[rbridge] C Bit Clarification
Donald Eastlake
d3e3e3 at gmail.com
Tue May 11 13:49:47 PDT 2010
Hi David,
On Tue, May 11, 2010 at 4:03 PM, David Michael Bond
<david.bond at iol.unh.edu> wrote:
> Hello Everyone,
>
> I have a question regarding the draft. It’s just a minor clarification.
>
>
>
> Section 4.1.1 says:
>
> The "C" bit shown in Figure 4.2 is not used in TRILL. It MUST be set to zero
> and is ignored by receivers.
>
>
>
> In a transit RBridge situation if an RBridge receives a TRILL data frame
> with the C bit set to one in either the Outer or Inner VLAN should it
> transparently forward the frame or clear the C bit?
I think this is mostly talking about the Inner.VLAN.
Although TRILL specifies the Outer VLAN ID and priority (which affects
queuing) to be used if a frame is sent tagged, it runs above 802.1Q
ports. For 802.3, which is what the standard focuses on, the C bit
should be clear, but I suspect the bit would just be ignored on
receipt, if it were a 1, by most existing 802.1Q/802.3 ports. However,
for more exotic link types, like 802.5, which I don't know much about
and which are not covered by the TRILL standard, I think you might
have to set the C (actually CFI) bit in the Outer.VLAN.
For the Inner.VLAN, I think transit RBridges should just transparently
forward it.
> Also, I believe “[802.1Qad] S-tags, …” Should be “[802.1ad] S-tags, …”
Yes. Only [802.1ad] appears in the References, so this should be caught later.
Thanks,
Donald
> Thanks,
>
> David
>
>
>
> David Bond
>
> .~~~~~.~~~~~.~~~~~.~~~~~~.~~~~~~.~~~~~.~~~~~.~~~~~.
> .University of New Hampshire .
> . - InterOperability Laboratory .
> . Research and Development Software Eng. .
> . Routing & Bridge Functions Consortiums .
> . - Computer Science PhD Student .
> . 121 Technology Drive, Suite 2, Durham NH, 03824 .
> . (O) 1-603-862-3525 (M) 1-603-845-7514 .
>
> . http://mokon.net http://iol.unh.edu .
> .~~~~~.~~~~~.~~~~~.~~~~~~.~~~~~~.~~~~~.~~~~~.~~~~~.
More information about the rbridge
mailing list