[
https://issues.apache.org/jira/browse/QPID-6063?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14121412#comment-14121412
]
ASF subversion and git services commented on QPID-6063:
-------------------------------------------------------
Commit 1622488 from [~k-wall] in branch 'qpid/trunk'
[ https://svn.apache.org/r1622488 ]
QPID-6063: [Java Broker] BDB HA - report node role correctly after node
restart in group without quorum or node error
* if a node is restarted but there are too few nodes attached to achieve
quorum, the role is now reported as WAITING (previously it was erroneously
reported as DETACHED)
* if a node goes into error state (for example in response to an intruder), the
role is now reported as DETACHED (previously it continued to report its
previous role)
> HA Node role UNKNOWN means both not-reachable and awaiting election
> -------------------------------------------------------------------
>
> Key: QPID-6063
> URL: https://issues.apache.org/jira/browse/QPID-6063
> Project: Qpid
> Issue Type: Improvement
> Components: Java Broker
> Reporter: Keith Wall
> Assignee: Alex Rudyy
> Fix For: 0.31
>
>
> Node role "UNKNOWN" is currently used to represent two distinct situations:
> * A remote node that is not reachable from this node
> * The local node is awaiting the result of an election, or awaiting
> sufficient nodes to contact the group in order for an election to be held.
> It is important that the user can distinguish between these two situations in
> order to have a complete understanding of the behaviour of the group.
> Role UNKNOWN will be replaced by UNREACHABLE and WAITING to represent the two
> distinct situations.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
