"No RP address for this group"

dandrzejewski · Fri Oct 16, 2015 7:15 pm

I'm seeing lots of these in my logs from pim:

12:12:36 pim,warning RX WHOLEPKT signal: vif_index = 31 src = 10.1.1.2 dst = 239.255.255.250 len = 201: no RP address for this group

I don't know what's causing them, and it doesn't appear to be affecting anything negatively.

Here's my PIM config.

[admin@MikroTik] /routing pim> export verbose
# oct/16/2015 12:14:36 by RouterOS 6.32.2
# software id = XXXX-XXXX
#
/routing pim
set switch-to-spt=yes switch-to-spt-bytes=0 switch-to-spt-interval=1m40s
/routing pim interface
add alternative-subnets="" assert-override-interval=3s assert-time=3m disabled=no dr-priority=1 hello-holdtime=1m45s hello-period=30s hello-trigerred-delay=5s igmp-version=IGMPv2 interface=vlan101 join-prune-holdtime=3m30s \
    join-prune-period=1m override-interval=250 preferred-source-address=0.0.0.0 propagation-delay=50 protocols=pim,igmp require-hello=yes tracking-support=yes
add alternative-subnets="" assert-override-interval=3s assert-time=3m disabled=no dr-priority=1 hello-holdtime=1m45s hello-period=30s hello-trigerred-delay=5s igmp-version=IGMPv2 interface=vlan102 join-prune-holdtime=3m30s \
    join-prune-period=1m override-interval=250 preferred-source-address=0.0.0.0 propagation-delay=50 protocols=pim,igmp require-hello=yes tracking-support=yes
add alternative-subnets="" assert-override-interval=3s assert-time=3m disabled=no dr-priority=1 hello-holdtime=1m45s hello-period=30s hello-trigerred-delay=5s igmp-version=IGMPv2 interface=vlan103 join-prune-holdtime=3m30s \
    join-prune-period=1m override-interval=250 preferred-source-address=0.0.0.0 propagation-delay=50 protocols=pim,igmp require-hello=yes tracking-support=yes

doneware · Fri Oct 16, 2015 8:01 pm

add RP entry in all routers which are participating in multicast forwarding.

/routing pim rp add address=x.x.x.x disabled=no

basically you pick one device, which is in the forwarding path between the source(s) and the receivers.
if you have a redundant network, create a "loopback" interface (a bridge without ports) in the chosen device,
assign a /32 address to it, and use this as RP address.

make sure that this address gets into the routing table of all other routers in the network, otherwise RPF checks will fail.

dandrzejewski · Fri Oct 16, 2015 9:36 pm

OK so this is all happening on one device - an RB850Gx2. I have 3 VLANs that I am routing between. So I would add this 'loopback' bridge there? What subnet would the address be on?

doneware · Sun Oct 18, 2015 4:19 pm

So I would add this 'loopback' bridge there? What subnet would the address be on?

loopback is only needed if you don't want your RP flapping because the interface where it gets its IP address went down. you can use whatever /32 address which is routed in your network.
or it's just a single device?

dandrzejewski · Mon Oct 19, 2015 11:22 pm

I only have one router on this network, and 3 VLANs on that router.

vikramdevatha · Thu Apr 28, 2016 5:28 am

hi

i have the same "problem".. im getting a "No RP address for this group" error message in the log. I executed the command mentioned in this email thread, viz. "rp add address=10.10.10.24 disabled=no".

Now the log gives a "upstream neighbour not found" message.

Im not a techie, and very new to Mikrotik, please advise.

Here is the "export verbose" output:
[admin@Router] /routing pim> export verbose
# jan/02/1970 16:00:52 by RouterOS 6.7
# software id = KYJ0-FXWP
#
/routing pim
set switch-to-spt=yes switch-to-spt-bytes=0 switch-to-spt-interval=1m40s
/routing pim interface
add alternative-subnets="" assert-override-interval=3s assert-time=3m disabled=\
no dr-priority=1 hello-holdtime=1m45s hello-period=30s \
hello-trigerred-delay=5s igmp-version=IGMPv2 interface=all \
join-prune-holdtime=3m30s join-prune-period=1m override-interval=250 \
preferred-source-address=0.0.0.0 propagation-delay=50 protocols=pim,igmp \
require-hello=yes tracking-support=yes
/routing pim rp
add address=10.10.10.24 disabled=no group=224.0.0.0/4 hash-mask-length=30 \
priority=192

regards
vikram

Thu Apr 28, 2016 6:19 am

RP = rendezvous point - it's the point in the network where all group join/leave messages (from IGMP) get relayed to, and where multicast streams are delivered to by default, even if there are no subscribers to it. This becomes the "root" of all of your multicast traffic pathways. Traffic goes "up" to the RP, and then rains back "down" through every interface which has at least one subscriber.

"rp add address=10.10.10.24 disabled=no".

This would work if you have a router with that IP address on it. Whatever your most upstream router is (the one connected to your ISP) - choose an IP address on the router (a static IP address) and set RP=that address. If you have multiple routers, then all of them must use the same RP address or you'll have issues.

It was mentioned earlier in this thread, making a loopback interface for this RP address. The reason is simple - suppose you use the IP of some interface, and that interface gets disconnected or goes down for whatever reason. You'd still like media that doesn't use this interface to keep working even while the interface is down, so what you do is make a virtual interface (a loop interface) with a /32 IP address on it. Make sure that all routers have a route for this /32 (or if it's on the most upstream router, then all other routers' default gateway routes will lead to it, so that is fine). Create a bridge interface and name it "loopback" - then put an IP address on it - e.g. 10.10.10.24/32 (given your example IP). It can be anything you want so long as it's unique in your network, and not some public IP address you picked at random....

Set that IP address as the RP for all routers in your network, and you'll have it working.

vikramdevatha · Fri Apr 29, 2016 5:47 am

Thanks for the reply. In my network, x.x.x.24 is my mikrotik router, and x.x.x.1 is the modem connected to my ISP. So, should rp be set to x.x.x.1 or to x.x.x.24?

Whatever your most upstream router is (the one connected to your ISP) - choose an IP address on the router (a static IP address) and set RP=that address. If you have multiple routers, then all of them must use the same RP address or you'll have issues.

hence, i guess rp should be set to x.x.x.1 then? if so, please help me with the code to remove x.x.x.24... i tried "remove" but dont seem to understand the syntax to use it. sorry for such basic questions..

also, what you have said about loopback interface makes sense. I have a NAS, and im not able to access media on it when my internet is down, even though it is on my home network. So i guess i need to make a loopback interface and point my NAS to that. I'll try this and report back.

thanks again
vikram

vikramdevatha · Fri Apr 29, 2016 6:10 am

hi

sorry for the silly question in the previous post. i managed to remove the rp address that i had added. Now, i'll look into the loopback interface...

thanks again
vikram

Fri Apr 29, 2016 6:19 pm

You should only use the .1 if that device is also participating in multicast - i.e. it is running PIM.

The loopback address is the best way to go, so let us know how that works out for you.

vikramdevatha · Sat Apr 30, 2016 10:28 am

hi

so i did the following:

a) added a loopback interface
b) assigned x.x.x.1 to this interface (this is the IP connected to my ISP)
c) executed rp add address=x.x.x.1 disabled=no

export verbose of routing pim is given below:

[admin@Router] /routing pim> export verbose
# apr/30/2016 12:50:29 by RouterOS 6.7
# software id = KYJ0-FXWP
#
/routing pim
set switch-to-spt=yes switch-to-spt-bytes=0 switch-to-spt-interval=1m40s
/routing pim interface
add alternative-subnets="" assert-override-interval=3s assert-time=3m disabled=no \
dr-priority=1 hello-holdtime=1m45s hello-period=30s hello-trigerred-delay=5s \
igmp-version=IGMPv2 interface=all join-prune-holdtime=3m30s join-prune-period=1m \
override-interval=250 preferred-source-address=0.0.0.0 propagation-delay=50 protocols=\
pim,igmp require-hello=yes tracking-support=yes
/routing pim rp
add address x.x.x.1 disabled=no group=224.0.0.0/4 hash-mask-length=30 priority=192
[admin@Router] /routing pim>

warning messages ("no rp address", and "nearest upstream neighbor not found") in system log have now stopped!

so now my queries are as follows:

1. all media devices on my network should point to x.x.x.1? and not to x.x.x.24 (which is the IP of my router)?

2. in the interface list, the loopback interface is showing no traffic. is that normal?

3. you had mentioned to assign a /32 ip to the loopback interface. Im not sure how to do this... i assigned a x.x.x.1 ip, not sure if that is /32 or /24?

thanks for your assistance

regards
vikram

thenoob · Mon Nov 12, 2018 9:33 pm

adding a loopback address :
=================
first lets see what is used:

> ip address print            
Flags: X - disabled, I - invalid, D - dynamic 
 #   ADDRESS            NETWORK         INTERFACE                                                                                              
 0   10.0.0.1/24        10.0.0.0        bridge-lan

lets see what is our bridges:

 > interface print where type=bridge
Flags: D - dynamic, X - disabled, R - running, S - slave 
 #     NAME                                TYPE       ACTUAL-MTU L2MTU  MAX-L2MTU MAC-ADDRESS      
 0  R  bridge-lan                              bridge           1500  1598            XX:XX:XX:XX:XX:XX

so we can see we have 10.0.0.1 on bridge-lan interface with a mask of /24
we could just do "/routing pim rp add address=10.0.0.1" but lets make things complicated

but lets add a loopback bridge.

#add the bridge interface:
>interface bridge add name=loopback0

#set the ip on the interface:
>ip address add address=192.168.0.254/32 interface=loopback0
#or same thing but explicit netmask
>ip address add address=192.168.0.254 netmask 255.255.255.255 interface=loopback0

#add lo0 as a rp : 
>routing pim rp add address=192.168.0.254

of course you can change the 192.168.0.254 address to any private address that is not used and not part of another subnet on another interface.
after that you need to route that /32 network to your routing protocol to other "sites" or routers in your network.

"No RP address for this group"

"No RP address for this group"

Re: "No RP address for this group"

Re: "No RP address for this group"

Re: "No RP address for this group"

Re: "No RP address for this group"

Re: "No RP address for this group"

Re: "No RP address for this group"

Re: "No RP address for this group"

Re: "No RP address for this group"

Re: "No RP address for this group"

Re: "No RP address for this group"

Re: "No RP address for this group"

Who is online