[infinispan-issues] [JBoss JIRA] (ISPN-3561) A joining cache should receive the rebalancedEnabled flag from the coordinator.

[RH Bugzilla Integration (JIRA)]

    [ https://issues.jboss.org/browse/ISPN-3561?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13026609#comment-13026609 ] 

RH Bugzilla Integration commented on ISPN-3561:
-----------------------------------------------

Dan Berindei <dberinde at redhat.com> changed the Status of [bug 1172550|https://bugzilla.redhat.com/show_bug.cgi?id=1172550] from NEW to ASSIGNED

> A joining cache should receive the rebalancedEnabled flag from the coordinator.
> -------------------------------------------------------------------------------
>
>                 Key: ISPN-3561
>                 URL: https://issues.jboss.org/browse/ISPN-3561
>             Project: Infinispan
>          Issue Type: Feature Request
>          Components: State Transfer
>    Affects Versions: 5.2.7.Final, 5.3.0.Final, 6.0.0.Beta1
>            Reporter: Erik Salter
>            Assignee: Erik Salter
>             Fix For: 7.1.0.Beta1, 7.1.0.Final
>
>
> There is an issue when starting up a set of nodes in a cluster where the coordinator has told the surviving members that state transfer has been disabled.  If rebalancing is disabled while the cluster is running it's disabled on all the 
> However, if a new set of nodes join afterwards, they don't know that rebalancing was disabled.
> This has consequences if there is a new coordinator elected (like during a MERGE) from the set of newly-started nodes.
> To prevent this and ensure the greatest probablility of success, a node joining should get the state of this flag from the response from the coordinator.



--
This message was sent by Atlassian JIRA
(v6.3.8#6338)