[
https://issues.apache.org/jira/browse/KAFKA-1112?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13821591#comment-13821591
]
Neha Narkhede commented on KAFKA-1112:
--------------------------------------
Thanks for the patch. Few comments -
1. In the most common case of needsRecovery, the position of the last entry
will be zero. In this case, we will search the entire log segment up until the
recovery point. This will slow down server startup but probably only when we
really need recovery.
2. LogSegment: We have to be carefully => We have to be careful
3. Log: If sanityCheck throws an exception, can we automatically invoke index
rebuild instead of bailing out?
4. Could you rebase?
5. Could you give the patch review tool a spin? The setup is minimal and we can
save time for this and future reviews -
https://cwiki.apache.org/confluence/display/KAFKA/Kafka+patch+review+tool#Kafkapatchreviewtool-1.Setup
Usage:
python kafka-patch-review.py -j KAFKA-1112 -b trunk
> broker can not start itself after kafka is killed with -9
> ---------------------------------------------------------
>
> Key: KAFKA-1112
> URL: https://issues.apache.org/jira/browse/KAFKA-1112
> Project: Kafka
> Issue Type: Bug
> Components: log
> Affects Versions: 0.8, 0.8.1
> Reporter: Kane Kim
> Assignee: Jay Kreps
> Priority: Critical
> Attachments: KAFKA-1112-v1.patch, KAFKA-1112.out
>
>
> When I kill kafka with -9, broker cannot start itself because of corrupted
> index logs. I think kafka should try to delete/rebuild indexes itself without
> manual intervention.
--
This message was sent by Atlassian JIRA
(v6.1#6144)
