- 3 hdfs hdfs 0 2016-01-05 16:39 /tmp/17558filename-2016-01-05.txt If that doesn't work, add this to the beginning: set +f ... View more