Github user srdo commented on the issue: https://github.com/apache/storm/pull/1679 @jfenc91 Ouch. I've never seen LEO drop like that unless unclean leader election was enabled. Out of curiosity which Kafka version are you running, and do you have unclean leader election enabled for this topic? If not, consider trying the kafka mailing list. I don't think the spout can or should handle the Kafka log suddenly getting truncated like that, since it's really a data loss scenario on the Kafka side. I can't really think of any way the spout could handle it other than resetting to earliest/latest, which it already does. If the culprit is unclean leader election then it would have also happened in Kafka 0.8.
--- If your project is set up for it, you can reply to this email and have your reply appear on GitHub as well. If your project does not have this feature enabled and wishes so, or if the feature is enabled but not working, please contact infrastructure at infrastruct...@apache.org or file a JIRA ticket with INFRA. ---