You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@kafka.apache.org by "Guozhang Wang (JIRA)" <ji...@apache.org> on 2017/06/08 04:17:21 UTC

[jira] [Resolved] (KAFKA-5314) Improve exception handling for state stores

     [ https://issues.apache.org/jira/browse/KAFKA-5314?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Guozhang Wang resolved KAFKA-5314.
----------------------------------
    Resolution: Fixed

Issue resolved by pull request 3135
[https://github.com/apache/kafka/pull/3135]

> Improve exception handling for state stores
> -------------------------------------------
>
>                 Key: KAFKA-5314
>                 URL: https://issues.apache.org/jira/browse/KAFKA-5314
>             Project: Kafka
>          Issue Type: Sub-task
>          Components: streams
>    Affects Versions: 0.11.0.0
>            Reporter: Eno Thereska
>            Assignee: Eno Thereska
>             Fix For: 0.11.1.0
>
>
> RocksDbStore.java throws a mix of exceptions like StreamsException and ProcessorStateException. That needs to be made consistent. 
> Also the exceptions thrown are not documented in the KeyValueStore interface. 
> All the stores (RocksDb, InMemory) need to have consistent exception handling.
> Today a store error is fatal and halts the stream thread that is processing the exceptions. We could explore alternatives, like haling part of the topology that passes through that faulty store, i.e., failing tasks, not the entire thread.



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)