-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/21899/#review44544
-----------------------------------------------------------



core/src/main/scala/kafka/utils/ReplicationUtils.scala
<https://reviews.apache.org/r/21899/#comment78969>

    Updating the zookeeper path if the contents are equal even if the version 
does not match seems to be widely applicable. So it should be generic API, not 
just limited to leader, isr or for that matter, any replication specific logic. 
How about making it a generic utility in ZkUtils that takes in a data 
comparator that the user implements and returns true with the new version only 
if the data matches? In fact, maybe even conditionalUpdatePersistentPath could 
be changed to accept a comparator. If the comparator is not passed in, it skips 
the check



core/src/main/scala/kafka/utils/ReplicationUtils.scala
<https://reviews.apache.org/r/21899/#comment78968>

    parseLeaderAndIsr can also be part of ReplicationUtils.
    
    


- Neha Narkhede


On May 31, 2014, 10:50 p.m., Sriharsha Chintalapani wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/21899/
> -----------------------------------------------------------
> 
> (Updated May 31, 2014, 10:50 p.m.)
> 
> 
> Review request for kafka.
> 
> 
> Bugs: KAFKA-1382
>     https://issues.apache.org/jira/browse/KAFKA-1382
> 
> 
> Repository: kafka
> 
> 
> Description
> -------
> 
> KAFKA-1382. Update zkVersion on partition state update failures.
> 
> 
> Diffs
> -----
> 
>   core/src/main/scala/kafka/cluster/Partition.scala 
> 518d2df5ae702d8c0937e1f9603fd11a54e24be8 
>   core/src/main/scala/kafka/controller/KafkaController.scala 
> e776423b8a38da6f08b2262c8141abf2064d37d2 
>   core/src/main/scala/kafka/utils/ReplicationUtils.scala PRE-CREATION 
> 
> Diff: https://reviews.apache.org/r/21899/diff/
> 
> 
> Testing
> -------
> 
> 
> Thanks,
> 
> Sriharsha Chintalapani
> 
>

Reply via email to