I have the following doubt: if an egress router knows how to reach a set of subnets, is it expected to bind all those remote destinations in just a FEC (one label) or has it to bind a label for each of those?
1 Answer
Depends on what you mean by a destination. In IPv4 world, FEC is actually an IPv4 prefix (or a route in other words). Simply put a router will map each prefix it learns via IGP to a label. RFC 3036 describes FECs as follows:
2.1. FECs
It is necessary to precisely specify which packets may be mapped to each LSP. This is done by providing a FEC specification for each
LSP. The FEC identifies the set of IP packets which may be mapped to that LSP.Each FEC is specified as a set of one or more FEC elements. Each FEC element identifies a set of packets which may be mapped to the corresponding LSP. When an LSP is shared by multiple FEC elements,
that LSP is terminated at (or before) the node where the FEC elements can no longer share the same path.Following are the currently defined types of FEC elements. New
element types may be added as needed:1. Address Prefix. This element is an address prefix of any length from 0 to a full address, inclusive. 2. Host Address. This element is a full host address.
-
Now I read and try to understand, However I have this in mind: an egress router knows how to reach x, y and z networks (in IPv4 world). Is it expected to bind the same label to each of those or one different for each of those? My doubt arose when I thought that in the MPLS domain it is not important which is the final destination since it is "ip-free", and therefore it is sufficient to carry the packet to that egress router. Thanks for answering (and forgive my english) Commented Apr 27, 2019 at 15:41
-
In other words, I'm assuming that x, y and z are "indipendent" networks, but reachable by this egress router Commented Apr 27, 2019 at 15:44
-
@Luigi2405, in that case, a router will generate a label per each network– ar_Commented Apr 28, 2019 at 8:28