KAFKA_DATA_LOSS error class


Some data may have been lost because they are not available in Kafka any more; either the data was aged out by Kafka or the topic may have been deleted before all the data in the topic was processed. If you don’t want your streaming query to fail on such cases, set the source option failOnDataLoss to false. Reason:


Added partition <topicPartition> starts from <startOffset> instead of 0.


Could not read records in offset [<startOffset>, <endOffset>) for topic partition <topicPartition>

with consumer group <groupId>.


Cannot find initial offsets for partitions <partitions>. They may have been deleted.


Partitions <partitions> have been deleted.


Partitions <partitions> have been deleted.

Kafka option ‘kafka.<groupIdConfig>’ has been set on this query, it is

not recommended to set this option. This option is unsafe to use since multiple concurrent

queries or sources using the same group id will interfere with each other as they are part

of the same consumer group. Restarted queries may also suffer interference from the

previous run having the same group id. The user should have only one query per group id,

and/or set the option ‘kafka.session.timeout.ms’ to be very small so that the Kafka

consumers from the previous query are marked dead by the Kafka group coordinator before the

restarted query starts running.


Partition <topicPartition> offset was changed from <prevOffset> to <newOffset>.


Starting offset for <topicPartition> was <offset> but consumer reset to <fetchedOffset>.