[
https://issues.apache.org/jira/browse/KAFKA-300?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Neha Narkhede updated KAFKA-300:
--------------------------------
Attachment: kafka-300.patch
This patch includes the following changes -
1. Implement leader election procedure using only assigned replicas list.
Basically, any replica in AR list can become a leader with preference to the
preferred replica (1st replica in AR list)
2. Introduced the Partition and Replica data structures with CUR, ISR, RAR
unused. Also, commitQ will be added as part of KAFKA-46
3. Each broker registers listeners on /brokers/topics and
/brokers/topics/[topic/partitions. Also for each partition assigned to the
broker, it registers leader change listener on
/brokers/topics/[topic]/partitions/[partition]/leader
4. Currently the interaction between KafkaZookeeper and LogManager is
unwieldly, but it will be fixed once KAFKA-307 is resolved.
5. Added LeaderElectionTest and helper code to wait until leader is elected for
a particular topic and partition
> Implement leader election
> -------------------------
>
> Key: KAFKA-300
> URL: https://issues.apache.org/jira/browse/KAFKA-300
> Project: Kafka
> Issue Type: Sub-task
> Reporter: Neha Narkhede
> Attachments: kafka-300.patch
>
>
> According to the Kafka replication design
> (https://cwiki.apache.org/confluence/display/KAFKA/Kafka+Replication), this
> JIRA will involve implementing the leaderElection() procedure.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
