[ 
https://issues.apache.org/jira/browse/QPID-7138?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15241023#comment-15241023
 ] 

Keith Wall commented on QPID-7138:
----------------------------------

Change {{ACO#onOpen()} to locate the tree's root, and then traverse the whole 
tree looking for a duplicate UUIDs.  If a duplicate UUID is encountered then 
the object should go into an ERRORed state.  We need to ensure that two 
virtualhosts, racing cannot both succeed in adding say a Queue with the same 
UUID to the tree at the same moment.  This might require that we hold global 
lock whilst registering a child's UUID, or traversing the tree.

> [Java Broker] Broker should detect duplicate IDs in configured objects
> ----------------------------------------------------------------------
>
>                 Key: QPID-7138
>                 URL: https://issues.apache.org/jira/browse/QPID-7138
>             Project: Qpid
>          Issue Type: Improvement
>          Components: Java Broker
>    Affects Versions: 0.28, 0.30, 0.32, qpid-java-6.0, qpid-java-6.0.1
>            Reporter: Alex Rudyy
>             Fix For: qpid-java-6.1
>
>
> Currently UUID uniqueness is enforced only amongst immediate siblings.  This 
> means that, say, a virtual host with queue with UUID A would reject a second 
> queue with the same UUID A.  Currently if a second virtual host had a queue 
> with same UUID A, this would be allowed.  This presents a problem for clients 
> (including the Web Management Console) that expect UUID to be universally 
> unique.
> Change ACO so that object UUID are universally so.    The validatation would 
> need to be applied as new objects are created, as the Broker starts up and as 
> Virtualhosts are recovered.  This latter may or may not be at start-up time.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to