Cisco Support Community
cancel
Showing results for 
Search instead for 
Did you mean: 
Announcements

Welcome to Cisco Support Community. We would love to have your feedback.

For an introduction to the new site, click here. And see here for current known issues.

New Member

Frame Relay Static Mapping and Inverse-ARP

Ref 1. : Configuring Frame Relay, http://www.cisco.com/en/US/docs/ios/12_2/wan/configuration/guide/wcffrely.html#Disabling_or_Reenabling_Frame_Relay_Inverse_ARP

Ref 2. : CCNA ICND2, Official Exam Certification Guide, Second Edition, ISBN 158720181X

Ref 3. : Frame Relay Question, http://forum.cisco.com/eforum/servlet/NetProf?page=netprof&forum=Network%20Infrastructure&topic=WAN%2C%20Routing%20and%20Switching&topicID=.ee71a06&CommCmd=MB%3Fcmd%3Dpass_through%26location%3Doutline%40%5E1%40%40.2cd2b0f5/17#selected_message

Folks,

Is there any way on a router in a frame relay network to verify through an IOS command that Inverse-ARP has been disabled? Ref 2, the ICND2 book, on Page 500, second paragraph says, "The router disables Inverse ARP on a subinterface when the frame-relay map command is configured". There is no mention of this in the 12.2 command reference for 'frame-relay map' and 'frame-relay inverse-arp' commands. At Ref 3, the last post says that Inverse Arp needs to be manually disabled when statically mapping DLCI's to IP addresses. The topology being asked about seems to be point-to-point with subinterfaces. Ref 1, Frame Relay Question, doesn't mention anything about a disabled inverse arp either.

Your advice is appreciated.

Regards,

Timothy

4 REPLIES

Re: Frame Relay Static Mapping and Inverse-ARP

inv-arp will be disabled if u use point-topoint subinterface

if u use the main interface or multipoint-point sub-interface

u need to disable it by using the command

no frame-relay invers-arp

good luck

if helpful rate

Re: Frame Relay Static Mapping and Inverse-ARP

Hy,

To see if there is Invers-Arp you can do a: show frame-relay map and see if the DLCIs are dynamic.

For example:

sh fram map

Serial0/0/0 (up): ip 129.1.124.4 dlci 204(0xCC,0x30C0), dynamic,

broadcast,, status defined, active

Always when you have dynamic learned DLCIs you have invers-arp enabled.

Even if you statically define DLCIs (frame-relay map ip 1.1.1.1 102 broadcast) invers-arp is not disabled. There is only one time when invers-arp is disabled by default: point-to-point subinterface:

int s0/0

encapsulation frame-relay

int s0/0.1 point-to-point

frame-relay interface-dlci 102

sh fram map

Serial0/0.1 (up): point-to-point dlci, dlci 102(0xC9,0x3090), broadcast

status defined, active

All the best,

Bogdan

Re: Frame Relay Static Mapping and Inverse-ARP

Hy again :),

and, of course, to disable invers-arp:

interface s0/0

encapsulation frame-relay

ip add 1.1.1.1 255.255.255.0

no frame-relay inverse-arp

So, there is a second command to see if inverse-arp is disabled or not:

sh run int s0/0

:)

All the best,

Bogdan

Cisco Employee

Re: Frame Relay Static Mapping and Inverse-ARP

Tim,

There seems to be some misunderstanding in the book you are referring to about the Inverse ARP.

Having a static mapping on a Frame Relay interface does not disable the InverseARP for the entire interface. The mapping merely causes that the router never sends InverseARP requests out the DLCI that is used in a static mapping. After all, that would be an useless thing to do, as the mapping is already defined statically. However, if an InverseARP request is received on a DLCI that is used in a static mapping, the router will send an InverseARP response without any limitation.

The point-to-point subinterfaces behave in a very similar way. They do not send InverseARP requests themselves because that does not make sense - for a point-to-point interface, everything can be reached through a single DLCI associated with that interface so there is no point in asking for mappings dynamically. However, if a point-to-point subinterface receives an InverseARP request, it happily responds with an InverseARP reply.

It is not true that in order to use static mappings, it is required to deactivate the InverseARP. You can freely mix dynamic and static mappings on an interface. Usually, though, when a static mapping is used, often the InverseARP is deactivated in order to have firm control about which DLCIs will or will not be used. Also, having a bunch of static and dynamic mappings on an interface can be confusing and difficult to debug so most often, it is best avoided.

So to summarize the main points here:

1.) A static mapping does not deactivate the InverseARP per se. It only stops sending InverseARP requests and only for statically mapped DLCI in question. It does not prevent sending InverseARP requests for other (not statically mapped) DLCIs, and it does not prevent sending responses at all.

2.) A point-to-point interface does not send InverseARP requests but it may send InverseARP responses if it receives a request itself.

3.) Static and dynamic mapping can be combined on a single physical interface or multipoint subinterface.

4.) The InverseARP is completely deactivated with all requests and responses if and only if the command "no frame-relay inverse-arp" is entered on an interface or subinterface.

723
Views
0
Helpful
4
Replies