I've never understood why setting DG to itself causes an ARP request for
the *target* IP address?
Normally, for non-local addresses, the ARP would be for the appropriate
gateway, (in this case, itself, which would be somewhat ridiculous), not
the target.
This must be a special hack to the lookup code -- at least I didn't find
it in the RFCs -- and as such, not necessarily universally supported.
Although, it makes as good sense as any as a way to cause an ARP for a
non-local address, which must be somehow configurable in the clients for
Proxy ARP to work in this case (the subnet mask thingy works because the
client thinks that the target is a local address, anyway).
-------------------------------------------------
Tks |
BV |
Sr. Technical Consultant, SBM, A Gates/Arrow Co.
Vox 770-623-3430 11455 Lakefield Dr.
Fax 770-623-3429 Duluth, GA 30097-1511
=================================================
-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]On Behalf Of
[EMAIL PROTECTED]
Sent: Thursday, May 24, 2001 6:05 PM
To: [EMAIL PROTECTED]
Subject: Re: ARP versus Proxy-arp [7:5664]
Chuck,
Proxy-arp is also useful for cases where you have multiple
candidate DG's on the same segment and for whatever reason you
can't or don't want to use HSRP/VRRP, IRDP or passive RIP.
You can achieve a certain amount of load-balancing and failover
using proxy-arp by pointing a hosts DG to its own address and
setting the timeout for arp entries very low on the end-stations.
Course, this increases the amount of ARP bcasts on the segment
and is, essentially, a "cool hack", but it does work.
Regards,
Kent
On 23 May 2001, at 22:36, Chuck Larrieu wrote:
> At the risk of becoming another Bob Vance......
>
> I'm reading Doug Comer's TCP/IP reference, on the assumption that it
> can't hurt to really get into how TCP/IP works.
>
> Proxy-arp versus normal arp.
>
> A host does not know the physical address of another host so it sends
> out an ARP request. If the host in question lies on another network, a
> router responds to that request. Proxy ARP, correct?
>
> A host through it's TCP stack does the XOR and determines that a host
> lies on another network. The host therefore sends the packet to the
> device indicated as its default gateway in its configuration. It sends
> an ARP request for the MAC of the default gateway. Normal ARP?
>
> So in other words, proxy arp may be viewed as something of an obsolete
> protocol / operation in that most modern TCP stacks contain the
> mechanisms for doing the network XOR determination, and then using the
> default gateway. A modern stack would recognize that a host is on a
> different network and go the default gateway route, so to speak.
>
> In other words, the necessity for proxy arp is eliminated for the most
> part because of the default gateway concept and the modern TCP stack.
>
> Has it sunk through this thick head finally?
>
> PS Comer states that proxy arp is aka arp hack. :->
>
> Chuck
>
> One IOS to forward them all.
> One IOS to find them.
> One IOS to summarize them all
> And in the routing table bind them.
>
> -JRR Chambers-
> FAQ, list archives, and subscription info:
> http://www.groupstudy.com/list/cisco.html Report misconduct and
> Nondisclosure violations to [EMAIL PROTECTED]
FAQ, list archives, and subscription info:
http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
Message Posted at:
http://www.groupstudy.com/form/read.php?f=7&i=5839&t=5664
--------------------------------------------------
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
