EoMPLS VC will be declared up when there is a MPLS LSP (assume we are talking about LDP) to remote and VC label (and other parameters) are negotiated, my question is how does EoMPLS know tunnel is ready? IOS LDP use independent control, which means even local router has a label mapping for the destination loopback prefix, it is not guaranteed that a label switched path exits end to end. So how does EoMPLS know tunnel LSP is ready or not? does it do a mpls ping internally to verify the LSP?
1) When you configure an ingress interface on a PE for L2Transport, the PE allocates a new label called VC label and binds it to the configured VCID.
2) Now this VC Label to VCID binding is exchanged over the directed LDP session with the remote end PE. This goes as a Label Mapping Message,
3) when each PE receives the message, they try to match the message value with the VCID configured locally. If it matches both sides the VC state comes to an established state.
4) And the Label values received in the process are used to switch data for that particular VC.
5) Now since the whole exchange of VC labels takes place over already existing MPLS domain,
the LSP is place when the exchange happens. If the LSP is not in place between the 2 PE's then the exchange would never happen. So it really doesnt matter what label distribution mode is used by MPLS, as if the IGP labels are not available end-to-end the VC labels cannot be exchanged. (This is why we get and have a label stack for ATOM, one is the outer IGP Label, and then the inner VC label, which is exchanged over the existing MPLS IGP Labels)
Now the MPLS LSP Ping/Traceroute or VCCV are methods used to do operation and management of L3 and l2 VPN's after they are deployed to verify and manage their end-to-end connection state.
I don't think you answered my question, my question is how does EoMPLS know end-to-end LSP exists? distribution mode matters because if IOS uses ordered distribution, local router can safely claim end-to-end LSP exists as long as it receives the label mapping from downstream.
And why the whole exchange of VC labels have to take place over already exisiting MPLS domain? VC label exchange happens over a targted LDP session which is just a TCP connection, you can use L2TP encap for transportation, right?
The original post was about the VC being ready. When the VCID with VC labels are exchanged on both sides the VC is ready.
And since the question was with LDP and MPLS, the answer revolved around Tunnel Label (IGP Label from the existing MPLS domain). Its although not mandatory, since as per the PWE3, other modes or transport can be used like L2TP PW.
And coming to your additional question, yes using independent control, in some rare conditions it may happen that the End-To-End LSP is not really there (broken in between, although I havent experienced this in real life), in that case we have to use the OAM (LSP Ping/Traceroute with VCCV capabilities) to verify and correct the same.
This document is an early notification of a behaviour change that will be introduced in IOS XR release 6.5.
IOS XR configuration principles relevant for this article are:
On router platforms all interfaces must be by defaul...
With XR 4.2.0 the ASR9000 is releasing a new line of hardware models. This amongst others is the RSP440, the next generation RSP with faster switch fabric along with Typhoon based Linecards, the next generation network processor.
The Cisco EPN system incorporates a network architecture designed to consolidate multiples services on a single Multiprotocol Label Switching (MPLS) transport network. This network is designed primarily based on Application ...