Re: [FROG] Source specific Multicast Support on FRR PIM implementation
`show ip pim neighbor` and `show ip pim join` We need to figure out why colo1 is not installing the mroute as well. Is there anything in the log on colo1? donald On Thu, Sep 20, 2018 at 7:25 AM, Jonas Dibbie <dibbie@web.de> wrote:
Hi Donald,
Thank you very much for your feedback.
SSM should work, configure the interface(s) with `ip pim ssm` and make sure you have a ssm range specified via the `ip pim ssm prefix-list <plistname>` if doing beyond the default SSM range.
Under the interface configuration I can only typ "ip pim sm" for the sparse mode. Do you mean this or do I have a missing option 'ip pim ssm'?
For a better understandig, my topology are two centos 7 linux hosts. One sitting int 'Colo1' [10.190.0.22] and one in 'Colo2' [10.55.0.33]. They have a direct MPLS link. This Link is Ipsec encapsulated (Strongswan) and has a GRE Interface for Multicast support. Each Linux Host is connected via Layer 3 Transferlink (BGP Relation) to the inside LAN Arista Switches, where the multicast sender and receiver are connected. All switches are PIM-SM SSM enabled.
Currently the receiver is in Colo1 [10.190.0.22] and the sender in Colo2 [10.55.0.33]
`show ip multicast`
Colo1: -----
Mroute socket descriptor: 7(Default-IP-Routing-Table) Mroute socket uptime: 19:50:34
Zclient update socket: 10 failures=0 Zclient lookup socket: 13 failures=0
Maximum highest VifIndex: 31
Upstream Join Timer: 60 secs Join/Prune Holdtime: 210 secs PIM ECMP: Disable PIM ECMP Rebalance: Disable
RPF Cache Refresh Delay: 50 msecs RPF Cache Refresh Timer: 0 msecs RPF Cache Refresh Requests: 2 RPF Cache Refresh Events: 2 RPF Cache Refresh Last: 00:52:38 Nexthop Lookups: 7 Nexthop Lookups Avoided: 0
Scan OIL - Last: 00:52:38 Events: 2 MFC Add - Last: 00:00:24 Events: 5 MFC Del - Last: 00:00:56 Events: 2
Interface Address ifi Vif PktsIn PktsOut BytesIn BytesOut Colo1:Colo2 172.21.21.75 16 0 0 0 0 0 ens4f1.308 172.22.190.10 12 1 0 0 0 0
Colo2: -----
Mroute socket descriptor: 7(Default-IP-Routing-Table) Mroute socket uptime: 46:01:22
Zclient update socket: 10 failures=0 Zclient lookup socket: 13 failures=0
Maximum highest VifIndex: 31
Upstream Join Timer: 60 secs Join/Prune Holdtime: 210 secs PIM ECMP: Disable PIM ECMP Rebalance: Disable
RPF Cache Refresh Delay: 50 msecs RPF Cache Refresh Timer: 0 msecs RPF Cache Refresh Requests: 16 RPF Cache Refresh Events: 13 RPF Cache Refresh Last: 19:40:22 Nexthop Lookups: 743 Nexthop Lookups Avoided: 0
Scan OIL - Last: 19:40:22 Events: 13 MFC Add - Last: --:--:-- Events: 0 MFC Del - Last: 00:02:32 Events: 371
Interface Address ifi Vif PktsIn PktsOut BytesIn BytesOut Colo2:Colo1 172.21.21.76 21 0 0 0 0 0 eth0 172.24.37.3 2 0 0 0 0 0 pimreg 0.0.0.0 45 0 0 0 0 0
`show ip pim upstream`
Colo1: -----
Iif Source Group State Uptime JoinTimer RSTimer KATimer RefCnt Colo1:Colo210.55.0.33 232.122.14.42 J 00:01:00 00:00:40 --:--:-- --:--:-- 1
Colo2: -----
Iif Source Group State Uptime JoinTimer RSTimer KATimer RefCnt eth0 10.55.0.33 232.122.14.42 J 00:02:35 00:00:04 --:--:-- --:--:-- 1
`show ip mroute`
Colo1: -----
Source Group Proto Input Output TTL Uptime 10.55.0.33 232.122.14.42 PIM Colo1:Colo2 ens4f1.308 1 00:01:24
Colo2: -----
Source Group Proto Input Output TTL Uptime
I personally am a bit interested in the rpf for 172.24.37.3 `show ip rpf 172.24.37.3`.
For a better debuging I changed the sender 172.24.37.3 to another host. This was on the FRR Host before. The new IP is 10.55.0.33. I did this command on the host in Colo2 (closer to the sender).
# sh ip rpf 10.55.0.33 Routing entry for 10.55.0.0/24 using Unicast RIB Known via "bgp", distance 200, metric 0, best Last update 1d22h03m ago * 172.24.37.102, via eth0
If you need any further debug output feel free to ask. Thank you very much for helping on this issue.
greetings, Jonas
_______________________________________________ frog mailing list frog@lists.frrouting.org https://lists.frrouting.org/listinfo/frog
participants (2)
-
Donald Sharp -
Jonas Dibbie