You are viewing a plain text version of this content. The canonical link for it is here.

Posted to commits@helix.apache.org by "Zhen Zhang (JIRA)" <ji...@apache.org> on 2013/10/04 20:08:45 UTC

[jira] [Created] (HELIX-265) helix controller not sending/delay sending transition messages

Zhen Zhang created HELIX-265:
--------------------------------

             Summary: helix controller not sending/delay sending transition messages
                 Key: HELIX-265
                 URL: https://issues.apache.org/jira/browse/HELIX-265
             Project: Apache Helix
          Issue Type: Bug
            Reporter: Zhen Zhang
            Assignee: Zhen Zhang


we are seeing a couple of cases where the participant is using leader-standby state model, and after controller sends OFFLINE->STANDBY transitions, it doesn't sending STANDBY->LEADER transitions for a long time (more than 30 minutes). During the 30 minutes, we are seeing controller rebalancer pipeline runs more 4000 times on that particular cluster.

a few observation:
1) No GC happens on Helix participant, helix controller, and zk server
2) lots of ideal-state changes and controller/rebalancer pipeline is super busy

we can't reproduce this 



--
This message was sent by Atlassian JIRA
(v6.1#6144)