Can I assume that we make this update for both M2.1 and trunk?
yes/no?
Carl.
Robert Godfrey wrote:
Hopefully any console will clearly show these queues as "temporary"
and "exclusive", so the "tmp_" part was always quite redundant. The
UUID at the end will guarantee uniqueness. I think adding "JMStopic"
is kind of redundant also - it doesn't give an operator any more
insight into the queue's purpose. My feeling is that the above naming
scheme is valid for all temporary queues that are involved in only one
binding.
Currently I don;t have any use cases for temp queues involved in
multiple bindings.
-- Rob
On 20/03/2008, Martin Ritchie <[EMAIL PROTECTED]> wrote:
As Mr. Grieg says this we should remember that this is for temp queues
of all sorts hence the generic format suggested. This doesn't really
have any bearing on JMS. We currently have a similar naming and
binding issue with the java direct exchange. The names start Temp but
the binding is tmp.
On 20/03/2008, Robert Greig <[EMAIL PROTECTED]> wrote:
> On 20/03/2008, Arnaud Simon <[EMAIL PROTECTED]> wrote:
>
> > Rob naming schema looks very good. I would however prepend "JMStopic_"
> > to it. As this scheme is used for simplifying the broker administration,
> > prepending "JMStopic_" would increase the understandability of queue
> > names.
>
> Sorry I don't quite follow this?
>
> The private queues are not a JMS-specific approach and neither are
> they limited to topics - for example the headers exchange has the same
> kind of topology.
>
> RG
>
--
Martin Ritchie
