[c-nsp] EoMPLS VC on Cisco 7609 is up, but with very small MTU (42 bytes)

Eric Xu ericxuweili at gmail.com
Wed May 18 22:23:32 EDT 2016


Hi George,
    The maximum MTU on SVI is 1500.

Hi Saku,
    Thanks for your suggestion. Definitely I will look into ELAM.

    Cheers.

Eric


On Thu, May 19, 2016 at 1:16 AM, George Giannousopoulos <ggiannou at gmail.com>
wrote:

> Hi,
>
> Have you tried setting the MTU on the SVI to something more than the
> default 1500?
> Although 42 bytes isn't normal, you should anyway raise the MTU on each
> SVI in order to reach 9000.
>
> --
> George
>
> On Wed, May 18, 2016 at 5:00 PM, Saku Ytti <saku at ytti.fi> wrote:
>
>> Hey Eric,
>>
>>
>> I'd capture the frames and see what is happening.
>>
>> 42B does not sound like completely arbitrary value. 46B is minimum
>> payload of L2, and you're adding one label, 4B to it, ending up with
>> exactly 46B of payload, which is minimum possible payload on ethernet.
>> Which implies perhaps somewhere MTU is configured below minimum, perhaps
>> 0.
>>
>> ELAM capture for lookup results at 42B and 43B should be interesting,
>> perhaps the lookup results are different (they should not be).
>>
>> You should check what the platform thinks MTU is, by looking at MLS
>> CEF forwarding, sometimes next-hop MTU is incorrectly programmed, not
>> same as in CLI. This can usually be fixed by manually changing MTU in
>> CLI to something arbitrary and then back to what you actually want,
>> which forces reprogramming of the HW adjacency MTU.
>>
>> The MTU you see in LDP signalling is completely meaningless, it's just
>> control-plane signalling, it has nothing to do with forwarding.
>>
>> You also can't exclude something very odd happening, on ES+ had once
>> bug on EoMPLS where it would change DADDR in payload by 1 every 5
>> packets on a /24 border.
>>
>>
>> On 18 May 2016 at 16:37, Eric Xu <ericxuweili at gmail.com> wrote:
>> > I have two directly connected Cisco 7609 routers running MPLS with OSPF
>> > being IGP,  and a EoMPLS VC has been created.
>> >
>> > On router 1:
>> >
>> > interface Vlan100
>> > ip address 192.168.168.2 255.255.255.0
>> > xconnect 10.200.0.158 100 encapsulation mpls
>> >
>> > On router 2:
>> >
>> > interface Vlan100
>> > ip address 192.168.168.3 255.255.255.0
>> > xconnect 10.200.0.157 100 encapsulation mpls
>> >
>> > *VC is up on both routers with MTU 1500:*
>> >
>> > *router1#sh mpls l2transport vc 100 detail*
>> > *Local interface: Vl100 up, line protocol up, Eth VLAN 100 up*
>> > *Interworking type is Ethernet*
>> > *Destination address: 10.200.0.158, VC ID: 100, VC status: up*
>> > *Output interface: Te1/0/0.18000064, imposed label stack {16}*
>> > *Preferred path: not configured*
>> > *Default path: active*
>> > *Next hop: 10.200.0.38*
>> > *Create time: 00:12:14, last status change time: 00:12:14*
>> > *Last label FSM state change time: 00:12:14*
>> > *Signaling protocol: LDP, peer 10.200.0.158:0 <http://10.200.0.158:0>
>> up*
>> > *Targeted Hello: 10.200.0.157(LDP Id) -> 10.200.0.158, LDP is UP*
>> > *Status TLV support (local/remote) : enabled/supported*
>> > *LDP route watch : enabled*
>> > *Label/status state machine : established, LruRru*
>> > *Last local dataplane status rcvd: No fault*
>> > *Last BFD dataplane status rcvd: Not sent*
>> > *Last BFD peer monitor status rcvd: No fault*
>> > *Last local AC circuit status rcvd: No fault*
>> > *Last local AC circuit status sent: No fault*
>> > *Last local PW i/f circ status rcvd: No fault*
>> > *Last local LDP TLV status sent: No fault*
>> > *Last remote LDP TLV status rcvd: No fault*
>> > *Last remote LDP ADJ status rcvd: No fault*
>> > *MPLS VC labels: local 214, remote 16*
>> > *Group ID: local 0, remote 0*
>> > *MTU: local 1500, remote 1500*
>> > *Remote interface description:*
>> > *Sequencing: receive disabled, send disabled*
>> > *Control Word: On (configured: autosense)*
>> >
>> > *router2#sh mpls l2transport vc 100 detail*
>> > *Local interface: Vl100 up, line protocol up, Eth VLAN 100 up*
>> > *Interworking type is Ethernet*
>> > *Destination address: 10.200.0.157, VC ID: 100, VC status: up*
>> > *Output interface: Te2/11.18000064, imposed label stack {214}*
>> > *Preferred path: not configured*
>> > *Default path: active*
>> > *Next hop: 10.200.0.37*
>> > *Create time: 00:13:06, last status change time: 00:12:53*
>> > *Last label FSM state change time: 00:12:53*
>> > *Signaling protocol: LDP, peer 10.200.0.157:0 <http://10.200.0.157:0>
>> up*
>> > *Targeted Hello: 10.200.0.158(LDP Id) -> 10.200.0.157, LDP is UP*
>> > *Status TLV support (local/remote) : enabled/supported*
>> > *LDP route watch : enabled*
>> > *Label/status state machine : established, LruRru*
>> > *Last local dataplane status rcvd: No fault*
>> > *Last BFD dataplane status rcvd: Not sent*
>> > *Last BFD peer monitor status rcvd: No fault*
>> > *Last local AC circuit status rcvd: No fault*
>> > *Last local AC circuit status sent: No fault*
>> > *Last local PW i/f circ status rcvd: No fault*
>> > *Last local LDP TLV status sent: No fault*
>> > *Last remote LDP TLV status rcvd: No fault*
>> > *Last remote LDP ADJ status rcvd: No fault*
>> > *MPLS VC labels: local 16, remote 214*
>> > *Group ID: local 0, remote 0*
>> > *MTU: local 1500, remote 1500*
>> > *Remote interface description:*
>> > *Sequencing: receive disabled, send disabled*
>> > *Control Word: On (configured: autosense)*
>> > *SSO Descriptor: 10.200.0.157/100 <http://10.200.0.157/100>, local
>> label:
>> > 16*
>> > *Dataplane:*
>> > *SSM segment/switch IDs: 4144/4142 (used), PWID: 9*
>> > *VC statistics:*
>> > *transit packet totals: receive 32, send 15*
>> > *transit byte totals: receive 2318, send 1020*
>> > *transit packet drops: receive 0, seq error 0, send 0*
>> >
>> > *MTU between the routers are more than 9000:*
>> >
>> > *router1#ping 10.200.0.158 size 9000 df-bit source l0*
>> > *Type escape sequence to abort.*
>> > *Sending 5, 9000-byte ICMP Echos to 10.200.0.158, timeout is 2 seconds:*
>> > *Packet sent with a source address of 10.200.0.157*
>> > *Packet sent with the DF bit set*
>> > *!!!!!*
>> > *Success rate is 100 percent (5/5), round-trip min/avg/max = 12/13/16
>> ms*
>> >
>> > *router2#ping 10.200.0.157 size 9000 df-bit source l0*
>> > *Type escape sequence to abort.*
>> > *Sending 5, 9000-byte ICMP Echos to 10.200.0.157, timeout is 2 seconds:*
>> > *Packet sent with a source address of 10.200.0.158*
>> > *Packet sent with the DF bit set*
>> > *!!!!!*
>> > *Success rate is 100 percent (5/5), round-trip min/avg/max = 12/20/44
>> ms*
>> >
>> > *However, MTU on the VC is only 42 bytes, which makes it unusable:*
>> >
>> > *router1#ping 192.168.168.3 size 42 df-bit source vlan 100*
>> > *Type escape sequence to abort.*
>> > *Sending 5, 42-byte ICMP Echos to 192.168.168.3, timeout is 2 seconds:*
>> > *Packet sent with a source address of 192.168.168.2*
>> > *Packet sent with the DF bit set*
>> > *!!!!!*
>> > *Success rate is 100 percent (5/5), round-trip min/avg/max = 8/9/12 ms*
>> >
>> > *router1#ping 192.168.168.3 size 43 df-bit source vlan 100*
>> > *Type escape sequence to abort.*
>> > *Sending 5, 43-byte ICMP Echos to 192.168.168.3, timeout is 2 seconds:*
>> > *Packet sent with a source address of 192.168.168.2*
>> > *Packet sent with the DF bit set*
>> > *....!*
>> > *Success rate is 20 percent (1/5), round-trip min/avg/max = 8/8/8 ms*
>> >
>> > *router1#ping 192.168.168.3 size 44 df-bit source vlan 100*
>> > *Type escape sequence to abort.*
>> > *Sending 5, 44-byte ICMP Echos to 192.168.168.3, timeout is 2 seconds:*
>> > *Packet sent with a source address of 192.168.168.2*
>> > *Packet sent with the DF bit set*
>> > *.....*
>> > *Success rate is 0 percent (0/5)*
>> >
>> >
>> > ESM20G 7600-ES20-10G3C card on router 1 and 7600 ES+ 76-ES+XC-20G3C
>> card on
>> > router 2.
>> >
>> >
>> > Any ideas why?
>> >
>> > Thank you.
>> > _______________________________________________
>> > cisco-nsp mailing list  cisco-nsp at puck.nether.net
>> > https://puck.nether.net/mailman/listinfo/cisco-nsp
>> > archive at http://puck.nether.net/pipermail/cisco-nsp/
>>
>>
>>
>> --
>>   ++ytti
>> _______________________________________________
>> cisco-nsp mailing list  cisco-nsp at puck.nether.net
>> https://puck.nether.net/mailman/listinfo/cisco-nsp
>> archive at http://puck.nether.net/pipermail/cisco-nsp/
>>
>
>


More information about the cisco-nsp mailing list