Does kafka automatically perform unclean leader election for partitions?
I have topic
./kafka-topics.sh --bootstrap-server XXX --describe --topic my-topic
Topic: my-topic TopicId: q7VUw5a6T_youSEL3WPLGQ PartitionCount: 60 ReplicationFactor: 6 Configs: min.insync.replicas=1,segment.bytes=1073741824,unclean.leader.election.enable=true
Topic: my-topic Partition: 0 Leader: none Replicas: 6,5,4,3,2,1 Isr: 6
Topic: my-topic Partition: 1 Leader: none Replicas: 5,6,4,3,2,1 Isr: 6
Topic: my-topic Partition: 2 Leader: none Replicas: 4,5,6,3,2,1 Isr: 6
Topic: my-topic Partition: 3 Leader: none Replicas: 6,5,4,3,2,1 Isr: 6
Topic: my-topic Partition: 4 Leader: none Replicas: 5,6,4,3,2,1 Isr: 6
Topic: my-topic Partition: 5 Leader: none Replicas: 4,5,6,3,2,1 Isr: 6
Topic: my-topic Partition: 6 Leader: none Replicas: 6,5,4,3,2,1 Isr: 6
I can run script kafka-leader-election.sh with option –election-type UNCLEAN. It helps – after that partitions have leader and ISR list contains all available replicas. But my question is about automatic leader election – without manual intervention – is it possible? Or, maybe, unclean leader election is too dangerious and I can perform it only manually?
alex whoodoo is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.