[
https://issues.apache.org/jira/browse/ZOOKEEPER-2866?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Jeffrey F. Lukman updated ZOOKEEPER-2866:
-----------------------------------------
Attachment: (was: ZK-2866.pdf)
> Reconfig Causes Newly Joined Node to Crash
> ------------------------------------------
>
> Key: ZOOKEEPER-2866
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-2866
> Project: ZooKeeper
> Issue Type: Bug
> Components: leaderElection, quorum, server
> Affects Versions: 3.5.3
> Reporter: Jeffrey F. Lukman
>
> When we run our Distributed system Model Checking (DMCK) in ZooKeeper v3.5.3
> by following the workload in ZK-2778:
> * initially start 2 ZooKeeper nodes
> * start 3 new nodes and let them join the cluster
> * do a reconfiguration where the newly joined will be PARTICIPANTS,
> while the previous 2 nodes change to be OBSERVERS
> We think our DMCK found this following bug:
> * one of the newly joined node crashes due to
> it receives an *unexpected* PROPOSAL message
> from the new leader in the cluster.
> For complete information of the bug, please see the document that is attached.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
