Good morning,

after switching over to the following filter, as mentioned in the last
mails:

filter static_and_bgp_and_ospf {
  if(source = RTS_STATIC || source = RTS_BGP || source = RTS_OSPF) then accept;
  reject;
}

I still see routes that are missing in the kernel, even though they are
coming in via OSPF and BGP:

--------------------------------------------------------------------------------
bird> show route all for 2a0a:e5c0:2:a::b
Table master6:
2a0a:e5c0:2:a::b/128 unicast [ospf6 2024-02-26] * E2 (150/10/10000) 
[147.78.194.129]
        via fe80::3eec:efff:fecb:d81a on eth0
        Type: OSPF-E2 univ
        OSPF.metric1: 10
        OSPF.metric2: 10000
        OSPF.tag: 0x00000000
        OSPF.router_id: 147.78.194.129
                     unicast [router1_place10_v6_v4 2024-02-26 from 
2a0a:e5c0:1:8::50] (100/10) [i]
        via fe80::21b:21ff:febc:bf36 on eth1
        Type: BGP univ
        BGP.origin: IGP
        BGP.as_path:
        BGP.next_hop: 2a0a:e5c0:1:8::50 fe80::21b:21ff:febc:bf36
        BGP.local_pref: 500
bird>
bird> show protocols all kernel_v6
Name       Proto      Table      State  Since         Info
kernel_v6  Kernel     master6    up     2024-02-05
  Channel ipv6
    State:          UP
    Table:          master6
    Preference:     10
    Input filter:   ACCEPT
    Output filter:  static_and_bgp_and_ospf
    Routes:         0 imported, 199204 exported, 0 preferred
    Route change stats:     received   rejected   filtered    ignored   accepted
      Import updates:              0          0          0          0          0
      Import withdraws:            0          0        ---          0          0
      Export updates:       11127835          0       7001        ---   11120834
      Export withdraws:      1131164        ---        ---        ---    1133092

bird> show route count filter static_and_bgp_and_ospf
2802064 of 2802068 routes for 937748 networks in table master4
583651 of 583679 routes for 199200 networks in table master6
Total: 3385715 of 3385747 routes for 1136948 networks in 2 tables
bird>

--------------------------------------------------------------------------------
(BGP can be ignored, the OSPF route is relevant)

Checking on the OS side, I see the local IPv6 address 2a0a:e5c0:2:a::a
on lo, but not the remote IPv6 address 2a0a:e5c0:2:a::b that should be added:

--------------------------------------------------------------------------------
[08:08] server122.place10:~# ip -6 r | grep 2a0a:e5c0:2:a
2a0a:e5c0:2:a::a dev lo proto kernel metric 256 pref medium
[08:08] server122.place10:~# ip route get 2a0a:e5c0:2:a::b
RTNETLINK answers: Host is unreachable
[08:17] server122.place10:~#
--------------------------------------------------------------------------------

What am I doing wrong here?

And is there actually a CLI syntax for
verifying that a route is accepted by a filter? In case of this
particular filter there are a lot of matches, so displaying them and
checking is not really feasible.

Greetings from Seoul,

Nico

--
Sustainable and modern Infrastructures by ungleich.ch

Reply via email to