BGP attribute-download and VPNv4
Using “bgp attribute-download” to get source AS number into netflow, and use that for traffic analysis.
According to cisco this command “bgp attribute-download” is only available for IPv4 unicast address family.
gewoon, mijn archief
Using “bgp attribute-download” to get source AS number into netflow, and use that for traffic analysis.
According to cisco this command “bgp attribute-download” is only available for IPv4 unicast address family.
Interface vlan777
ipv6 enable
Otherwise, the config looks spot on
Our config looks like:
interface Vlan110
standby version 2
standby 110 ipv6 FE80::1
standby 110 timers 1 3
standby 110 priority 110
standby 110 preempt delay minimum 180
standby 110 authentication xxxx
ipv6 address dead:beef:1::FFFE/64
ipv6 enable
ipv6 nd other-config-flag
ipv6 nd router-preference High
ipv6 pim dr-priority 4294967295
ipv6 dhcp relay destination dead:beef:0::1
ipv6 dhcp relay destination dead:beef:0::2
Ended up with the following for DSL customers using DS-Lite:
block size 512
max blocks per user 16
block timeout 120
address-sharing-ratio 8:1
avg usage is 34 ports per block and 1,3 blocks per address, but the top 1% are at least x10.
Some years ago we had started with more relaxed numbers (to be on the safe side), but last year we squeezed them a little. No complaints until now.
Mon Feb 29 14:25:16.812 CST
Building configuration…
!! IOS XR Configuration 5.3.1
service cgn cgn1
service-location preferred-active 0/3/CPU0 service-type nat44 nat1
portlimit 5000
alg ActiveFTP
inside-vrf six
!
!
!
end
RP/0/RSP0/CPU0:eng-lab-9k-1(config-cgn-invrf)#bulk-port-alloc ?
size Bulk size for allocation.
Should be greater or equal to one fourth of the port limit
and less than twice the port limit.
For optimum results, should be close to half the port limit RP/0/RSP0/CPU0:eng-lab-9k-1(config-cgn-invrf)#bulk-port-alloc size ?
none No bulk allocation
8 Allocate 8 ports in bulk
16 Allocate 16 ports in bulk
32 Allocate 32 ports in bulk
64 Allocate 64 ports in bulk
128 Allocate 128 ports in bulk
256 Allocate 256 ports in bulk
512 Allocate 512 ports in bulk
1024 Allocate 1024 ports in bulk
2048 Allocate 2048 ports in bulk
4096 Allocate 4096 ports in bulk
vrf export route-policy doesn’t seem to directly deny prefixes from advertisement… but according to this one site you can indirectly deny prefixes… it seems that vrf export route-policy is mainly used to more granularly assign rt’s and add rt’s to rt’s (additive).
http://www.akbintel.com/mediawiki/index.php/VRF/Config#export_map
http://nagendrakumar-nagendra.blogspot.com/2011/09/mpls-vpn-vrf-export-map.html
Thought it was pretty clever to not put a route-target export x:x and then whichever prefixes are denied in the export route-policy will drop out of the route-map logic, and this will not be advertised since there isn’t a route-target export statement in the vrf. Cool.
For some reason especially on3.7 code we have also seen this message on ports which are left no shut, and they have an SFP in it.
That’s because they introduced DOM support for some transceivers:
As of July 28, 2008, Cisco PIX Security Appliance platforms/bundles are no longer being sold. Customers can still purchase accessories and licenses until January 27, 2009.
It is important to note that Cisco will continue to support Cisco PIX Security Appliance customers through July 27, 2013.
Based on understanding, the CPU on 1001-X is more powerful and can handle more concurrent session establishments. It also has 2 10G interfaces which can come in handy. However, the queue-count is considerably lower in the 1001s which makes QoS difficult, depending on your policies. I would recommend a 1002-X if you are looking at ASR 1K.
Have a look at this document here for more details:
This is a live working 897 using QinQ over VDSL:
interface Ethernet0
no ip address
!
interface Ethernet0.400
encapsulation dot1Q 101 second-dot1q 400 ip vrf forwarding test ip address 1.1.1.2 255.255.255.252 !
interface Ethernet0.401
encapsulation dot1Q 101 second-dot1q 401 ip vrf forwarding test-2 ip address 2.2.2.2 255.255.255.252
#show ver | i IOS
Cisco IOS Software, C800 Software (C800-UNIVERSALK9-M), Version 15.5(3)M, RELEASE SOFTWARE (fc1)
#show ver | i C897
Cisco C897VA-K9
#show license
Index 1 Feature: advipservices
Period left: Life time
License Type: Permanent
License State: Active, In Use
License Count: Non-Counted
License Priority: Medium
#show controllers VDSL 0 | i VDSL
Controller VDSL 0 is UP
Trained Mode: G.993.2 (VDSL2) Profile 17a
Here are some pictures of the ASR920 Console kit A920-CONS-KIT-S
The Adapter Plugs in the Top Left USB Console Port and we have it wired up to a Perle IOLAN SCS48C console server using a rollover cable.
Here are some pictures of it, since I can only find a brief mention of it in all the cisco docs.
service instance 940 ethernet
description description TEST_OUTER_940_Inner_942
encapsulation dot1q 940 second-dot1q 942
rewrite ingress tag pop 2 symmetric
bridge-domain 942
interface Vlan942
description TEST_OUTER_INNER_TAGs
mtu 9100
ip address 10.97.97.1 255.255.255.252
no ip proxy-arp
end
#ping 10.97.97.2
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 10.97.97.2, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 1/1/4 ms
#sh mac address-table dyn bridge-domain 942
Mac Address Table
——————————————-
BD Mac Address Type Ports
—- ———– ——– —–
942 5087.89fd.d320 DYNAMIC Gi0/24+Efp940
http://d2zmdbbm9feqrf.cloudfront.net/2012/usa/pdf/BRKSPG-2209.pdf
A Catalyst Switch will only look at the first VLAN Tag(Outter), it doesn’t care about the inner vlan tag and will forward the frame on. Just watch your MTU Size, because you lose 4btyes to the inner vlan tag.
A Good Example. http://blog.jhe.co/2009/11/dot1q-tunneling.html
Look at the Routed VPLS/EoMPLS section
What you are looking for a is a VPLS Routed Interface Configuration, just association the VPLS instance with the bridge group. I have never tried this, so I could be wrong, but it’s worth a shot.
Try adding this. This creates’ a VPLS Instance and associates it to the Bridge Group 941.
l2 vfi TESTING manual
vpn id 941
bridge-domain 941
Then under your VLAN Interface associate the VLAN Interface with the VPLS Instanace.
interface Vlan941
xconnect vfi TESTING <<<<<<<<<<<< ADD THIS
description INNER_OUTER_TAG_TEST
mtu 9100
ip address xxx.xxx.xxx.xxx 255.255.255.252 no ip proxy-arp
Also watch your MTU Size’s a mismatch can cause the VPLS instance to be down. You can check it with “show mpls l2transport vc vcid 941 detail”
This looks good
interface GigabitEthernet0/24 <- Connects to 4500X
service instance 940 ethernet
description description Inner_outer_tag_test_Outer_940_Inner_941
encapsulation dot1q 940 second-dot1q 941
rewrite ingress tag pop 2 symmetric
bridge-domain 941
Here is a step by step,
Step By Step Description Commands “semi-colon is a new line”
————————————————————————————————————————–
Global Config
6a. ASR920 Created L3 Routed Interface bridge-group 820; interface bdi820
6b. ME3800 Create L3 Routed Interface interface vlan 820
6b. ME3800 Connect Interface to BridgeGroup xconnect vfi TESTING
Cisco 4500X Provider Port
Cisco 4500X Port to ME3800.
Cisco ME3800 Port Config
I Did lab this up on a ASR920, the commands should be fairly close, I noted the difference between the ASR920 and ME3800
Carrier Side: QinQ Interface Outer VLAN 800 InnerVLAN 20 IP 192.168.0.1/24
l2 vfi TESTING manual
vpn id 820
bridge-domain 820
interface GigabitEthernet0/0/11
no ip address
negotiation auto
service instance 820 ethernet
encapsulation dot1q 800 second-dot1q 20
rewrite ingress tag pop 2 symmetric
bridge-domain 820
!!! ON A ASR920
bridge-domain 820
interface BDI820
ip vrf forwarding TESTING2
ip address 192.168.0.2 255.255.255.0
no shut
!!!!ON A ME3800
int vlan 820
xconnect vfi TESTING
ip address 192.168.0.2 255.255.255.0
no shut
ASR920#ping vrf TESTING2 192.168.0.1
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 192.168.0.1, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 1/2/4 ms
ASR920#show mac-address-table bdomain 820
Nile Mac Address Entries
BD mac addr type ports
———————————————————————————————-
820 0012.448e.8062 DYNAMIC Gi0/0/11.Efp820
ASR920#sh ip arp vrf TESTING2
Protocol Address Age (min) Hardware Addr Type Interface
Internet 192.168.0.1 2 0012.448e.8062 ARPA BDI820
Internet 192.168.0.2 – 00f2.8bdd.603f ARPA BDI820
You must do switchport commands
Services Instance is just a number, we keep it the same as the VLAN Id Under the Service Instance you specify the VLAN ID with endcapsulation dot1q x Rewrite ingress Tag POP 1 symmetric — This removes the first VLAN Tag on in incoming frame, if you do pop 2, it removes the 2 VLAN Tags.
First Example is VPLS with Bridge Domains. Bridge domain ID does not have to the same as the VLAN Id Second Example is EoMPLS XConnect
VPLS
——
l2 vfi VPLS1 manual
vpn id 41
bridge-domain 41
neighbor 1.2.3.4 encapsulation mpls
interface GigabitEthernet0/19
switchport trunk allowed vlan none
switchport mode trunk
mtu 9180
!VPLS Example
service instance 41 ethernet
encapsulation dot1q 41
rewrite ingress tag pop 1 symmetric
bridge-domain 41
!EoMPLS Example
service instance 117 ethernet
encapsulation dot1q 117
rewrite ingress tag pop 1 symmetric
xconnect 3.4.5.6 275 encapsulation mpls pw-class L2VPN
xconnect <Remote router IP> <Unique ID in the AS>
Thanks for the quick reply – We use service instances quite a bit, but only popping first tag, then creating vlan int (for vrf), or PW….so fairly basic stuff 🙂
In this circumstance, where we would receive vlan 800 as outer tag, and we want to access inner vlans 10,20,30 how would this look under a service instance?
service instance 800 ethernet
description description LINK_TO_CARRIER_X_VIA_4500X
encapsulation dot1q 800 second-dot1q 10,20,30
rewrite ingress tag pop 2 symmetric
bridge-domain ?
or a separate service instance and pop inner vlans on each one?
service instance 800 ethernet
description description LINK_TO_CARRIER_X_VIA_4500X
encapsulation dot1q 800 second-dot1q 10
rewrite ingress tag pop 2 symmetric
bridge-domain 10
service instance 801 ethernet
description description LINK_TO_CARRIER_X_VIA_4500X
encapsulation dot1q 800 second-dot1q 20
rewrite ingress tag pop 2 symmetric
bridge-domain 20
You are probably better using a service instance for each vlan. Then you can send each VLAN where ever.
service instance 800 ethernet
description description LINK_TO_CARRIER_X_VIA_4500X
encapsulation dot1q 800 second-dot1q 10
rewrite ingress tag pop 2 symmetric
bridge-domain 10
service instance 801 ethernet
description description LINK_TO_CARRIER_X_VIA_4500X
encapsulation dot1q 800 second-dot1q 20
rewrite ingress tag pop 2 symmetric
bridge-domain 20