[KAFKA-12775] StreamsPartitionAssignor / ClientState throws an exception when a new Task gets added to a KStreams Application in a Backwards-Compatible Topology Change

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Invalid
• Component/s: streams
• Assignee: Unassigned
• Reporter: Nico Habermann
• Created: 12/May/21 12:29
• Updated: 14/May/21 06:58
• Resolved: 12/May/21 18:09

Description

KAFKA-6145 and KAFKA-10079 added an exception if the Partition Assignor tries to look up the lag for a TaskId that seemingly does not exist.

 

I believe this is a functional regression.

Before, it was possible for Streams users to make backwards-compatible topology changes and roll those out, without having to do a complete restore or reload.

For example:

Existing sample topology:
 

stream1 = stream(topic)

stream1
  .map(...)
  .to(output)

And doing this backwards-compatible change:

stream1 = stream(topic)
++ table = stream(topic2).through(repartition-topic)/repartition().toTable()

stream1
  .map(...)
++  .join(table)
  .to(output)

 

This effectively creates a new subtopology with a new task for the table repartition.

In older KStreams versions, it would have been possible to simply roll this change out.
Since 2.6, rolling this out will crash the stream because the linked exception gets thrown when StreamsPartitionAssignor#getPreviousTasksByLag tries to look up the lag for the new table-repartition-task

 

At this time, the only possible way to avoid this exception seems to be deleting all local state and doing a complete restore with the new topology change included.