Re: [qpimd-users] No next hop in RIB with Quagga 1.1.0

[Donald Sharp]

Joerg -

Do you have 'debug pim zebra' on?  I'd like to see output from that
debug.  Additionally what is the output of 'show ip rpf 10.55.0.132'?

donald

On Mon, Jan 23, 2017 at 10:59 AM,  <address@hidden> wrote:
> Hi.
>
> I hope this mailinglist is still active.
>
> I am using Zebra, eBGP, Pimd version 0.99.25-dev so far and everything is
> working.
> eBGP is getting the routes from the neighbors over physical and logical
> interfaces (GRE).
> Zebra is writing these routes into the kernel and pimd is getting the next
> hop via pim_zlookup.
>
> Example:
> 2017/01/23 10:48:59 PIM: pim_zlookup.c zclient_lookup_nexthop: lookup=0/3:
> zebra returned recursive nexthop 172.26.79.4 for address 10.6.7.25 dist=20
> met=0
>
> I tried to upgrade to version 1.1.0 at the weekend.
>
> The routes are there, also mroutes. "show ip rib XXXX" is showing the right
> outgoing interface, but the next hop is always 0.0.0.0
>
> I do not see any pim_zlookup in the logs, just
>
> 2017/01/21 14:17:48 PIM: pim_socket_recvfromto: HAVE_IP_PKTINFO
> to=224.0.0.13,103
> 2017/01/21 14:17:48 PIM: Recv IP PIM pkt size=54 from 172.24.83.250 to
> 224.0.0.13 on fd=14 on ifindex=12 (sock_ifindex=10)
> 2017/01/21 14:17:48 PIM: Scheduling READ event on PIM socket fd=14
> 2017/01/21 14:17:48 PIM: pim_socket_recvfromto: HAVE_IP_PKTINFO
> to=224.0.0.13,103
> 2017/01/21 14:17:48 PIM: Recv IP PIM pkt size=54 from 172.24.83.250 to
> 224.0.0.13 on fd=15 on ifindex=12 (sock_ifindex=12)
> 2017/01/21 14:17:48 PIM: Recv IP packet from 172.24.83.250 to 224.0.0.13 on
> NY4FR2: size=54 ip_header_size=20 ip_proto=103
> 2017/01/21 14:17:48 PIM: Recv PIM packet from 172.24.83.250 to 224.0.0.13 on
> NY4FR2: ttl=1 pim_version=2 pim_type=0 pim_msg_size=34 checksum=bb9
> 2017/01/21 14:17:48 PIM: neighbor_timer_off: cancelling timer for neighbor
> 172.24.83.250 on NY4FR2
> 2017/01/21 14:17:48 PIM: pim_neighbor_timer_reset: starting 105 sec timer
> for neighbor 172.24.83.250 on NY4FR2
> 2017/01/21 14:17:48 PIM: Scheduling READ event on PIM socket fd=15
> 2017/01/21 14:17:48 PIM: pim_socket_recvfromto: HAVE_IP_PKTINFO
> to=224.0.0.13,103
> 2017/01/21 14:17:48 PIM: Recv IP PIM pkt size=54 from 172.24.83.250 to
> 224.0.0.13 on fd=16 on ifindex=12 (sock_ifindex=13)
> 2017/01/21 14:17:48 PIM: Scheduling READ event on PIM socket fd=16
> 2017/01/21 14:17:48 PIM: pim_socket_recvfromto: HAVE_IP_PKTINFO
> to=224.0.0.13,103
> 2017/01/21 14:17:48 PIM: Recv IP PIM pkt size=54 from 172.24.83.250 to
> 224.0.0.13 on fd=18 on ifindex=12 (sock_ifindex=3)
> 2017/01/21 14:17:48 PIM: Scheduling READ event on PIM socket fd=18
> 2017/01/21 14:17:49 PIM: pim_socket_recvfromto: HAVE_IP_PKTINFO
> to=224.0.0.22,2
> 2017/01/21 14:17:49 PIM: pim_socket_recvfromto: HAVE_IP_PKTINFO
> to=224.0.0.22,2
> 2017/01/21 14:17:49 PIM: pim_socket_recvfromto: HAVE_IP_PKTINFO
> to=224.0.0.22,2
> 2017/01/21 14:17:49 PIM: pim_socket_recvfromto: HAVE_IP_PKTINFO
> to=224.0.0.22,2
> 2017/01/21 14:17:49 PIM: send_join: can't send join upstream:
> RPF'(10.55.0.132,239.0.100.100)=0.0.0.0
> 2017/01/21 14:17:49 PIM: pim_joinprune_send: sending
> Join(S,G)=(10.55.0.132,239.0.100.100) to upstream=0.0.0.0 on interface eth1
> 2017/01/21 14:17:49 PIM: pim_joinprune_send:
> Join(S,G)=(10.55.0.132,239.0.100.100): upstream=0.0.0.0 is myself on
> interface eth1
> 2017/01/21 14:17:49 PIM: join_timer_start: starting 60 sec timer for
> upstream (S,G)=(10.55.0.132,239.0.100.100)
> 2017/01/21 14:17:49 PIM: send_join: can't send join upstream:
> RPF'(10.6.7.25,239.0.100.102)=0.0.0.0
> 2017/01/21 14:17:49 PIM: pim_joinprune_send: sending
> Join(S,G)=(10.6.7.25,239.0.100.102) to upstream=0.0.0.0 on interface eth1
> 2017/01/21 14:17:49 PIM: pim_joinprune_send:
> Join(S,G)=(10.6.7.25,239.0.100.102): upstream=0.0.0.0 is myself on interface
> eth1
> 2017/01/21 14:17:49 PIM: join_timer_start: starting 60 sec timer for
> upstream (S,G)=(10.6.7.25,239.0.100.102)
>
> But the BGP routes are there.
> If I configure static routes, the next hop is populated and multicast
> routing is working.
>
> Maybe someone can help?
>
> Regards,
> Joerg
>
>
>