Community discussions

MUM Europe 2020
 
tholbrook
just joined
Topic Author
Posts: 5
Joined: Thu Jun 04, 2009 12:52 am

Cisco VPLS (BGP) and MTU

Thu Jun 04, 2009 6:03 pm

Is there a way to set the MPLS MTU on dynamically created VPLS interfaces? Specifically in setting up a Cisco style VPLS using BGP.

Obviously when you statically create the interface you do in the "interface VPLS" statement but when you set a VPLS up using BGP there doesn't seem to be anywhere to set it:
[admin@A] > /interface vpls cisco-bgp-vpls set cisco-bgp-vpls1 ?
Change properties of one or several items.

<numbers> -- List of item numbers
bridge -- 
bridge-cost -- 
bridge-horizon -- 
comment -- Set comment for items
disabled -- 
export-route-targets -- 
import-route-targets -- 
l2router-id -- 
name -- Interface name
route-distinguisher -- 
vpls-id -- 


Thanks
 
Mplsguy
MikroTik Support
MikroTik Support
Posts: 226
Joined: Fri Jun 06, 2008 5:06 pm

Re: Cisco VPLS (BGP) and MTU

Thu Jun 04, 2009 9:56 pm

Currently there is no way to do that. Is there any particular reason why you need this feature?
 
tholbrook
just joined
Topic Author
Posts: 5
Joined: Thu Jun 04, 2009 12:52 am

Re: Cisco VPLS (BGP) and MTU

Thu Jun 04, 2009 11:33 pm

In a generic case we might need it for all the normal reasons you need to be able to adjust protocol MTU. For example I may be running multiple labels and/or 802.1Q tags through a network where have a device like a switch that has a lower physical MTU than required and cannot fragment.

In this particular case though I have set up a simple 3 node BGP MPLS network of 493AHs and am unable to pass traffic larger than 1474 bytes (IP). I expect that they would just fragment the frames but they aren't. Setting the MPLS MTU was just my first attempt at troubleshooting. Any suggestions would be welcome.
 
tholbrook
just joined
Topic Author
Posts: 5
Joined: Thu Jun 04, 2009 12:52 am

Re: Cisco VPLS (BGP) and MTU

Fri Jun 05, 2009 1:21 am

This should be easy to reproduce btw. The RBs are connected in a straight line:

/--e9-(A)-e2----------------e3-(B)-e2---------------e3-(C)-e9---/

The intent is to create a dynamic VPLS from one side to the other. Router B is a route reflector. As I mentioned the network comes up and operates fine. There is just an issue with the frame size that will successfully pass. Here are the non-default configurations settings:
/system identity set name=A
/interface bridge add name=lo0
/interface bridge add name=customer1
/ip address add interface=lo0 address=10.127.0.1 netmask=255.255.255.255
/ip address add interface=ether2 address=10.1.0.1 netmask=255.255.255.0
/routing ospf instance set default router-id=10.127.0.1
/routing ospf network add area=backbone disabled=no network=10.1.0.0/24
/routing ospf network add area=backbone disabled=no network=10.127.0.1/32
/routing bgp set default router-id=10.127.0.1
/routing bgp peer add remote-as=65530 update-source=lo0 instance=default remote-address=10.127.0.2 address-families=l2vpn-cisco
/mpls ldp set enabled=yes transport-address=10.127.0.1 lsr-id=10.127.0.1
/mpls ldp interface add interface=ether2
/interface vpls bgp-vpls add bridge=customer1 bridge-horizon=1 route-distinguisher=10:1 site-id=1 

/system identity set name=B
/interface bridge add name=lo0
/ip address add interface=lo0 address=10.127.0.2 netmask=255.255.255.255
/ip address add interface=ether3 address=10.1.0.2 netmask=255.255.255.0
/ip address add interface=ether2 address=10.2.0.1 netmask=255.255.255.0
/routing ospf instance set default router-id=10.127.0.2
/routing ospf network add area=backbone disabled=no network=10.1.0.0/24
/routing ospf network add area=backbone disabled=no network=10.2.0.0/24
/routing ospf network add area=backbone disabled=no network=10.127.0.2/32
/routing bgp set default router-id=10.127.0.2
/routing bgp peer add remote-as=65530 update-source=lo0 instance=default remote-address=10.127.0.1 address-families=l2vpn-cisco route-reflect=yes
/routing bgp peer add remote-as=65530 update-source=lo0 instance=default remote-address=10.127.0.3 address-families=l2vpn-cisco route-reflect=yes
/mpls ldp set enabled=yes transport-address=10.127.0.2 lsr-id=10.127.0.2
/mpls ldp interface add interface=ether2
/mpls ldp interface add interface=ether3

/system identity set name=C
/interface bridge add name=lo0
/interface bridge add name=customer1
/ip address add interface=lo0 address=10.127.0.3 netmask=255.255.255.255
/ip address add interface=ether3 address=10.2.0.2 netmask=255.255.255.0
/routing ospf instance set default router-id=10.127.0.3
/routing ospf network add area=backbone disabled=no network=10.2.0.0/24
/routing ospf network add area=backbone disabled=no network=10.127.0.3/32
/routing bgp set default router-id=10.127.0.3
/routing bgp peer add remote-as=65530 update-source=lo0 instance=default remote-address=10.127.0.2 address-families=l2vpn-cisco
/mpls ldp set enabled=yes transport-address=10.127.0.3 lsr-id=10.127.0.3
/mpls ldp interface add interface=ether3
/interface vpls bgp-vpls add bridge=customer1 bridge-horizon=1 route-distinguisher=10:1 site-id=2
 
Mplsguy
MikroTik Support
MikroTik Support
Posts: 226
Joined: Fri Jun 06, 2008 5:06 pm

Re: Cisco VPLS (BGP) and MTU

Fri Jun 05, 2009 9:56 am

What exact version and packages are you using? Which routerboards and which ports are used in your setup? What is mpls-mtu setup on all routers?

When you use BGP AF l2vpn-cisco, BGP distributes NLRIs created by /interface cisco-bgp-vpls, when you use BGP AF l2vpn, BGP distributes NLRIs created by /interface bgp-vpls.
 
tholbrook
just joined
Topic Author
Posts: 5
Joined: Thu Jun 04, 2009 12:52 am

Re: Cisco VPLS (BGP) and MTU

Mon Jun 08, 2009 7:00 pm

I posted the wrong config above. The working configuration has the correct "/interface vpls cisco-bgp-vpls" statements. As I mentioned though there isn't a problem with its basic functionality, only the large frames.

Here are the other details:

-These are 3 identical 493AH boards with the following software:
[admin@A] > system routerboard print
       routerboard: yes
             model: "493AH"
     serial-number: "1A8E01EECE87"
  current-firmware: "2.18"
  upgrade-firmware: "2.20"
[admin@A] /system package> print
Flags: X - disabled 
 #   NAME                    VERSION                    SCHEDULED              
 0   system                  3.24                                              
 1   routerboard             3.24                                              
 2 X ipv6                    3.24                                              
 3   mpls-test               3.24                                              
 4   dhcp                    3.24                                              
 5   routing-test            3.24                                              
 6 X routing                 3.24                                              
 7   hotspot                 3.24                                              
 8   wireless                3.24                                              
 9   routeros-mipsbe         3.24                                              
10   security                3.24                                              
11   ppp                     3.24                                              
12   advanced-tools          3.24      
-The ports are reflected in my line diagram above. I'll redraw it here:

/--e9-(A)-e2----------------e3-(B)-e2---------------e3-(C)-e9---/

-I have not changed any mtu settings so they are all at defaults:
/mpls interface
add comment="" disabled=no interface=all mpls-mtu=1508

Who is online

Users browsing this forum: No registered users and 9 guests