Amit Ramesh created SPARK-12002:
-----------------------------------
Summary: offsetRanges attribute missing in Kafka RDD when resuming from checkpoint
Key: SPARK-12002
URL: https://issues.apache.org/jira/browse/SPARK-12002
Project: Spark
Issue Type: Bug
Components: PySpark, Streaming
Reporter: Amit Ramesh
SPARK-8389 added offsetRanges to Kafka direct streams. And SPARK-10122 fixed the issue of not ending up with non-Kafka RDDs when chaining transforms to Kafka RDDs. It appears that this issue remains for the case where a streaming application using Kafka direct streams is initialized from the checkpoint directory. The following is a representative example where everything works as expected during the first run, but exceptions are thrown on a subsequent run when the context is being initialized from the checkpoint directory.
{code:title=test-checkpoint.py|language=python}
from pyspark import SparkContext
from pyspark.streaming import StreamingContext
from pyspark.streaming.kafka import KafkaUtils
def attach_kafka_metadata(kafka_rdd):
offset_ranges = kafka_rdd.offsetRanges()
return kafka_rdd
def create_context():
sc = SparkContext(appName='kafka-test')
ssc = StreamingContext(sc, 10)
ssc.checkpoint(CHECKPOINT_URI)
kafka_stream = KafkaUtils.createDirectStream(
ssc,
[TOPIC],
kafkaParams={
'metadata.broker.list': BROKERS,
},
)
kafka_stream.transform(attach_kafka_metadata).count().pprint()
return ssc
if __name__ == "__main__":
ssc = StreamingContext.getOrCreate(CHECKPOINT_URI, create_context)
ssc.start()
ssc.awaitTermination()
{code}
{code:title=Exception on resuming from checkpoint}
Traceback (most recent call last):
File "/home/spark/spark/python/lib/pyspark.zip/pyspark/streaming/util.py", line 62, in call
r = self.func(t, *rdds)
File "/home/spark/spark/python/lib/pyspark.zip/pyspark/streaming/kafka.py", line 344, in <lambda>
File "/home/spark/ad_realtime/batch/test_checkpoint.py", line 12, in attach_kafka_metadata
offset_ranges = kafka_rdd.offsetRanges()
AttributeError: 'RDD' object has no attribute 'offsetRanges'
{code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org