[jira] [Commented] (ZOOKEEPER-3314) Document the possibility of MultiCallback receiving a null pointer

["Hudson (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/ZOOKEEPER-3314?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16813839#comment-16813839 ] 

Hudson commented on ZOOKEEPER-3314:
-----------------------------------

SUCCESS: Integrated in Jenkins build Zookeeper-trunk-single-thread #304 (See [https://builds.apache.org/job/Zookeeper-trunk-single-thread/304/])
ZOOKEEPER-3314: Document the possibility of MultiCallback receiving a (andor: rev 9a7cebb96a8fe810fbeea18d911212076424feaa)
* (edit) zookeeper-server/src/main/java/org/apache/zookeeper/AsyncCallback.java


> Document the possibility of MultiCallback receiving a null pointer
> ------------------------------------------------------------------
>
>                 Key: ZOOKEEPER-3314
>                 URL: https://issues.apache.org/jira/browse/ZOOKEEPER-3314
>             Project: ZooKeeper
>          Issue Type: Improvement
>    Affects Versions: 3.4.7, 3.4.8, 3.4.9, 3.4.10, 3.5.0, 3.5.1, 3.5.2, 3.5.3, 3.4.11, 3.5.4, 3.4.12, 3.4.13
>            Reporter: Steven McDonald
>            Assignee: Steven McDonald
>            Priority: Trivial
>              Labels: pull-request-available
>             Fix For: 3.6.0
>
>         Attachments: 0001-Document-the-case-where-opResults-is-null-in-multi-c.patch
>
>          Time Spent: 0.5h
>  Remaining Estimate: 0h
>
> A {{MultiCallback}} can receive a null pointer on failure, rather than a list of {{org.apache.zookeeper.OpResult.ErrorResult}} as documented. This is evident from [the implementation|https://github.com/apache/zookeeper/blob/master/zookeeper-server/src/main/java/org/apache/zookeeper/ClientCnxn.java#L689].
> This causes NullPointerExceptions in Kafka 2.1.x (see [KAFKA-7898|https://issues.apache.org/jira/browse/KAFKA-7898]). Kafka 2.0.x does not use the async multi interface, and Kafka 2.2.x handles the null pointer case.
> However, this is enough of a hazard that it should be documented. I have a patch for that which I will try to attach in a moment (JIRA won't allow me to attach it now for some reason).



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)