One thing i am not sure for the operator state is ...
# troubleshooting
s
One thing i am not sure for the operator state is the recovery part: in case the job get restarted, or we scale up number of kafka partitions, how do we gurantee the operator instance that process a certain kafka partition before will still get the state of the previous operator instance that process the same partition before the restart/scaling? Thread in Slack Conversation
p
Flink does that on the startup process. This post has a good explanation of how this works.