Scala/Spark serialization error - streaming data to HBASE - scala

Question

I am a newbie to Scala/Spark. In the following code, I am extracting Twitter public stream content to the HBase.
On commenting the last four lines (put commands in HBase), I am able to print content of tweet on the terminal, however unable to dump it to the HBase table.
I need help in on the following regards:
1. How can I overcome the serialilzation error?
2. Are there efficient methods (may be useing Kryo serialilzation) to overcome this error?
Caused by: java.io.NotSerializableException:
org.apache.hadoop.conf.Configuration Serialization stack:
- object not serializable (class: org.apache.hadoop.conf.Configuration, value: Configuration:
core-default.xml, core-site.xml, mapred-default.xml, mapred-site.xml,
yarn-default.xml, yarn-site.xml, hdfs-default.xml, hdfs-site.xml)
import twitter4j.auth._
import twitter4j.conf._
import twitter4j._
import twitter4j.json._
import scala.io.Source
import org.apache.hadoop.conf.Configuration
import org.apache.hadoop.fs.Path;
import org.apache.hadoop.hbase.client.ConnectionFactory
import org.apache.hadoop.hbase.client.HBaseAdmin
import org.apache.hadoop.hbase.client.HTable;
import org.apache.hadoop.hbase.client.Put;
import org.apache.hadoop.hbase.HColumnDescriptor
import org.apache.hadoop.hbase.io.ImmutableBytesWritable
import org.apache.hadoop.hbase.KeyValue
import org.apache.hadoop.hbase.mapred.TableOutputFormat
import org.apache.hadoop.hbase.mapreduce.HFileOutputFormat2
import org.apache.hadoop.hbase.mapreduce.LoadIncrementalHFiles
import org.apache.hadoop.hbase.mapreduce.TableInputFormat
import org.apache.hadoop.hbase.TableName
import org.apache.hadoop.hbase.util.Bytes
import org.apache.hadoop.hbase.{HBaseConfiguration, HTableDescriptor}
import org.apache.hadoop.mapreduce.Job
import org.apache.spark._
import org.apache.spark.rdd.NewHadoopRDD
import org.apache.spark.SparkConf
import org.apache.spark.SparkContext
import org.apache.spark.streaming._
import org.apache.spark.streaming.StreamingContext._
import org.apache.spark.streaming.{Seconds, StreamingContext}
import org.apache.hadoop.hbase.util.Bytes
import java.io._
import org.apache.spark.streaming.twitter.TwitterUtils
////////////////////////////
val conf = new SparkConf().setAppName("model1").setMaster("local[*]")
// val sc = new SparkContext(conf)
val TABLE_NAME = "publicrd"
val CF_USER = "user"
val CF_TWEET = "tweet"
val CF_ENTITIES = "entities"
val CF_PLACES = "places"
val hadoopConf = new Configuration
val conf = HBaseConfiguration.create(hadoopConf)
val admin = new HBaseAdmin(conf)
val tableDesc = new HTableDescriptor(Bytes.toBytes(TABLE_NAME))
// Define column family descriptor
val ColumnFamilyDesc1 = new HColumnDescriptor(Bytes.toBytes(CF_USER))
val ColumnFamilyDesc2 = new HColumnDescriptor(Bytes.toBytes(CF_TWEET))
val ColumnFamilyDesc3 = new HColumnDescriptor(Bytes.toBytes(CF_ENTITIES))
val ColumnFamilyDesc4 = new HColumnDescriptor(Bytes.toBytes(CF_PLACES))
// Add column family in table descriptor
tableDesc.addFamily(ColumnFamilyDesc1)
tableDesc.addFamily(ColumnFamilyDesc2)
tableDesc.addFamily(ColumnFamilyDesc3)
tableDesc.addFamily(ColumnFamilyDesc4)
// Check if the table exists
if (admin.tableExists(TABLE_NAME)){
print(">>>>>" + TABLE_NAME + " already exists <<<<<")
admin.disableTable(TABLE_NAME)
admin.deleteTable(TABLE_NAME)
}
// Create HBASE table
admin.createTable(tableDesc)
val table = new HTable(conf, TABLE_NAME)
/////
val timewindow = 2 // seconds
val ssc = new StreamingContext(sc, Seconds(timewindow))
val cb = new ConfigurationBuilder
val ckey = "ckey"
val csecret = "csecret"
val atoken = "atoken"
val atokensecret = "atokensecret"
cb.setDebugEnabled(true).
setOAuthConsumerKey(ckey).
setOAuthConsumerSecret(csecret).
setOAuthAccessToken(atoken).
setOAuthAccessTokenSecret(atokensecret).
setJSONStoreEnabled(true)
val auth = new OAuthAuthorization(cb.build)
val tweets = TwitterUtils.createStream(ssc,Some(auth))
val status = tweets.filter(_.getLang()=="en")
status.foreachRDD(foreachFunc = rdd => {
rdd.foreachPartition {
records => while (records.hasNext) {
var record = records.next
print("\n\n>>>>"+record)
var tweetID = record.getUser().getId().toString//.isInstanceOf[Int]
print("\ntweetID : "+tweetID)
var tweetBody = record.getText()//.toString
print("\ntweetBody : "+tweetBody)
var favoritesCount = record.getFavoriteCount()//.toInt
print("\nfavoritesCount : "+favoritesCount)
var keyrow = "t_"+tweetID //"t_"+
print("\nkeyrow : "+keyrow+"\n")
var theput= new Put(Bytes.toBytes(keyrow))
theput.add(Bytes.toBytes(CF_TWEET),Bytes.toBytes("tweetid"),Bytes.toBytes(tweetID))
theput.add(Bytes.toBytes(CF_TWEET),Bytes.toBytes("tweetid"),Bytes.toBytes(tweetBody))
theput.add(Bytes.toBytes(CF_USER),Bytes.toBytes("tweetid"),Bytes.toBytes(favoritesCount))
table.put(theput)
}
}
}
)
The code is run on the terminal via:
spark-shell --driver-class-path /opt/hadoop/hbase-1.2.1/lib/hbase-server-1.1.4.jar:/opt/hadoop/hbase-1.2.1/lib/hbase-protocol-1.0.0-cdh5.5.0.jar:/opt/hadoop/hbase-1.2.1/lib/hbase-hadoop2-compat-1.0.0-cdh5.5.0.jar:/opt/hadoop/hbase-1.2.1/lib/hbase-client-1.0.0-cdh5.5.0.jar:/opt/hadoop/hbase-1.2.1/lib/hbase-common-1.0.0-cdh5.5.0.jar:/opt/hadoop/hbase-1.2.1/lib/htrace-core-3.2.0-incubating.jar:/home/cloudera/Desktop/hbase/twitter4jJARS/guava-19.0.jar:/home/cloudera/Desktop/hbase/twitter4jJARS/spark-streaming-twitter_2.10-1.6.1.jar:/home/cloudera/Desktop/hbase/twitter4jJARS/twitter4j-async-4.0.4.jar:/home/cloudera/Desktop/hbase/twitter4jJARS/twitter4j-core-4.0.4.jar:/home/cloudera/Desktop/hbase/twitter4jJARS/twitter4j-examples-4.0.4.jar:/home/cloudera/Desktop/hbase/twitter4jJARS/twitter4j-media-support-4.0.4.jar:/home/cloudera/Desktop/hbase/twitter4jJARS/twitter4j-stream-4.0.4.jar

Answer 1

It says the object org.apache.hadoop.conf.Configuration is not serialisable which mean it does not implement the Serializable interface while it's required. To get rid of that add #transient keyword.
#transient val hadoopConf = new Configuration