You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@helix.apache.org by "Hudson (JIRA)" <ji...@apache.org> on 2014/02/24 20:21:21 UTC
[jira] [Commented] (HELIX-390) Race condition: Checking
HelixManager#isConnected is unreliable
[ https://issues.apache.org/jira/browse/HELIX-390?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13910691#comment-13910691 ]
Hudson commented on HELIX-390:
------------------------------
UNSTABLE: Integrated in helix #1233 (See [https://builds.apache.org/job/helix/1233/])
[HELIX-390] Force message handler registration even when ZK connection state is flaky (kanak: rev a302564a345f4c5ff92559fe9cad126acad740c9)
* helix-core/src/main/java/org/apache/helix/messaging/DefaultMessagingService.java
* helix-core/src/main/java/org/apache/helix/manager/zk/ZkHelixParticipant.java
* helix-core/src/main/java/org/apache/helix/manager/zk/ParticipantManagerHelper.java
> Race condition: Checking HelixManager#isConnected is unreliable
> ---------------------------------------------------------------
>
> Key: HELIX-390
> URL: https://issues.apache.org/jira/browse/HELIX-390
> Project: Apache Helix
> Issue Type: Bug
> Reporter: Kanak Biscuitwala
> Assignee: Kanak Biscuitwala
>
> This led to a NPE in trying to execute state transitions due to the fact that the state model was never re-registered after a post-GC reconnection.
> The fix is to invoke the message handler's onConnected method in handleNewSession, and to put all necessary on-connect logic inside this method.
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)