Hi, Yao.
The cleanup procedure will be updated.
Because some of current cleanup procedures is not same as the specification.
We must follow the specification.
Actual specification will take the following.
TN NUT
| |
|---> | unsolicited NA (multicast, S=0&O=1, different TLL)
| |
|------>| Echo Request
|<------| Echo Reply
| |
* | Wait (DELAY_FIRST_PROBE_TIME)
| |
|<------| NS (unicast)
|<------| NS (unicast)
|<------| NS (unicast)
| |
V V
It will work for everything.
// Personally, I wanted to consider NA is sending a packet.
// But, specification accepts both. :-)
Thanks,
On Mon, 28 Jan 2008 17:03:24 -0700
Yinghui Yao <Yinghui.Yao@alcatel-lucent.com> wrote:
> Hi,
>
> Has anyone especially TAHI guys ever questioned wisdom behind the
> sequence that TAHI clears NUT's ND entry
>
> TN NUT
> | |
> | ----> | NS (unicast) (different cached address)
> | <---- | NA
> | |
> * | Wait (DELAY_FIRST_PROBE_TIME)
> | |
> | <---- | NS (unicast)
> | <---- | NS (unicast)
> | <---- | NS (unicast)
> | |
> V V
>
> I understand it's based on RFC 2461 section 7.3.3
>
> "The first time a node sends a packet to a neighbor whose entry is
> STALE, the sender changes the state to DELAY and a sets a timer to
> expire in DELAY_FIRST_PROBE_TIME seconds."
>
> Does "NA packet sent from NUT" count as "sending a packet". I would
> think "sending a packet"
> really means sending packets other than ND e.g. ICMP, TCP/UDP packets.
>
> Thanks,
> Yinghui Yao
> Alcatel-Lucent
>
>
>
>
>
>
>
>
>
------------------------------------------------------------------------
Yukiyo Akisada <akisada@tahi.org>