[
https://issues.apache.org/jira/browse/KAFKA-612?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Jun Rao updated KAFKA-612:
--------------------------
Attachment: kafka-612-v2.patch
Thanks for the review. Attach patch v2.
1.1 This is to handle the case that a partition is removed from the fetchMap
while a fetch request is issued to the broker (since we don't hold the lock
when making fetch requests). When this happens, we just need to ignore this
partition.
1.2 Good suggestion. Fixed.
2. That info message is really used to track how long it takes a broker to
complete the leaderAndIsr request. From the broker's perspective, once the
leader is set in a partition, the partition can serve read/write requests.
3.1 Agreed that the naming is confusing. Changed it to partitionMap.
3.2 Done.
> move shutting down of fetcher thread out of critical path
> ---------------------------------------------------------
>
> Key: KAFKA-612
> URL: https://issues.apache.org/jira/browse/KAFKA-612
> Project: Kafka
> Issue Type: Bug
> Components: core
> Affects Versions: 0.8
> Reporter: Jun Rao
> Assignee: Jun Rao
> Attachments: kafka-612.patch, kafka-612-v2.patch
>
>
> Shutting down a fetch thread seems to take more than 200ms since we need to
> interrupt the thread. Currently, we shutdown fetcher threads while processing
> a leaderAndIsr request. This can delay some of the partitions to become a
> leader.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
