What version of code on the two devices. Jude ... View more

if your mpls cloud is 7600 with ES+ cards you can use service instance 10 ethernet encapsulation dotq1 10 rewrite ingress tag translate 1-to-1 dot1q 5 xconnect 4.4.4.4 encapsulation mpls ASR 9k interface GigabitEthernet0/0/0/0/3.10 l2transport encapsulation dot1q 10 rewrite ingress tag translate 1-to-1 dot1q 5 symmetric Regards Jude ... View more

I agree with Atif_hafeez I have also used service groups with success. However keep in mind usually/hopefully the bandwidth is present in the core network.  Let the core just swap labels.  they will cary any policies you may have presented on packets Configure all of your policies at the edge.  Ingress/Egress. Regards Jude Bryant ... View more

on your metro device perform a show interface transceiver to show light levels of sfp. Also if you are using a program like cacti you can creat thresholds on light levels to alarm you as the light may be degrading. Always worked for me Regards Jude Bryant ... View more

Thats interesting, as I have configured thousands of Service Instances and never knew that option existed. Starting with SRA back in the day (didnt mean to rhyme) to SRE and now of course 15.xx if you find out post for my curiosity Jude ... View more

May I ask why you want to do such. Sub interface interface GigabitEthernet3/0/1.700 description Customer A Traffic encapsulation dot1Q 700 ip address 10.1.1.1 255.255.255.192 then route this traffic somewhere.  However when you put ip address on this interface it is now in routed mode and you cannot create/use EVC services. Service Instance (EVC) service instance 700 ethernet encapsulation dot1q 700 !if you want to hairpin on same card to different port xconnect gig3/0/1 !if you want to xconnect to different router with lo0 of 4.4.4.4 xconnect 4.4.4.4 700 encapsulation mpls !if you will be configuring an SVI for vlan 700 rewrite ingress tag pop 1 symmetric bridge-domain 700 interface vlan700 ip address 10.1.1.1 255.255.255.192 This does the same thing, but you will use an SVI to router vlan 700 traffic if you want. using sub interfaces you are limited to 4096 and you cannot specify internal tag or customer tag if using QinQ. EVC you have 8000 service instances and the use of specifying second vlan in QinQ. Service Instance (EVC) service instance 7004 ethernet encapsulation dot1q 700 second dot1q 400 xconnect.... service instance 7006 ethernet encapsulation dot1q 700 second dot1q 600 xconnect.... better options this way.. Jude Bryant ... View more

Wbashlor, Sorry I misunderstood.  The ASR9k configuration is very similar to the 3800X Look at the picture I posted of the Multiple services on same port example.  The E-LAN(VPLS) sample config is what you need to use. ... View more

        Answered    ... View more

No need to make any changes.  I will write a config and post it soon. The config I wrote was for point to point pseudowire/EFP.  You are attempting VPLS so the config changes somewhat. Please read this documentation, http://www.cisco.com/en/US/docs/switches/metro/me3600x_3800x/software/release/15.2_2_S/configuration/guide/swmpls.html#wp1213789 Look for section; EFPs and Ethernet over Multiprotocol Layer Switching (EoMPLS) Enabling EoMPLS This section includes this information about configuring EoMPLS on a switch used as a provider-edge router: • Default EoMPLS Configuration • EoMPLS Configuration Guidelines • Configuring EoMPLS • Configuring the Pseudowire Using Pseudowire Class • Configuring L2VPN Interworking • EoMPLS and EVC • Packet Flow in an EoMPLS Network • Configuring L2VPN Pseudowire Redundancy • Routed VPLS/EoMPLS I particularly like the EVC XConnection configuration because you have the option of creating VLan based service instances without globally defining the VLan itself.  I understand the 3800x has Asics for bridge-domain processing, not done with software like the 7600 router. Also when encapsulating with MPLS the ASR9Ks are only label swapping.  No additional VLan based configuration required.  This option also gives you the ability to create traffic engineering from 3800x to 3800x. Primary, secondary, and tertiary using FRR.  Or you cant just let the IGP to take care of LSP dynamically. I understand the 3800x has Asics for bridge-domain processing, not done with software, however you have to globally define the vlan you are bridging. I hope this helps you. -Regards Jude Jude ... View more

Thanks for the additional info. I will look at this carefully and reply when the opportunity arises.  I assume you are running the ME3800X models with 15.1 or higher code and the advanced services with allows full mpls configurations. I will also assume your IGP (OSPF) is configured correctly as well. Chat with you soon Jude        attached ASR9K Config Sheet ... View more

Wbashlor Here is an example of how this works.  I hope this will be helpful too you. user/client<==>(switch)<==>(me3800x)<==>(asr9k1)<=mpls=>(asr9k2)<==>(me3800x)<==>(switch)<==>user/client asr9k1 has loopback0 ip address 1.1.1.1 asr9k2 has loopback0 ip address 2.2.2.2                                          client (Cisco switch) vlan 200 ! interface gig0/1 description "traffic to_from user_customer" switchport mode access switchport access vlan 200 ! interface gig0/24 description "trunk port to_from me3800x" switchport trunk encapsulation dot1q switchport mode trunk switchport trunk allowed vlan 200 (cisco me3800x) interface gig0/1 description "trunk port to_from switch" switchport trunk allowed vlan none switchport mode trunk mtu 9800 no cdp enable service instance 200 ethernet   encapsulation untagged , dot1q 200 (untagged is if you want to tunnel cdp, stp, vtp *please look below)   l2protocol tunnel cdp stp vtp   bridge-domain 200 ! interface gig0/2 description "connection to asr9k" switchport trunk allowed vlan none switchport mode trunk no cdp enable service instance 200 ethernet   encapsulation untagged , dot1q 200   rewrite ingress tag pop 1 symmetric   bridge-domain 200 (asr9k1) RP/0/RSP0/CPU0:asr9k1(config)#interface te0/0/0/2.200 l2transport RP/0/RSP0/CPU0:asr9k1(config-if)#description "connection to_from me3800x" RP/0/RSP0/CPU0:asr9k1(config-if)#encapsulation dot1q 200 RP/0/RSP0/CPU0:asr9k1(config)#l2vpn RP/0/RSP0/CPU0:asr9k1(config)# xconnect asr9k_mpls RP/0/RSP0/CPU0:asr9k1(config)#  p2p asr9k2 RP/0/RSP0/CPU0:asr9k1(config)#   interface te0/0/0/2.200 RP/0/RSP0/CPU0:asr9k1(config)#   neighber 2.2.2.2 pw-id 200 (2.2.2.2 is loopback of asr9k2) RP/0/RSP0/CPU0:asr9k1(config)#end * if the switches on the edge has a change to its native vlan, it needs to be reflected where the "untagged" word is* -Regards Jude ... View more

Interesting, since you are using bridge-domain, you may have to add the "rewrite ingress tag pop 1 symmetric" line to your configuration. Ill have to look at documentation. Jude Bryant ... View more

Michalis Milaitis, I like your drawing, however on the UPE why not use dot1q tunnel.  That way you dont have go globally define all vlans comming from the CPE.  The UPE can define one VLan (outer tag or Service Tag), and all incomming vlans ( inner tag or Customer Tag) gets double tagged. Also instead of using the bridge-domain command in your service instance, use the xconnect command.  The xconnect command will forward all incoming traffic from one port to another port, vs bridge-domain where you have to globally define the vlan and learn all mac-addressses on the incomming ports.  However if you need to share more than two ports then bridge-domain is the correct answer. These two suggestions will save overhead and resources on your devices. Regards Jude ... View more

along with the above comments, for future reference you might start adding the following line to the interface storm-control multicast level 10.00 (10 being a %, change to your preference) good practice for multicast storm control. Jude ... View more