Jesse DuPont
Owner
/ Network
Architect
email:
jesse.dup...@celeritycorp.net
Celerity
Networks LLC / Celerity
Broadband LLC
Like us!
facebook.com/celeritynetworksllc
Like us! facebook.com/celeritybroadband
The P2P provider is saying that they find no technical reason why the IPv6 traffic is not passing, and they're pointing to my equipment.
I've plugged a couple more routers into the CRS317 to test with. What I found is that if I have the PVID set to the Vlan, and Admit all packets on the interface, then I can ping with IPv6 between the 2 routers. But if I define a VLan on either router, no IPv6 traffic passes.
Router1 <- Ethernet Cable -> Router2 Ipv6 works
Router1 <-CRS317 port1 PVID=3000 <-> CRS317 Port2 PVID=3000 -> Router2 IPv6 works
Router1 <-CRS317 Port1 PVID=3000 <-> CRS317 Port2 -> Vlan3000 Router 2 IPv6 fails.
But even if I have the PVID of my P2P port and Router1 set the same, I still can't ping an IPv6 address on the other end of the P2P circuit. I'm about ready to give up and Kill my Dualstack.
Has anyone tried IPv6 with Vlans on a CRS317? Am I missing something simple?
On 2/15/2021 11:34 AM, Jesse Dupont wrote:
The CRS will pass all ethertypes by default.
Sent from my iPhone
On Feb 15, 2021, at 9:52 AM, Nate Burke <n...@blastcomm.com> wrote:
After escalating with the provider, they think that might be what is happening. If what they're telling me is true, Their Core network is running MPLS, but it's in a QinQ to the NID that I'm plugged into. So all IPv4 is fine, but might not pass IPv6.
So do I need to set anything special on my CRS317 bridge to pass the IPv6 Traffic on the Vlans? The Only Ethertype setting I see in the bridge interface itself, and you can only choose 0x8100, 0x88a8, or 0x9100
On 2/15/2021 10:12 AM, Jesse DuPont wrote:
It is possible your P2P provider is filtering Ethertypes. IPv4's Ethertype if 0x0800 whereas IPv6's Ethertype is 0x86dd. This field is right after the 802.1Q VLAN header in the L2 header.
Jesse DuPont
Owner / Network Architect
email: jesse.dup...@celeritycorp.net
Celerity Networks LLC / Celerity Broadband LLC
Like us! facebook.com/celeritynetworksllcLike us! facebook.com/celeritybroadband
![]()
On 2/15/21 7:36 AM, Nate Burke wrote:
I don't 'think' it's an MTU issue, even a simple IPv6 ping won't go across the P2P circuit.
On the CRS317, if I torch Vlan3000 on the ccr1009 interface, I see the IPv6 icmp packets coming in from my CCR1009, but if I torch the Interface going to the P2P circuit on the CRS317, I see no IPv6 traffic. Is that a problem, or because of the way the Mikrotik packet flows work, I would only see traffic as it enters the router, and since there is no reply from the far end, nothing would show up.
On 2/15/2021 6:15 AM, Adam Moffett wrote:
I don't think MTU would be the issue. If the endpoint has a 1500 byte MTU then it will send IP packets up to 1500 bytes long. IPv6 header is a little longer, but that means the data portion will just have to be a little shorter. So if you had an MTU issue with v6 you'd have an MTU issue with v4 also.
On 2/14/2021 5:27 PM, Steve Jones wrote:
The ip6 traffic is encapsulated in the vlan? Ip6 headers I'm assuming are larger because the number itself is longer, mtu. But I'm just now learning to use vlans and I lick the tasty windows so I dont know a lot
On Sun, Feb 14, 2021, 3:00 PM Adam Moffett <dmmoff...@gmail.com> wrote:
Unless you've defined it as a protocol based VLAN then no it shouldn't
matter.
On 2/14/2021 3:12 PM, Nate Burke wrote:
> I have a P2P circuit plugged into a CRS317-1G-16S+ and setting
> PVID3000 on that bridge interface. A Mikrotik CCR1009 is plugged
> into another port on the CRS317 set for trunked VLan3000. I can move
> IPv4 traffic on Vlan3000 and to the other end of the P2P circuit just
> fine, but I cannot move any IPv6 traffic on Vlan3000.
>
> The Vlan shouldn't care about IP4 vs IP6 traffic, right? I just want
> to make sure I'm not causing my own problem before pointing my finger
> at the Circuit.
>
> The IPv6 package is not installed on the CRS317, because it's just
> acting as a switch, it's not doing any routing. That shouldn't
> matter, right?
>
--
AF mailing list
AF@af.afmug.com
http://af.afmug.com/mailman/listinfo/af_af.afmug.com
--
AF mailing list
AF@af.afmug.com
http://af.afmug.com/mailman/listinfo/af_af.afmug.com
-- AF mailing list AF@af.afmug.com http://af.afmug.com/mailman/listinfo/af_af.afmug.com
