Avro Files Working

Step 1: download avro tool jar
wget http://mirrors.sonic.net/apache/avro/avro-1.7.7/java/avro-tools-1.7.7.jar

Step 2: Generate schema
java -jar avro-tools-1.7.7.jar getschema /home/hdfs/genre1/part-m-00000.avro

Step 3: 
sqoop import --connect jdbc:mysql://172.16.2.164/movielens --username hive -P --table genre --as-avrodatafile

this imports genre data from mysql to hdfs as .avro files and generates .avsc schema in local filesystem
data files need to be in hdfs but .avsc file can be either in local or hdfs

Step 4: creating a hive table for .avro data file

create external table genre row format serde 'org.apache.hadoop.hive.serde2.avro.AvroSerDe' stored as inputformat 'org.apache.hadoop.hive.ql.io.avro.AvroContainerInputFormat' outputformat 'org.apache.hadoop.hive.ql.io.avro.AvroContainerOutputFormat' location '/user/hdfs/genre' tblproperties('avro.schema.url'='hdfs:///user/hdfs/genre.avsc');

Step 5:
in the genre directory except _SUCCESS and part files nothing should be kept.