From: Erick B. (erickbe@xxxxxxxxx)
Date: Wed Aug 30 2000 - 17:03:40 GMT-3
Hi, let me explain my understanding of inverse-arp and
DLCI mappings. Alot of this understanding comes from
my experience with Bay FR service records on a regular
basis. IOS works in a similar fashion. Don't kill me
for saying Bay :)
By default all DLCIs not mapped or specified with the
interface-dlci are assigned/dynamically mapped to the
major/primary interface. Inverse arp enabled for
DLCIs.
If you use the interface-dlci command it will assign
that dlci to that subinterface/interface and use
inverse-arp still on that DLCI (dynamic).
If you use the map command then inverse-arp is
disabled on that DLCI and protocol (static). I've
heard mixed stories about this. Some say map
statements will disable inverse-arp on that
subinterface and others say just on that DLCI. I need
to look into this in more detail myself for
clarification.
If you use frame map or interface-dlci statements then
those DLCIs will be moved/assigned to the
subinterfaces instead of the major/primary interface.
Lastly, 'show frame map' is a godsend.
HTH, Erick 9/21-22 SJ
PS: On another note, it is common knowledge that
static routes on the lab are not allowed/recommended.
How does staticly mapping frame relay, etc differ? If
it's static and your DLCI, remote address changes it's
not going to work - just like a static route if the
next hop goes away.
--- mark salmon <masalmon@cisco.com> wrote:
> I thought that the point of the frame relay
> interface dlci command
> negates the need for ARP. It seems to me that the
> frame relay interface
> dlci command maps the dlci to the subinterface. I
> am referring to pure
> inverse arp.
>
> "Erick B." wrote:
> >
> > Simon's config is using inverse arp on s0.2
> > subinterface and the show frame map statement
> displays
> > that the address's are dynamically learned on s0.2
> so
> > inverse arp is working.
> >
> > --- mark salmon <masalmon@cisco.com> wrote:
> > > No can do you are using map statements. My
> > > contention is to use inverse
> > > arp. I realize that you can use map statements
> to
> > > achieve
> > > reachability. I wish ot use inverse arps on the
> hub
> > > router.
> > >
> > > Simon Baxter wrote:
> > > >
> > > > Yip, just added it just for you!!
> > > >
> > > > interface Serial0
> > > > ip address 192.1.1.2 255.255.255.0
> > > > encapsulation frame-relay
> > > > no ip mroute-cache
> > > > ip policy route-map policy
> > > > frame-relay traffic-shaping
> > > > frame-relay priority-dlci-group 1 100 200 300
> 400
> > > > frame-relay map bridge 400 broadcast
> > > > frame-relay map ip 192.1.1.1 100 broadcast
> > > > frame-relay map ipx A.0000.0c01.1235 300
> > > broadcast
> > > > frame-relay map appletalk 300.1 200 broadcast
> > > > no frame-relay inverse-arp
> > > > frame-relay qos-autosense
> > > > !
> > > > interface Serial0.2 multipoint
> > > > ip address 202.1.1.2 255.255.255.0
> > > > cdp enable
> > > > frame-relay interface-dlci 500
> > > > !
> > > >
> > > > RTRB#
> > > > RTRB#show frame map
> > > > Serial0 (up): bridge dlci 400(0x190,0x6400),
> > > static,
> > > > broadcast,
> > > > CISCO, status defined, active
> > > > Serial0 (up): ip 192.1.1.1 dlci
> 100(0x64,0x1840),
> > > static,
> > > > broadcast,
> > > > CISCO, status defined, active
> > > > Priority DLCI Group 1, DLCI 100 (HIGH), DLCI
> 200
> > > (MEDIUM)
> > > > DLCI 300 (NORMAL), DLCI 400 (LOW)
> > > > Serial0.2 (up): ip 202.1.1.1 dlci
> > > 500(0x1F4,0x7C40), dynamic,
> > > > broadcast,, status defined,
> active
> > > > Serial0 (up): ipx A.0000.0c01.1235 dlci
> > > 300(0x12C,0x48C0), static,
> > > > broadcast,
> > > > CISCO, status defined, active
> > > > Serial0 (up): appletalk 300.1 dlci
> > > 200(0xC8,0x3080), static,
> > > > broadcast,
> > > > CISCO, status defined, active
> > > > RTRB#ping 202.1.1.1
> > > >
> > > > Type escape sequence to abort.
> > > > Sending 5, 100-byte ICMP Echos to 202.1.1.1,
> > > timeout is 2 seconds:
> > > > !!!!!
> > > > Success rate is 100 percent (5/5), round-trip
> > > min/avg/max = 56/59/60 ms
> > > > RTRB#
> > > >
> > > > as you'll see, everything else apart from s0.2
> is
> > > static and no inverse
> > > > arped...
> > > >
> > > > Simon
> > > >
> > > > -----Original Message-----
> > > > From: mark salmon [mailto:masalmon@cisco.com]
> > > > Sent: Wednesday, August 30, 2000 2:31 PM
> > > > To: ccielab@groupstudy.com
> > > > Subject: Inverse ARP and Subinterfaces
> > > >
> > > > HAs anyone been able to get inverse arp to
> work
> > > with frame relay
> > > > multipoint subinterfaces? According to
> Caslow,
> > > multipoint subinterfaces
> > > > do inverse arp by default. I have not been
> able
> > > to set it up that way
> > > > in a hub and spoke environment (both sides
> > > multipoint subinterfaces).
> > > >
> > > > Any ideas?
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 08:24:33 GMT-3