【发布时间】:2015-03-29 03:26:51
【问题描述】:
我正在尝试使用 Spark-cassandra-Connector 将 RDD[CassandraRow] 写入现有的 Cassandra 表。这是我的一段代码
val conf = new SparkConf().setAppName(getClass.getSimpleName)
.setMaster("local[*]")
.set("spark.cassandra.connection.host", host)
val sc = new SparkContext("local[*]", keySpace, conf)
val rdd = sc.textFile("hdfs://hdfs-host:8020/Users.csv")
val columns = Array("ID", "FirstName", "LastName", "Email", "Country")
val types = Array("int", "string", "string", "string", "string")
val crdd=rdd.map(p => {
var tokens = p.split(",")
new CassandraRow(columns,tokens)
})
val targetedColumns = SomeColumns.seqToSomeColumns(columns)
crdd.saveToCassandra(keySpace, tableName, targetedColumns, WriteConf.fromSparkConf(conf))
当我运行这段代码时,我得到以下异常
Exception in thread "main" java.util.NoSuchElementException: Column not found ID in table demo.usertable
这是表的实际架构
CREATE TABLE usertable (
id int,
country text,
email text,
firstname text,
lastname text,
PRIMARY KEY ((id))
)
有什么建议吗? 谢谢
【问题讨论】:
标签: scala cassandra apache-spark