Can ping from CMTS, but not from individual cable interface | docsis.org

You are here

Can ping from CMTS, but not from individual cable interface

6 posts / 0 new
Last post
networkvisionary
Can ping from CMTS, but not from individual cable interface

Hi guys. Thought maybe you all might have some insight into this one. I can ping an address from the CMTS itself, and from its loopback. But I cannot ping from the individual cable interface. Any ideas about the root of the problem here? I'm not understanding how the cable interface looks at routing differently than the rest of the CMTS. The problem is not from being blocked upstream, as I have another CMTS that can reach the same address, with the same source specified.

hinzoo
Show run please.

Is Cable x/x bundled?

networkvisionary
applicable configs

Here is how this portion of the config looks. I'm not sure where else I should be looking for why the "source-ip" addition to the ping command changes the E6000 to ignoring default gateway. There is a default gateway configured. All other traffic exits the device using the default gateway. So why would changing the source-ip impact the routing decision to use the default route?

E6000-CMTS# show ipv6 route xxxx:xxxx:xxx:2089::103

Dist/
IPv6 Route Dest / mask Act Next Hop Metric Protocol RouteAge Interface
====================== === ======================================= ========= ========= ======== ==============
::/0 ACT xxxx:xxxx:xxxx:1::61 1/0 netmgmt 25:25:10 ether 6/0.0

E6000-CMTS# show running-config interface cable-mac 1.0

# show running-config interface cable-mac 1
configure
interface cable-mac 1
cable insertion-interval 10
cable cm-ip-prov-mode ipv4only
cable mcast-fwd-by-dsid
cable verbose-cm-rcp
cable dynamic-rcc
cable downstream-bonding-group dynamic enable
cable upstream-bonding-group dynamic enable
cable mult-tx-chl-mode
cable cm-status event-type all max-event-holdoff 32000

interface cable-mac 1.0
ip address 10.155.198.49 255.255.255.252
ipv6 enable
ip address 10.16.4.1 255.255.255.192 secondary
ip address 10.16.6.1 255.255.255.192 secondary
ip address 10.16.13.1 255.255.255.0 secondary
ip address 10.132.1.1 255.255.255.128 secondary
ip address 10.132.1.129 255.255.255.128 secondary
ip address 10.155.208.161 255.255.255.224 secondary
ip address 10.155.215.65 255.255.255.192 secondary
ipv6 address xxxx:xxxx:xxxx:x::1/64
ipv6 address xxxx:xxxx:xxxx:x::1/64
cable helper-address xxx.xx.xxx.xxx
cable helper-address xxx.xx.x.xxx
ip igmp
ip igmp version 3
ipv6 dhcp relay destination xxx:xxxx:xxx:2089::103
ipv6 dhcp relay destination xxxx:xxxx:xxx:3add::a
ipv6 nd dad attempts 0
ipv6 nd managed-config-flag
ipv6 nd other-config-flag
ipv6 nd ra interval 60 20
exit
interface cable-mac 1 cable bundle master
interface cable-mac 2 cable bundle cable-mac 1
interface cable-mac 3 cable bundle cable-mac 1
interface cable-mac 4 cable bundle cable-mac 1
interface cable-mac 1 no shutdown
interface cable-mac 1.0
ip pim sparse-mode-ssm passive
ip pim dr-priority 0
ip pim hello-holdtime 65535
exit
interface cable-mac 1
cable privacy mandatory bpi-plus
exit
end
E6000-CMTS# ping xxx:xxxx:xxx:2089::103
ping ( xxx:xxxx:xxx:2089::103): 100 data bytes
!!!!!
5 packets transmitted, 5 packets received
round-trip min/avg/max = 10/13/29 ms

traceroute: max hops reached prior to reaching destination

E6000-CMTS# ping xxx:xxxx:xxx:2089::103 source-ip xxxx:xxxx:xxxx:3::1
ping (xxxx:xxxx:xxxx:3::1): 100 data bytes
.....
5 packets transmitted, 0 packets received

E6000-CMTS# traceroute ipv6 xxx:xxxx:xxx:2089::103 source xxxx:xxxx:xxxx:3::1
1: * * *
2: * * *
3: * * *

networkvisionary
yes, it is set up a cable-mac

yes, it is set up as cable-mac 1. with cable mac 1.0 over/under it. interesting point. on my other chassis (C4).. there is not bundling configured. this works on the C4.. but not the E6000 (which has bundling).. so maybe I'm missing a line of config on the bundle.

the show run for the cable-mac

configure
interface cable-mac 1
description "Environtment3"
cable insertion-interval 10
cable cm-ip-prov-mode ipv4only
cable mcast-fwd-by-dsid
cable verbose-cm-rcp
cable dynamic-rcc
cable downstream-bonding-group dynamic enable
cable upstream-bonding-group dynamic enable
cable mult-tx-chl-mode
cable cm-status event-type all max-event-holdoff 32000
ip igmp static-group 239.223.47.49 source 10.155.207.101 # layer 2
exit
interface cable-mac 1.0
ip address 10.155.198.49 255.255.255.252
ipv6 enable
ip address 10.16.4.1 255.255.255.192 secondary
ip address 10.16.6.1 255.255.255.192 secondary
ip address 10.16.13.1 255.255.255.0 secondary
ip address 10.132.1.1 255.255.255.128 secondary
ip address 10.132.1.129 255.255.255.128 secondary
ip address 10.155.208.161 255.255.255.224 secondary
ip address 10.155.215.65 255.255.255.192 secondary
ipv6 address xxxx:xxxx:210d:3::1/64
ipv6 address xxxx:xxxx:210d:4::1/64
cable helper-address xxx.24.108.103
cable helper-address xxx.14.7.151
ip igmp
ip igmp version 3
ipv6 dhcp relay destination xxxx:xxxx:202:2089::103
ipv6 dhcp relay destination xxxx:xxxx:604:3add::a
ipv6 nd dad attempts 0
ipv6 nd managed-config-flag
ipv6 nd other-config-flag
ipv6 nd ra interval 60 20
exit
interface cable-mac 1 cable bundle master
interface cable-mac 2 cable bundle cable-mac 1
interface cable-mac 3 cable bundle cable-mac 1
interface cable-mac 4 cable bundle cable-mac 1
interface cable-mac 1 no shutdown
interface cable-mac 1.0
ip pim sparse-mode-ssm passive
ip pim dr-priority 0
ip pim hello-holdtime 65535
exit
interface cable-mac 1
cable privacy mandatory bpi-plus
exit
end

networkvisionary
looks like this

here is what I'm seeing

E6000-CMTS# ping xxxx:xxxx:60e:1:acd:0:547:1
ping (xxxx:xxxx:60e:1:acd:0:547:1): 100 data bytes
!!!!!
5 packets transmitted, 4 packets received
round-trip min/avg/max = 10/10/10 ms
E6000-CMTS# traceroute ipv6 xxxx:xxxx:60e:1:acd:0:547:1 source xxxx:xxxx:210d:4::1
1: * * *
2: * * *
3: * * *
4: * * *
5: * * *
6: * * *
7: * * *
8: * * *
9: * * *
10: * * *

networkvisionary
Duh!

Problem was in a router several devices upstream. Router somehow got into its head that the route for all my ipv6 addresses in the local network was to Null0. Since that router had a route to a black hole.. it spread the word via ospf. since E6000 is getting updates via ospf as well.. it had the idea that the route to the local ipv6 address was into a black hole. so it made no attempt to forward any traffic with an address in my local ipv6 scope. which is why I could get out with just the dhcp relay address.. since it is not in the local scope. but attaching the source of the local range caused the traffic to just get dropped on its way out.

Log in or register to post comments