Re: Arps and grat arps was (RE: tsv-dir review ofdraft-ietf-vrrp-unified-spec-02.txt)

Sorry - my confusion - the text does say ARP *request*, but I was thinking
of ARP responses.  You are of course correct.

 - Mark

On Tue, Nov 4, 2008 at 9:04 PM, don provan <[email protected]> wrote:

>  Do they sniff MAC addresses in ARP packets? I didn't know that
> was legal.
>
> The source MAC address in the ARP packet's MAC header must always
> be the physical MAC address of the system, of course, never the
> VR MAC address. I've only ever heard of switches learning based
> on the source MAC address *in* the MAC header. Am I about to have
> a learning experience?
> -don
>
> -----Original Message-----
> *From:* [email protected] [mailto:[email protected]]*On
> Behalf Of *Mark Handley
> *Sent:* Tuesday, November 04, 2008 12:36 PM
> *To:* don provan
> *Cc:* Stephen Nadas; TSV Dir; vrrp
> *Subject:* Re: [VRRP] Arps and grat arps was (RE: tsv-dir review
> ofdraft-ietf-vrrp-unified-spec-02.txt)
>
>
>
> On Tue, Nov 4, 2008 at 7:27 PM, don provan <[email protected]> wrote:
>
>>
>> But it doesn't really matter: all routers *can* send gratuitous
>> ARP messages for the VR. The VR ARP information is constant: it
>> does not change when the VRRP master changes. No black hole would
>> be created: all the entries map the VR IP address to the VR MAC
>> address. The current master is the node that receives all packets
>> sent to the VR MAC address no matter what node actually distributed
>> the VR ARP mapping.
>
>
> Won't you blackhole the traffic if you do this on a LAN composed of
> MAC-learning switches?
>
>  - Mark
>
>

_______________________________________________
vrrp mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/vrrp