Experimental Physics and
Industrial Control System

"Jeff Hill" <[email protected]> · Fri, 4 Feb 2005 12:21:46 -0700

> > The CA repeater's purpose is to receive CA server
> beacons and fan them out
> > to all of the CA clients running in the same host.
> Most modern IP kernels
> > now allow multiple listeners on the same UDP port,
> so that is no longer the
> > reason for the CA Repeater. Unfortunately, there is
> still a defect in most
> > IP kernels where UDP broadcasts reach multiple
> listeners on the same UDP
> > port, but UDP unicasts (single host addressed
> datagrams) do not. Therefore,
> > the CA repeater must still exist, and there can only
> be one of them on each
> > host.
>
> This is the case for process-based IOCs as well as
> vxWorks and RTEMS,
> right?
> 

It's actually fairly easy to test this. Just start two
soft IOCs with the same database on the same host.
Launch a client sending searches for PVs in these IOCS.
Configure this client to only send searches to the
unicast address of this host (no broadcasts) and watch
to see if it gets the "PV name duplicated on multiple
servers" message. That message would indicate that the
defect does *not* exist.

As I recall, in the past, at least both vxWorks and
Solaris (and probably also windows) suffered from this
defect. 

It will probably be easier to pursue multicasting
support than it will be to support multiple modes of
unicast operation on different OS.

Jeff

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System