5

I am using Structured Streaming + Kafka for realtime data analytics in our project. I am using Spark 2.2, kafka 0.10.2.

I am facing an issue during streaming query recovery from checkpoint at application startup. As there are multiple streaming queries derived from a single kafka streaming point and there are different checkpint directories for every streaming query. So in case of job failure, when we restart the job there are some streaming queries which fails to recover from checkpoint location hence throw an exception of Error reading delta file. Here are the logs :

Job aborted due to stage failure: Task 2 in stage 13.0 failed 4 times, most recent failure: Lost task 2.3 in stage 13.0 (TID 831, ip-172-31-10-246.us-west-2.compute.internal, executor 3): java.lang.IllegalStateException: Error reading delta file /checkpointing/wifiHealthPerUserPerMinute/state/0/2/1.delta of HDFSStateStoreProvider[id = (op=0, part=2), dir = /checkpointing/wifiHealthPerUserPerMinute/state/0/2]: /checkpointing/wifiHealthPerUserPerMinute/state/0/2/1.delta does not exist
at org.apache.spark.sql.execution.streaming.state.HDFSBackedStateStoreProvider.org$apache$spark$sql$execution$streaming$state$HDFSBackedStateStoreProvider$$updateFromDeltaFile(HDFSBackedStateStoreProvider.scala:410)
at org.apache.spark.sql.execution.streaming.state.HDFSBackedStateStoreProvider$$anonfun$org$apache$spark$sql$execution$streaming$state$HDFSBackedStateStoreProvider$$loadMap$1$$anonfun$6.apply(HDFSBackedStateStoreProvider.scala:362)
at org.apache.spark.sql.execution.streaming.state.HDFSBackedStateStoreProvider$$anonfun$org$apache$spark$sql$execution$streaming$state$HDFSBackedStateStoreProvider$$loadMap$1$$anonfun$6.apply(HDFSBackedStateStoreProvider.scala:359)
at scala.Option.getOrElse(Option.scala:121)
at org.apache.spark.sql.execution.streaming.state.HDFSBackedStateStoreProvider$$anonfun$org$apache$spark$sql$execution$streaming$state$HDFSBackedStateStoreProvider$$loadMap$1.apply(HDFSBackedStateStoreProvider.scala:359)
at org.apache.spark.sql.execution.streaming.state.HDFSBackedStateStoreProvider$$anonfun$org$apache$spark$sql$execution$streaming$state$HDFSBackedStateStoreProvider$$loadMap$1.apply(HDFSBackedStateStoreProvider.scala:358)
at scala.Option.getOrElse(Option.scala:121)
at org.apache.spark.sql.execution.streaming.state.HDFSBackedStateStoreProvider.org$apache$spark$sql$execution$streaming$state$HDFSBackedStateStoreProvider$$loadMap(HDFSBackedStateStoreProvider.scala:358)
at org.apache.spark.sql.execution.streaming.state.HDFSBackedStateStoreProvider$$anonfun$org$apache$spark$sql$execution$streaming$state$HDFSBackedStateStoreProvider$$loadMap$1$$anonfun$6.apply(HDFSBackedStateStoreProvider.scala:360)
at org.apache.spark.sql.execution.streaming.state.HDFSBackedStateStoreProvider$$anonfun$org$apache$spark$sql$execution$streaming$state$HDFSBackedStateStoreProvider$$loadMap$1$$anonfun$6.apply(HDFSBackedStateStoreProvider.scala:359)
at scala.Option.getOrElse(Option.scala:121)
at org.apache.spark.sql.execution.streaming.state.HDFSBackedStateStoreProvider$$anonfun$org$apache$spark$sql$execution$streaming$state$HDFSBackedStateStoreProvider$$loadMap$1.apply(HDFSBackedStateStoreProvider.scala:359)
at org.apache.spark.sql.execution.streaming.state.HDFSBackedStateStoreProvider$$anonfun$org$apache$spark$sql$execution$streaming$state$HDFSBackedStateStoreProvider$$loadMap$1.apply(HDFSBackedStateStoreProvider.scala:358)
at scala.Option.getOrElse(Option.scala:121)
at org.apache.spark.sql.execution.streaming.state.HDFSBackedStateStoreProvider.org$apache$spark$sql$execution$streaming$state$HDFSBackedStateStoreProvider$$loadMap(HDFSBackedStateStoreProvider.scala:358)
at org.apache.spark.sql.execution.streaming.state.HDFSBackedStateStoreProvider$$anonfun$org$apache$spark$sql$execution$streaming$state$HDFSBackedStateStoreProvider$$loadMap$1$$anonfun$6.apply(HDFSBackedStateStoreProvider.scala:360)
at org.apache.spark.sql.execution.streaming.state.HDFSBackedStateStoreProvider$$anonfun$org$apache$spark$sql$execution$streaming$state$HDFSBackedStateStoreProvider$$loadMap$1$$anonfun$6.apply(HDFSBackedStateStoreProvider.scala:359)
at scala.Option.getOrElse(Option.scala:121)
at org.apache.spark.sql.execution.streaming.state.HDFSBackedStateStoreProvider$$anonfun$org$apache$spark$sql$execution$streaming$state$HDFSBackedStateStoreProvider$$loadMap$1.apply(HDFSBackedStateStoreProvider.scala:359)
at org.apache.spark.sql.execution.streaming.state.HDFSBackedStateStoreProvider$$anonfun$org$apache$spark$sql$execution$streaming$state$HDFSBackedStateStoreProvider$$loadMap$1.apply(HDFSBackedStateStoreProvider.scala:358)
at scala.Option.getOrElse(Option.scala:121)
at org.apache.spark.sql.execution.streaming.state.HDFSBackedStateStoreProvider.org$apache$spark$sql$execution$streaming$state$HDFSBackedStateStoreProvider$$loadMap(HDFSBackedStateStoreProvider.scala:358)
at org.apache.spark.sql.execution.streaming.state.HDFSBackedStateStoreProvider$$anonfun$org$apache$spark$sql$execution$streaming$state$HDFSBackedStateStoreProvider$$loadMap$1$$anonfun$6.apply(HDFSBackedStateStoreProvider.scala:360)
at org.apache.spark.sql.execution.streaming.state.HDFSBackedStateStoreProvider$$anonfun$org$apache$spark$sql$execution$streaming$state$HDFSBackedStateStoreProvider$$loadMap$1$$anonfun$6.apply(HDFSBackedStateStoreProvider.scala:359)
at scala.Option.getOrElse(Option.scala:121)
at org.apache.spark.sql.execution.streaming.state.HDFSBackedStateStoreProvider$$anonfun$org$apache$spark$sql$execution$streaming$state$HDFSBackedStateStoreProvider$$loadMap$1.apply(HDFSBackedStateStoreProvider.scala:359)
at org.apache.spark.sql.execution.streaming.state.HDFSBackedStateStoreProvider$$anonfun$org$apache$spark$sql$execution$streaming$state$HDFSBackedStateStoreProvider$$loadMap$1.apply(HDFSBackedStateStoreProvider.scala:358)
at scala.Option.getOrElse(Option.scala:121)
at org.apache.spark.sql.execution.streaming.state.HDFSBackedStateStoreProvider.org$apache$spark$sql$execution$streaming$state$HDFSBackedStateStoreProvider$$loadMap(HDFSBackedStateStoreProvider.scala:358)
at org.apache.spark.sql.execution.streaming.state.HDFSBackedStateStoreProvider$$anonfun$org$apache$spark$sql$execution$streaming$state$HDFSBackedStateStoreProvider$$loadMap$1$$anonfun$6.apply(HDFSBackedStateStoreProvider.scala:360)
at org.apache.spark.sql.execution.streaming.state.HDFSBackedStateStoreProvider$$anonfun$org$apache$spark$sql$execution$streaming$state$HDFSBackedStateStoreProvider$$loadMap$1$$anonfun$6.apply(HDFSBackedStateStoreProvider.scala:359)

Please help me out for the same. There may be workarounds for this issue, please suggest me if any, or may be it is a bug.

zero323
  • 322,348
  • 103
  • 959
  • 935
Rohit Verma
  • 51
  • 1
  • 4

1 Answers1

1

What's your checkpoint location? This is usually because you are using the local file system to store checkpoints. Make sure you set the "checkpointLocation" option and it points to a distributed file system (such as HDFS) that can be accessed by all nodes. [1]

[1] http://spark.apache.org/docs/latest/structured-streaming-programming-guide.html#recovering-from-failures-with-checkpointing

zsxwing
  • 20,270
  • 4
  • 37
  • 59
  • 1
    I have the same issue. My checkpoint location is in Google cloud storage which can be accessed by all the nodes. Still I face this issue. What would be the problem? – passionate Dec 04 '17 at 03:33
  • I had the same situation. Switching from local filesystem to hdfs worked for me. But if you ask me, using local filesystem for checkpointing must either work or be technically forbidden in case it's not supported. At the moment, a local path for checkpointing is supported but does not really work. This is not really a good solution. – Aydin K. May 08 '18 at 12:05
  • 1
    I am also facing the same issue and my checkpoint location is AWS S3 Still appearing the same issue. What would be the possible root cause? – Rajesh Kumar Sep 02 '22 at 03:56