Fault Mgmt on L2 Ethernet over MPLS

[lharman]

With typical ethernet it is a point to point solution and when the link goes down both interfaces on the link go down. With L2 Ethernet over MPLS, if a link goes down in the path of the network my Ethernet link is still up/up. I will have a backup link implemented for failover but the primary will still show up/up with the traffic using the backup link. I have considered GRE but the tunnel would just route over the back up link in the event of a failure. Any help would be greatly appreciated as I am at a loss of ideas.

Thanks,

lharman

[Harold Ritter]

You should be able to test the status of the link by running a routing protocol over the ethernet circuit and failover if the RP detects a link failure.

Hope this helps,

[lharman]

hritter,

thanks for the reply. If I run a RP over think and the link fails wouldn't the RP be routed over the b/u link.

[Harold Ritter]

Yes, isn't that what you want?

Thanks,

[lharman]

Actually I want a trap or alarm to notify my mgmt system that the link is down. recently we had an issue where the primary link went down inside the carrier cloud and traffic started routing over the b/u link. The b/u link also failed inside the cloud but both of my links ethernet links showed up/up. I am currently working with multiple carriers on a solution but non currently have anything. So I need sometype of layer-2 ping or some equivalent.

thanks

Thanks

[Harold Ritter]

You can use the syslog message generated by the RP to actually detect that the interface has not end to end connectivity although its status is up/up. For instance, if you loose end to end connectivity and you run OSPF you would get something similar to the following message:

Jul 5 12:58:50.029: %OSPF-5-ADJCHG: Process 1, Nbr 192.168.2.1 on GigabitEth

ernet3/1 from FULL to DOWN, Neighbor Down: Dead timer expired

This could then be used by your NMS to generate alarms.

Hope this helps