[ 
https://issues.apache.org/jira/browse/PROTON-491?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13988085#comment-13988085
 ] 

ASF subversion and git services commented on PROTON-491:
--------------------------------------------------------

Commit 1592009 from r...@apache.org in branch 'proton/trunk'
[ https://svn.apache.org/r1592009 ]

PROTON-491: applied patch from Dominic for receiver name

> proton-c: Messenger uses hard-coded link names of sender-xxx & receiver-xxx
> ---------------------------------------------------------------------------
>
>                 Key: PROTON-491
>                 URL: https://issues.apache.org/jira/browse/PROTON-491
>             Project: Qpid Proton
>          Issue Type: Bug
>          Components: proton-c
>    Affects Versions: 0.6
>            Reporter: Dominic Evans
>            Assignee: Rafael H. Schloming
>              Labels: easyfix
>         Attachments: 01_append_topic_to_receiver_link_name.patch
>
>
> # Abstract
> The qpid-proton 'messenger' implementation currently uses hard-coded 
> receiver-xxx and sender-xxx link names for all attachments.
> # Description
> I believe the AMQP 1.0 spec says here that:- although client-ids do not need 
> to be unique, the triplet of <client-id>, <link-name> and <server-id>  *is* 
> guaranteed to be unique.
> http://docs.oasis-open.org/amqp/core/v1.0/os/amqp-core-transport-v1.0-os.html#doc-idp298752
> However, the qpid-proton Messenger code seems to have all link-name's 
> hard-coded to sender-xxx and receiver-xxx
> As an easyfix, I would suggest that by default the topicname part of the 
> address should be appended to the link-name (instead of -xxx)
> e.g., amqp://localhost/topicName --> sender-topicName / receiver-topicName
> However, that would still cause a clash if the same messenger wanted to make 
> two subscriptions to the same topic, so perhaps the Messenger implementation 
> needs to track what link names it has allocated and ensure a unique name is 
> assigned each time?



--
This message was sent by Atlassian JIRA
(v6.2#6252)

Reply via email to