[rabbitmq-discuss] A new exchange type

[Ben Hood]

Ben,

On Thu, Jan 29, 2009 at 2:59 PM, Ben Hyde <bhyde at pobox.com> wrote:
> This cache of last msg of a given key exchange type is useful only if
> the messages are
> absolute, rather than a series of diffs on the state of the thing
> denoted by the key.  Is that
> the typical pattern in these parts?

I don't know if that generally holds, I guess you could envisage some
LCA scheme. However, for all practical purposes, I think this is more
in the application domain, or is at least an overlay concern rather
than something intrinsic to the caching mechanism. What I am trying to
say is that this is has more to do with a usage pattern rather than
the fundamental messaging infrastructure.

> I guess
> it might be a good rule of thumb to add guards to enforce sell-by-
> dates on all
> the message handling.  Is that common?

No, but it should be :-)

We are currently re-implementing the queue data structure to be able
to set TTLs on a per-message basis. No ETA on that feature, though.

> I don't know if all that means this should be in the broker or not.

I think this kind of thing should be a layered concern.

HTH,

Ben