java.lang.RuntimeException: java.lang.NullPointerException in InflightRecordTracker

[geota]

We are seeing periodically these error messages in our logs. Brain dumping this issue here for now, will follow up with more information as we debug this issue.

In our KinesisSpout fail function, we call "Record record = stateManager.getInflightRecord(shardId, seqNum);" which leads to a NullPointerException.

{
    "timestamp": "2015-06-26T05:08:58.820+0000",
    "level": "ERROR",
    "thread": "Thread-91-kinesis-spout",
    "logger": "backtype.storm.util",
    "message": "Async loop died!",
    "context": "default",
    "exception": "java.lang.RuntimeException: java.lang.NullPointerException\n\tat backtype.storm.utils.DisruptorQueue.consumeBatchToCursor(DisruptorQueue.java:128)\n\tat backtype.storm.utils.DisruptorQueue.consumeBatch(DisruptorQueue.java:87)\n\tat backtype.storm.disruptor$consume_batch.invoke(disruptor.clj:76)\n\tat backtype.storm.daemon.executor$fn__4654$fn__4669$fn__4698.invoke(executor.clj:542)\n\tat backtype.storm.util$async_loop$fn__458.invoke(util.clj:463)\n\tat clojure.lang.AFn.run(AFn.java:24)\n\tat java.lang.Thread.run(Thread.java:745)\nCaused by: java.lang.NullPointerException: null\n\tat com.amazonaws.services.kinesis.stormspout.state.zookeeper.InflightRecordTracker.getInflightRecord(InflightRecordTracker.java:210)\n\tat com.amazonaws.services.kinesis.stormspout.state.zookeeper.LocalShardState.getInflightRecord(LocalShardState.java:92)\n\tat com.amazonaws.services.kinesis.stormspout.state.zookeeper.ZookeeperStateManager.getInflightRecord(ZookeeperStateManager.java:229)\n\tat com.amazonaws.services.kinesis.stormspout.KinesisSpout.fail(KinesisSpout.java:260)\n\tat backtype.storm.daemon.executor$fail_spout_msg.invoke(executor.clj:372)\n\tat backtype.storm.daemon.executor$fn$reify__4657.expire(executor.clj:432)\n\tat backtype.storm.utils.RotatingMap.rotate(RotatingMap.java:73)\n\tat backtype.storm.daemon.executor$fn__4654$tuple_action_fn__4660.invoke(executor.clj:437)\n\tat backtype.storm.daemon.executor$mk_task_receiver$fn__4645.invoke(executor.clj:404)\n\tat backtype.storm.disruptor$clojure_handler$reify__1446.onEvent(disruptor.clj:58)\n\tat backtype.storm.utils.DisruptorQueue.consumeBatchToCursor(DisruptorQueue.java:125)\n\t... 6 common frames omitted\n",
    "source_tag": "tail.dat-cer-sup.workerlog",
    "@timestamp": "2015-06-26T05:08:58+00:00"
  }

[geota]

Believe this is an issue in our forked code. Closing for now.