On 01/06/2011 06:31 AM, Amit Shah wrote:
Hello,
On (Thu) Dec 16 2010 [09:31:43], Ted Ross wrote:
> There has been some discussion about what the strategy for
> Qpid-over-virtio-serial should be. One of the newer considerations
> is that the local broker (on the guest) should be optional and is
> considered an optimization (for connection concentration). Not
> using a local broker makes for a significantly simpler system
> configuration.
>
> This leaves two possible approaches:
>
> 1) Build a virtio-serial plug-in for Qpid
> 2) Build a tcp-socket proxy that multiplexes tcp connections over
> virtio-serial
>
> The advantage of (1) is that it is a better architectural solution
> and does not require that we create a new package and a new item
> that needs to be configured on guests.
>
> The advantage of (2) is that it might be easier to write in a short time.
I feel (1) is the better option as well.
> Here are some notes on each approach:
>
> 1) virtio-serial plug-in
>
> Qpid already has two value-added transport plug-ins, one for RDMA
> and one for SSL. Both of these are packaged as a set of
> shared-object libraries (a common, a client-side, and a
> server-side). Either of these transports can serve as a template
> for a virtio-serial transport.
For virtio-serial, only a client-side library would be needed.
I'm not sure I understand this. It seems to me that a client-side
library would be needed on the guest and a server-side library would be
needed for the broker on the host.
> The source for these can be found in the Qpid SVN (there's a
git also):
>
>
http://svn.apache.org/repos/asf/qpid/trunk/qpid
Thanks; I'll get some familiarity with this code and post my thoughts.
[...]
Amit