Kafka indexing service lock tasks and do not consume data

Hi, all! I use kafka indexing service for data ingesion. Periodically some indexing task stop consuming and lose data.

In the tasks log i see this error message.

2018-08-22T12:14:15,643 ERROR [task-runner-0-priority-0] io.druid.indexing.kafka.KafkaIndexTask - Resetting Kafka offsets for datasource [dsp_traf_supervisor]: {class=io.druid.indexing.kafka.KafkaIndexTask, partitions=[1]}

2018-08-22T12:14:15,645 ERROR [task-runner-0-priority-0] io.druid.emitter.graphite.GraphiteEmitter - The following alert is dropped, description is [Resetting Kafka offsets for datasource [dsp_traf_supervisor]], severity is [component-failure]

2018-08-22T12:14:15,645 INFO [task-runner-0-priority-0] io.druid.indexing.kafka.KafkaIndexTask - Pausing ingestion until resumed

What is the reason for blocking tasks? How to restart the task and continue consuming process for blocked kafka partition automatically? Without supervisor restart?

Version of druid is 0.12.2

indexing-task-log-index_kafka_dsp_traf_supervisor_86ec66d8e97ba6b_dnjgcjgg.txt (126 KB)

overloard.log (5.39 MB)

supervisor-spec.txt (3.79 KB)

supervisor-status.txt (6.96 KB)

did you find the root cause of the killing?

I’ve the same issue currently on my environment.