Skip to main content

10. ONT tests

10.0 Pre-checks​

Before running the tests from ont1 and pc1, identify the currently active IPv6 delegated prefix on wan1.

Quick option using docker exec from the host:

docker exec ont1 sh -lc 'ip -6 addr show dev eth1.150; echo; ip -6 addr show dev eth2'
docker exec pc1 sh -lc 'ip -6 addr show dev eth1; echo; ip -6 route show dev eth1'

What to validate:

  • eth1.150 on ont1 has a WAN IPv6 /128
  • eth2 on ont1 has a LAN address from the current delegated prefix, for example 2001:db8:200:3::1/64
  • pc1 should receive that same prefix on eth1

Important:

  • do not assume the delegated prefix will always be 2001:db8:200:1::/64
  • after a reconnect or lease change, pc1 may temporarily keep a previous prefix in addition to the current one
  • if pc1 shows more than one global prefix on eth1, use the prefix currently configured on ont1 eth2 as the source of truth

Connectivity from ONT1​

Thanks to the fact that the ONT image has PBR (Policy Based Routing), it is possible to ping each of the DHCP gateways corresponding to its services:

Ping IPv4 CGNAT Gateway (Dual-Stack WAN2)​

Option using docker exec from the host:

docker exec ont1 ping -c 4 100.80.0.1
user@ont1  ~  ping 100.80.0.1
PING 100.80.0.1 (100.80.0.1) 56(84) bytes of data.
64 bytes from 100.80.0.1: icmp_seq=1 ttl=64 time=2.14 ms
64 bytes from 100.80.0.1: icmp_seq=2 ttl=64 time=1.62 ms
64 bytes from 100.80.0.1: icmp_seq=3 ttl=64 time=1.81 ms
64 bytes from 100.80.0.1: icmp_seq=4 ttl=64 time=1.43 ms
^C
--- 100.80.0.1 ping statistics ---
4 packets transmitted, 4 received, 0% packet loss

Ping IPv4 VIP Gateway (WAN3)​

Option using docker exec from the host:

docker exec ont1 ping -c 4 192.168.5.1
user@ont1  ~  ping 192.168.5.1
PING 192.168.5.1 (192.168.5.1) 56(84) bytes of data.
64 bytes from 192.168.5.1: icmp_seq=1 ttl=64 time=2.33 ms
64 bytes from 192.168.5.1: icmp_seq=2 ttl=64 time=1.81 ms
^C
--- 192.168.5.1 ping statistics ---
4 packets transmitted, 4 received, 0% packet loss

IPv6 WAN Ping (IPv6-only WAN1)​

Option using docker exec from the host:

docker exec ont1 ip -6 -o addr show dev eth1.150 scope global
docker exec ont1 ping -6 -c 4 -I <CURRENT_WAN1_IPV6> 2001:db8:aaaa::2

Expected result:

  • ont1 reaches 2001:db8:aaaa::2 using the current WAN IPv6 on eth1.150 as source

10.1 Check Prefix Delegation on PC1​

On PC1 you can see the delegated prefix assigned via DHCPv6-PD.

Option using docker exec from the host:

docker exec ont1 ip -6 addr show dev eth2
docker exec pc1 ip -6 addr show dev eth1

What to validate:

  • ont1 advertises the LAN gateway for the current delegated prefix on eth2, for example 2001:db8:200:3::1/64
  • pc1 receives a global address from that same prefix on eth1

Example:

  • if ont1 shows 2001:db8:200:3::1/64 on eth2
  • then pc1 should show an address like 2001:db8:200:3:xxxx:xxxx:xxxx:xxxx/64

This prefix is routed to ont1 by the BNG, and ont1 advertises it to pc1 on the LAN through Router Advertisement.

Note:

  • if pc1 keeps a previous global prefix in addition to the current one, that is not necessarily a service failure
  • it may be a stale prefix temporarily kept due to RA lifetime
  • in that case, use the current prefix configured on ont1 as the reference for the reachability test
  • if a clean validation is required, restart pc1 before continuing:
docker restart pc1

10.2 Check Reachability from PC1​

Once the current delegated prefix has been identified on ont1, pc1 must be able to reach:

  • the current IPv6 LAN gateway on ont1
  • the internal lab DNS server

IPv6 LAN Gateway​

Option using docker exec from the host:

docker exec ont1 ip -6 -o addr show dev eth2 scope global
docker exec pc1 ping -6 -c 4 <CURRENT_LAN_GW_ON_ONT1>

Example:

docker exec pc1 ping -6 -c 4 2001:db8:200:3::1

Expected result:

  • pc1 reaches the current IPv6 LAN gateway on ont1

Internal Lab DNS​

Option using docker exec from the host:

docker exec pc1 ping -6 -c 4 2001:db8:aaaa::2
[*]─[pc1]─[~]
└──> ping -6 2001:db8:aaaa::2
PING 2001:db8:aaaa::2(2001:db8:aaaa::2) 56 data bytes
64 bytes from 2001:db8:aaaa::2: icmp_seq=1 ttl=61 time=28.1 ms
64 bytes from 2001:db8:aaaa::2: icmp_seq=2 ttl=61 time=1.18 ms
64 bytes from 2001:db8:aaaa::2: icmp_seq=3 ttl=61 time=1.42 ms
64 bytes from 2001:db8:aaaa::2: icmp_seq=4 ttl=61 time=3.30 ms

--- 2001:db8:aaaa::2 ping statistics ---
4 packets transmitted, 4 received, 0% packet loss

Expected result:

  • ont1 has an active WAN IPv6 on eth1.150
  • ont1 has an active LAN IPv6 gateway on eth2 derived from the current delegated prefix
  • pc1 receives the current delegated prefix on eth1
  • pc1 reaches the current LAN gateway on ont1
  • PC1 reaches the internal lab DNS at 2001:db8:aaaa::2

Operational note:

  • if pc1 shows more than one global IPv6 prefix, validation must be done against the prefix currently present on ont1
  • an old prefix still visible on pc1 may persist for some time due to the previous RA lifetime and does not necessarily mean a BNG or ONT failure