This post has NOT been accepted by the mailing list yet.
My spark streaming has two input stream ,one is custom hdfs inputstream, the other is kafka input stream.
For the hdfs input stream, i read the hfds file ,and broadcast the List<String> in the foreachRDD.
And in the kafka input stream, for the map join. I use the broadcast to filter. but Now the broadcast is null
How to confim the broadcast is finished when read the broadcast value ?