default route over frame relay

hi all

if you have two routers 1 and 2 connected over frame relay.

the connection is fine, and you are able to ping eachother.

if R1 has a loopback interface 10.10.10.10/24

and R2 has looopback interface 20.20.20.20/24

if you setup dynamic or static route from R1 to R2 loopback interface, then you can ping it.

but if setup default route like IP Route 0.0.0.0 0.0.0.0 interfce f0/0

then it will not work.

why is that ????

Comments

  • if you setup dynamic or static route from R1 to R2 loopback interface, then you can ping it.

    but if setup default route like IP Route 0.0.0.0 0.0.0.0 interfce f0/0

    then it will not work.

    why is that ????

    Hi,

    Are you sure that the outgoing interface is fa0/0 ? In general FR interface is serial not fastethernet. Be sure that the ougoing interface to either R1 or R2 is in the default route or add the next-hop address in the default route.

    Don't hesitate to share your configuration with us.

  • sorry it is a mistake

    it is IP Route 0.0.0.0 0.0.0.0 interfce serial 0/0

     

    serial 0/0 is the interface of R1 that connected to the Frame-Relay

  • Assuming you mean s0/0 in your example, try:

    ip route 0.0.0.0 0.0.0.0 s0/0 10.1.1.1

    where 10.1.1.1 is replaced with the interface IP on the far side.

    If not, configz.

  • serial 0/0 is the interface of R1 that connected to the Frame-Relay

    So you try to configure the default route in R1 toward R2. Don't hesitate to add R2 FR ip address (the next-hop) to default route.

    For example on R1:

    ip route 0.0.0.0 0.0.0.0 serial 0/0  <ip_address_of_R2_serial0/0>

    Assumming that R2 FR ip address is on s0/0.

    Don't hesitate to share the config an topology with us.

     

  • In general there is a difference of how ARP works when using a static route point to an interface rather than valid next hop address.  When you use the valid next hop address you will ARP for that address, then rewrite l2 header and send it out.  When you use the interface name you still arp, but the difference is you are ARPing for the destination address instead.  So when you ARP, unless that IP is directly connected you are not going to resolve.  

    With that said, the nature of frame relay is multipoint.  You are most likely failing with a l2/l3 resolution which in frame multipoint is going to use either a static ip/dlci mapping, or dynamic inverse arp.  

    start by using debug ip packet detail.  If l2/l3 resolution is failing you will see encapsulation failed. 

    From there you can investigate the various l2/l3 resolution methods for your learning pleasure 

    Just so you know in an ethernet environment you have other options like static arp definitions and proxy arp.

    with your config we can can explain in more depth your specific issue.  

    Check out this support article as well - https://supportforums.cisco.com/docs/DOC-27825

    it relates to ethernet, but the idea behind point-to-point mediums vs. multipoint is generally true.  The difference in your frame-relay situation is that there is now layer2/layer3 mapping (static or dynamic) so the local device cannot encapsulate.  There is a chance you know all of this and may just have statically mapped the IP over the wrong dlci.

    r/w

    On Aug 11, 2013, at 12:29 PM, oudmaster <[email protected]> wrote:

    sorry it is a mistake

    it is IP Route 0.0.0.0 0.0.0.0 interfce serial 0/0

     

    serial 0/0 is the interface of R1 that connected to the Frame-Relay




    INE - The Industry Leader in CCIE Preparation
    http://www.INE.com

    Subscription information may be found at:
    http://www.ieoc.com/forums/ForumSubscriptions.aspx

  • Hi,

    In frame-relay multipoint network packet is forwarded using route information as mentioned below:

    1. Route recursion and next hop identification.

    2. Frame-relay layer3 to Layer 2 resolution for the particular next hop.

    3. Layer 2 encapsulation is done according to DLCI value.

    If it fails to find the DLCI mapping for the next hop, you will get "encapsulation failed" error.

    R1#ping 2.2.2.2 repeat 1

     

    Type escape sequence to abort.

    Sending 1, 100-byte ICMP Echos to 2.2.2.2, timeout is 2 seconds:

     

    *Mar  1 00:23:11.179: IP: tableid=0, s=10.1.12.1 (local), d=2.2.2.2 (Serial0/0), routed via RIB

    *Mar  1 00:23:11.179: IP: s=10.1.12.1 (local), d=2.2.2.2 (Serial0/0), len 100, sending

    *Mar  1 00:23:11.183: IP: s=10.1.12.1 (local), d=2.2.2.2 (Serial0/0), len 100, encapsulation failed.

    Success rate is 0 percent (0/1)

     

    Since this is a multipoint network, its obvious that there can be multiple possible destinations in remote location. So, it can't route the packet just having outgoing information without knowing its corresponding DLCI value for the next hop.

    If you were configuring such a scenario using point-to-point subinterface, it would be able to send traffic as you expected with your default route. Because point-to-point network has only one possible destination and layer3 to layer 2 resolution is not required for the particular next hop. Once traffic leaves local router, it should reach its destination even if it doesn't have DLCI information for the next hop.

    Refer to the following configuration:

    R1#sh ip route st

    S*   0.0.0.0/0 is directly connected, Serial0/0.12

     

    R1#sh frame-relay map

    Serial0/0.12 (up): point-to-point dlci, dlci 102(0x66,0x1860), broadcast

              status defined, active

     

    R1#sh  ip route st

    S*   0.0.0.0/0 is directly connected, Serial0/0.12

     

     

    R2#sh frame-relay map

    Serial0/1.12 (up): point-to-point dlci, dlci 201(0xC9,0x3090), broadcast

              status defined, active

     

    R1#ping 2.2.2.2 source l0

     

    Type escape sequence to abort.

    Sending 5, 100-byte ICMP Echos to 2.2.2.2, timeout is 2 seconds:

    Packet sent with a source address of 1.1.1.1

    !!!!!

    Success rate is 100 percent (5/5), round-trip min/avg/max = 24/40/68 ms

     

    Hope this helps!

    Good luck!

     

     

     

  • Here is how you can learn this by doing.

    1. Configure the static route pointing to the interface (not IP) on an Ethernet segment and then ping the loopback of the other router.  Then look at the ARP table to see how the loopback was reached and in turn what layer 2 address was used.  Bonus: clear ip arp <loopback ip> on the local router and then disable proxy ARP on the remote router and try to ping again.
    2. Configure the same static route pointing to a Frame Relay point-to-point sub interface and then ping the loopback of the other router.  Then look at the Frame Relay mapping (show frame map) to the remote router and see what layer 2 address was used. Debug frame packet will show you the layer 2 address that was used.
    3. Finally configure the same static route pointing to a Frame Relay multipoint sub interface (or physical), turn on debug frame packet and then try to ping the loopback of the other router.  Why did it fail?  Bonus:  Put a Frame Relay map using the IP address of the remote loopback and the DLCI of the remote router and try to ping.

    -- 
    Brian Dennis, CCIEx5 #2210 (R&S/ISP-Dial/Security/SP/Voice)

    INE, Inc.

    From: oudmaster <[email protected]>
    Reply-To: "[email protected]" <[email protected]>
    Date: Sunday, August 11, 2013 9:54 AM
    To: Brian Dennis <[email protected]>
    Subject: [CCIE R&S General] default route over frame relay

    hi all

    if you have two routers 1 and 2 connected over frame relay.

    the connection is fine, and you are able to ping eachother.

    if R1 has a loopback interface 10.10.10.10/24

    and R2 has looopback interface 20.20.20.20/24

    if you setup dynamic or static route from R1 to R2 loopback interface, then you can ping it.

    but if setup default route like IP Route 0.0.0.0 0.0.0.0 interfce f0/0

    then it will not work.

    why is that ????




    INE - The Industry Leader in CCIE Preparation

    http://www.INE.com


    Subscription information may be found at:

    http://www.ieoc.com/forums/ForumSubscriptions.aspx
  • I would also like to add:

    if you set the correct frame relay map up only in one direction, your pings will be successful as long as you are not sourcing from a specific interface, this is because the device on the other end sees the source as the frame relay interface address.  If you want to be able to go from loopback 0 to loopback 0 you are going to need the frame-map for opposing ends loopback0 interface.

    Try it one way, then on both sides.  run the debug ip packet detail as well, so you can see the s/d and encapsulation messages too.


    See this sample config:


    network 10.10.52.x/24

    dlci 502                   dlci 205
    R5--------------FR---------------R2
    lo0                             lo0
    5.5.5.5/32               2.2.2.2/32





    R5#sho run int s0/1/0
    Building configuration...

    Current configuration : 125 bytes
    !
    interface Serial0/1/0
     ip address 10.10.52.5 255.255.255.0
     encapsulation frame-relay
     frame-relay map ip 2.2.2.2 502
    end



    R2#sho run int s0/1/0
    Building configuration...

    Current configuration : 125 bytes
    !
    interface Serial0/1/0
     ip address 10.10.52.2 255.255.255.0
     encapsulation frame-relay
     frame-relay map ip 5.5.5.5 205
    end


    On Aug 11, 2013, at 1:02 PM, wayne <[email protected]> wrote:
    In general there is a difference of how ARP works when using a static route point to an interface rather than valid next hop address.  When you use the valid next hop address you will ARP for that address, then rewrite l2 header and send it out.  When you use the interface name you still arp, but the difference is you are ARPing for the destination address instead.  So when you ARP, unless that IP is directly connected you are not going to resolve.  

    With that said, the nature of frame relay is multipoint.  You are most likely failing with a l2/l3 resolution which in frame multipoint is going to use either a static ip/dlci mapping, or dynamic inverse arp.  

    start by using debug ip packet detail.  If l2/l3 resolution is failing you will see encapsulation failed. 

    From there you can investigate the various l2/l3 resolution methods for your learning pleasure 

    Just so you know in an ethernet environment you have other options like static arp definitions and proxy arp.

    with your config we can can explain in more depth your specific issue.  

    Check out this support article as well - https://supportforums.cisco.com/docs/DOC-27825

    it relates to ethernet, but the idea behind point-to-point mediums vs. multipoint is generally true.  The difference in your frame-relay situation is that there is now layer2/layer3 mapping (static or dynamic) so the local device cannot encapsulate.  There is a chance you know all of this and may just have statically mapped the IP over the wrong dlci.

    r/w

    On Aug 11, 2013, at 12:29 PM, oudmaster <[email protected]> wrote:

    sorry it is a mistake

    it is IP Route 0.0.0.0 0.0.0.0 interfce serial 0/0

     

    serial 0/0 is the interface of R1 that connected to the Frame-Relay




    INE - The Industry Leader in CCIE Preparation
    http://www.INE.com

    Subscription information may be found at:
    http://www.ieoc.com/forums/ForumSubscriptions.aspx




    INE - The Industry Leader in CCIE Preparation
    http://www.INE.com

    Subscription information may be found at:
    http://www.ieoc.com/forums/ForumSubscriptions.aspx

  • Hi oudmaster,

    Use of an exit interface rather the the next hop address when configuring a default route will result in an ARP request for each destination. A ping routed through a default route cnofigured with an exit interface will FAIL if the connected router has ip proxy arp disabled; Expected condition for security reasons. So correcting this will enable the echo request to get to the next hop router.

    If more than two routers are involved, each would have to have dynamic routing configured or have an appropriate static route configured to enable the echo request to get to its intended destination. The reverse is also true. The echo reply cannot be sent back to the source of the echo request unless dynamic routing or an appropriate static route is configured on each router in the transit path.

    Finally, in an Ethernet environment, the return path could be totally difference than the path to the destination.

    HTH,

    Ernie

  • if you have two routers 1 and 2 connected over frame relay.

    the connection is fine, and you are able to ping eachother.

    if R1 has a loopback interface 10.10.10.10/24

    and R2 has looopback interface 20.20.20.20/24

    if you setup dynamic or static route from R1 to R2 loopback interface, then you can ping it.

    but if setup default route like IP Route 0.0.0.0 0.0.0.0 interfce f0/0

    then it will not work.

    why is that ????

    May be you are configuring the exit interface which doesn't have IP address i.e you have configured the sub-interface for Frame-relay and you are using main interface for static route.

    R1#show ip int brief
    Interface                  IP-Address      OK? Method Status                Protocol
    Serial1/0                  unassigned      YES TFTP   up                    up 
    Serial1/0.1                12.12.12.1      YES manual up                    up 
    Loopback0                  1.1.1.1         YES manual up                    up 

    In this case I can reach to remote router but if  I configure static route with serial1/0 exit interface packet can;t reach to destination becuase there is no proper route installed in routing table - reason is we are not using interface has L3 address.

    R1(config)#ip route 0.0.0.0 0.0.0.0 serial1/0 12.12.12.2
    R1(config)#do show ip route static

    R1(config)#do debug ip packet detail
    IP packet debugging is on (detailed)
    R1(config)#do ping 3.3.3.3

    Type escape sequence to abort.
    Sending 5, 100-byte ICMP Echos to 3.3.3.3, timeout is 2 seconds:

    *Mar  1 00:09:42.299: IP: s=1.1.1.1 (local), d=3.3.3.3, len 100, unroutable
    *Mar  1 00:09:42.303:     ICMP type=8, code=0.
    *Mar  1 00:09:44.299: IP: s=1.1.1.1 (local), d=3.3.3.3, len 100, unroutable
    *Mar  1 00:09:44.299:     ICMP type=8, code=0.
    *Mar  1 00:09:46.299: IP: s=1.1.1.1 (local), d=3.3.3.3, len 100, unroutable
    *Mar  1 00:09:46.303:     ICMP type=8, code=0.

    *Mar  1 00:09:48.299: IP: s=1.1.1.1 (local), d=3.3.3.3, len 100, unroutable
    *Mar  1 00:09:48.299:     ICMP type=8, code=0.
    *Mar  1 00:09:50.299: IP: s=1.1.1.1 (local), d=3.3.3.3, len 100, unroutable
    *Mar  1 00:09:50.299:     ICMP type=8, code=0.
    Success rate is 0 percent (0/5)

    So here you must use the subinterface:

    R1(config)#no ip route 0.0.0.0 0.0.0.0 serial1/0 12.12.12.2
    R1(config)#ip route 0.0.0.0 0.0.0.0 serial1/0.1 12.12.12.2
    R1(config)#do ping 3.3.3.3

    Type escape sequence to abort.
    Sending 5, 100-byte ICMP Echos to 3.3.3.3, timeout is 2 seconds:
    !!!!!
    Success rate is 100 percent (5/5), round-trip min/avg/max = 32/66/132 ms
    R1(config)#

    Good luck with your studies!

Sign In or Register to comment.