| T.R | Title | User | Personal
Name | Date | Lines |
|---|
| 579.1 | no can do | STAR::SALKEWICZ | It missed... therefore, I am | Wed May 20 1992 11:36 | 8 |
| Nothing has been implemented yet for this,.. including "future"
releases we are runnning in our lab.
What is the need? Just curiousity? A "monitoring" program? A customer
request? Something else?
/Bill
|
| 579.2 | | KONING::KONING | Paul Koning, A-13683 | Wed May 20 1992 15:05 | 10 |
| Do you mean the status of the A/C indicators when a frame you transmitted
comes back around (and is stripped)?
If so: the hardware does not support that, partly because it would have been
very difficult to implement, and partly because this feature has no known
application.
What do you want to do with this?
paul
|
| 579.3 | thanks1 | XLIB::SCHAFER | Mark Schafer, ISV Tech. Support | Fri May 22 1992 12:33 | 37 |
| <<< ISVNET::SYS$SYSDEVICE:[NOTES$LIBRARY]VMS.NOTE;1 >>>
-< VMS Operating System >-
================================================================================
Note 387.2 FDDI - State of "A"/"C" bits on transmits 2 of 2
ISVHUB::VOLZ 30 lines 22-MAY-1992 11:11
-< Answer to your questions >-
--------------------------------------------------------------------------------
>Do you mean the status of the A/C indicators when a frame you
>transmitted comes back around (and is stripped)?
Yes. These are the "A" and "C" indicators I meant.
>Nothing has been implemented yet for this. What is the need? Just
>curiousity? A "monitoring" program? A customer request? Something else?
Well ... there is always the argument "why was it added to the
FDDI specification if it wasn't of any use!"
There are some applications. For example, in TCP/IP it would have
been very nice to invalidate the ARP cache entry for an IP address
when the "A" indicated the address was not recognized. This would
allow for supporting dual-MAC stations (in particular, when the
ring wraps and the "old" cached address is no longer on the ring).
Also, the proNET-10/80 hardware supported this bit (it actually
just had the "C" (copied) indication) and this was useful in that
a higher level application could retransmit the packet since it
knew the peer did not receive it).
>If so: the hardware does not support that, partly because it would have
>been very difficult to implement, and partly because this feature has
>no known application.
Yuck! (again ... why was it added to the FDDI specification?)
But ... since it ain't in there, we'll just have to live with
the restriction.
|
| 579.4 | Wouldn't work anyway... | STAR::SALKEWICZ | It missed... therefore, I am | Fri May 22 1992 13:36 | 11 |
|
Generally speaking,.. the "acknowledging" of frame reception
is a "higher" (higher than the data link) function.
In your example,. if you did see the A bit set, it doesn't
guarantee that the packet actually ever got up to the ARP layer.
It only means that it was recognized by some station out there,. and
not necessarily the station your were interested in...
/Bill
|
| 579.5 | | KONING::KONING | Paul Koning, A-13683 | Tue May 26 1992 17:57 | 7 |
| The A indicator tells you nothing, since there may be bridges.
The reason it's in the FDDI spec is partly history and partly politics; just
remember that there are quite a lot of things in those standards that serve
no observable purpose.
paul
|
| 579.6 | approaching zero value? | STAR::SALKEWICZ | It missed... therefore, I am | Wed May 27 1992 20:04 | 4 |
| ... OK ,.. so its of even less value than I thought :-) :-)
/Bill
|