You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@qpid.apache.org by Martin Ritchie <ri...@apache.org> on 2009/01/27 13:21:40 UTC
ACL Queue Browser consumer
Hi,
Another item I was thinking of from the JMS point of view was QueueBrowsing.
Currently the Java Broker with 0-8/9 uses a custom property 'no_consume' to
determine that the new consumer should be a browser. Ideally we should be
able to control this independently from queue consumption with the ACLs
however that is not currently possible.
I've JIRA'd the fact (QPID-1617) and one possible solution of using a
no_consume property. However, I wasn't sure how this would work with 0-10 or
even 1-0. It would be good to get a format that we can use in all of the
protocol versions.
I imagine it would be quite difficult/confusing to configure the ACLs in the
Java broker (when we support multiple version) if we need a special queue
browsing acl for each protocol version.
Thoughts on the best way to represent a queue browser in the acl file?
Of course that is all dependant on the ability to determine that a
connecting user is actually a browser in 0-10 / 1-0 and not just a consumer
that hasn't yet consumed. Checking to ensure the acquire mode is NOT_AQUIRED
would seem like a possible solution.
Regards
Martin
--
Martin Ritchie
Re: ACL Queue Browser consumer
Posted by Martin Ritchie <ri...@apache.org>.
2009/1/27 Carl Trieloff <cc...@redhat.com>:
>
> yes, we should be able to handle this with case already on the C++ side, it
> would be an 'access' policy
> on the queue. This will allow browsing via the correct acquire-mode
> properties for 0-10 and covered
> by the existing ACL schema
>
> Might be worth adding an explicit test for it, if we don't have one already.
>
> make sense?
> Carl
How would that then tie in with management rights? Would that user not
the have rights to access the queue to see the properties of the
queue?
Cheers
Martin
> Martin Ritchie wrote:
>>
>> Hi,
>>
>> Another item I was thinking of from the JMS point of view was
>> QueueBrowsing.
>> Currently the Java Broker with 0-8/9 uses a custom property 'no_consume'
>> to
>> determine that the new consumer should be a browser. Ideally we should be
>> able to control this independently from queue consumption with the ACLs
>> however that is not currently possible.
>>
>> I've JIRA'd the fact (QPID-1617) and one possible solution of using a
>> no_consume property. However, I wasn't sure how this would work with 0-10
>> or
>> even 1-0. It would be good to get a format that we can use in all of the
>> protocol versions.
>>
>> I imagine it would be quite difficult/confusing to configure the ACLs in
>> the
>> Java broker (when we support multiple version) if we need a special queue
>> browsing acl for each protocol version.
>>
>> Thoughts on the best way to represent a queue browser in the acl file?
>>
>> Of course that is all dependant on the ability to determine that a
>> connecting user is actually a browser in 0-10 / 1-0 and not just a
>> consumer
>> that hasn't yet consumed. Checking to ensure the acquire mode is
>> NOT_AQUIRED
>> would seem like a possible solution.
>>
>> Regards
>>
>> Martin
>>
>>
>
>
> ---------------------------------------------------------------------
> Apache Qpid - AMQP Messaging Implementation
> Project: http://qpid.apache.org
> Use/Interact: mailto:dev-subscribe@qpid.apache.org
>
>
--
Martin Ritchie
---------------------------------------------------------------------
Apache Qpid - AMQP Messaging Implementation
Project: http://qpid.apache.org
Use/Interact: mailto:dev-subscribe@qpid.apache.org
Re: ACL Queue Browser consumer
Posted by Carl Trieloff <cc...@redhat.com>.
yes, we should be able to handle this with case already on the C++ side,
it would be an 'access' policy
on the queue. This will allow browsing via the correct acquire-mode
properties for 0-10 and covered
by the existing ACL schema
Might be worth adding an explicit test for it, if we don't have one already.
make sense?
Carl
Martin Ritchie wrote:
> Hi,
>
> Another item I was thinking of from the JMS point of view was QueueBrowsing.
> Currently the Java Broker with 0-8/9 uses a custom property 'no_consume' to
> determine that the new consumer should be a browser. Ideally we should be
> able to control this independently from queue consumption with the ACLs
> however that is not currently possible.
>
> I've JIRA'd the fact (QPID-1617) and one possible solution of using a
> no_consume property. However, I wasn't sure how this would work with 0-10 or
> even 1-0. It would be good to get a format that we can use in all of the
> protocol versions.
>
> I imagine it would be quite difficult/confusing to configure the ACLs in the
> Java broker (when we support multiple version) if we need a special queue
> browsing acl for each protocol version.
>
> Thoughts on the best way to represent a queue browser in the acl file?
>
> Of course that is all dependant on the ability to determine that a
> connecting user is actually a browser in 0-10 / 1-0 and not just a consumer
> that hasn't yet consumed. Checking to ensure the acquire mode is NOT_AQUIRED
> would seem like a possible solution.
>
> Regards
>
> Martin
>
>
---------------------------------------------------------------------
Apache Qpid - AMQP Messaging Implementation
Project: http://qpid.apache.org
Use/Interact: mailto:dev-subscribe@qpid.apache.org