Abstract: Techniques are described for using route target constraint to filter routes advertised to a node in a seamless Multiprotocol Label Switching (MPLS) network. For example, a first router of a first network may generate a first border gateway protocol (BGP) message to advertise routing information for a first node of the first network, the first BGP message indicating a transport class and specifying an address-specific route target, the transport class comprising one or more tunnels to the first node that share common characteristics. In response to receiving a second BGP message originated by second node of a second network, the second BGP message comprising the address-specific route target, the first router sends the first BGP message to a second router of the second network for sending to the second node to cause the second node to import the routing information.

Abstract: Techniques are described by which a routing protocol, such as border gateway protocol (BGP), is extended to control propagation and importation of information using route targets (RTs) specified as bitmasks that encode link administrative group information. For example, a network control device (e.g., controller) is configured to allocate one or more subset of resources (e.g., nodes or links) of an underlay network to each of one or more virtual networks established over the underlay network. The controller generates a bitmask encoded with link administrative group information of the one or more links. The controller then outputs, to a plurality of provider edge (PE) routers that are participating in a respective virtual network, a routing protocol message to advertise the one or more subset of resources, wherein the routing protocol message includes a route target specified as the bitmask.

Abstract: Techniques are described for selectively pinging certain devices along a segment routing label switched path (LSP) to detect failures in the segment routing LSP. For example, an ingress device comprises one or more processors operably coupled to a memory that are configured to: in response to a request to verify connectivity of a segment routing LSP, configure a FEC stack specifying a stack of segment routing labels for the segment routing LSP; for each of the one or more devices identified from the FEC stack: generate a respective MPLS connectivity request packet for a respective device identified from an outermost FEC of the FEC stack; send the MPLS connectivity request packet to the respective device; receive an MPLS connectivity response packet that verifies connectivity of the respective device; and in response, update the FEC stack by removing the outermost FEC of the FEC stack that identifies the respective device.

Abstract: Techniques are described for providing end-to-end segment routing paths across metropolitan area networks. For example, a method comprises receiving, by an area border router (ABR) connected to one or more metropolitan area networks and a core network, a packet including a segment routing label stack including at least a label of the ABR, a context label associated with a routing instance of the ABR, and a subsequent label identifying a device in the segment routing path, determining, from a lookup of the context label in the metro routing table, a table next hop to the core routing table (or metro routing table); in response to determining the table next hop, determining, from a lookup of the subsequent label in the core routing table (or metro routing table), a next hop in the segment routing path; and sending, by the ABR, the packet toward the device in the segment routing path.

Abstract: Techniques are described for advertising constraint-based path computation (e.g., flexible-algorithm) through a constrained network topology. For example, a network device comprises a memory and one or more programmable processors operably coupled to the memory, wherein the one or more programmable processors are configured to generate a packet including a segment identifier (SID) offset, wherein the SID offset is an offset value associated with the flexible-algorithm. The one or more programmable processors of the network device are also configured to send, to at least one other network device of the plurality of network devices, the SID offset to enable the at least one other network device to derive a node segment identifier for the at least one other network device to participate in the flexible-algorithm.

Abstract: A network device may receive topology data identifying a spine and leaf topology of network devices, and may set link metrics to a common value to generate modified topology data. The network device may remove data identifying connections from leaf network devices to any devices outside the topology from the modified topology data to generate further modified topology data, and may process the further modified topology data, with a model, to determine path data identifying paths to destinations. The network device may determine particular path data identifying shorter paths and longer paths to corresponding destinations, and may determine hop counts associated with the paths. The network device may determine whether the hop counts are all odd values, all even values, or odd and even values, and may perform actions based on whether the hop counts are all odd values, all even values, or odd and even values.

Abstract: Echo or traceroute functionality is supported in a path spanning multiple autonomous systems (ASes) having segment routing (SR) enabled, the path including an ingress node and an egress node, by: (a) obtaining a return label stack to reach the ingress node from either (A) the egress node, or (B) a transit node in the path; (b) obtaining a label stack to reach, from the ingress node, either (A) the egress node, or (B) the transit node; (c) generating a request message including the return label stack; and (d) sending the request message towards either (A) the egress node, or (B) the transit node using the label stack.

Abstract: Ping or traceroute functionality is supported in a path spanning multiple autonomous systems (ASes) having segment routing (SR) enabled, the path including an ingress node in a first autonomous system (AS) and an egress node in an AS other than the first AS, using a reverse path label pair including (1) a node segment identifier (SID) corresponding to an AS Border Router (ASBR) of the second AS (second ASBR), and (2) an egress peer engineering (EPE) SID corresponding to a segment between the second ASBR to an ASBR of the first AS (first ASBR). Responsive to receiving a ping or traceroute request by a router in the second AS, the router generates a ping or traceroute reply including the reverse path label pair. The ping or traceroute reply is forwarded to the second ASBR using the node SID of the reverse path label pair. The ping or traceroute reply is then forwarded from the second ASBR to the first ASBR using the EPE SID of the reverse path label pair.

Abstract: A first network device may determine that a link-state database (LSDB), associated with the first network device, includes a first link-state advertisement (LSA) instance associated with a second network device. The first network device may determine that the first network device has not received a second LSA instance, associated with the second network device, that does not include information identifying a fully adjacent link between the second network device and the first network device. The first network device may receive the second LSA instance associated with the second network device and may transmit, to the second network device, a third LSA instance, associated with the first network device, that includes the information identifying the fully adjacent link between the second network device and the first network device, only after the second LSA instance is received.

Abstract: Techniques are described for advertising constraint-based path computation (e.g., flexible-algorithm) through a constrained network topology. For example, a network device comprises a memory and one or more programmable processors operably coupled to the memory, wherein the one or more programmable processors are configured to generate a packet including a segment identifier (SID) offset, wherein the SID offset is an offset value associated with the flexible-algorithm. The one or more programmable processors of the network device are also configured to send, to at least one other network device of the plurality of network devices, the SID offset to enable the at least one other network device to derive a node segment identifier for the at least one other network device to participate in the flexible-algorithm.

Abstract: In general, various aspects of the techniques described in this disclosure provide a sequence number checksum for link state protocols. In one example, the disclosure describes an apparatus, such as a network device, having a control unit operative to obtain link state information describing links between pairs of the network devices in a network topology, the link state information being fragmented into a plurality of link state protocol (LSP) fragments; compute a sequence number checksum from sequence numbers of the link state protocol (LSP) fragments; receive an LSP data unit from another network device in the network; determine whether a sequence number checksum in the LSP data unit matches a sequence number checksum computed from the link state information; and configure a delay for processing the LSP data unit in response to determining a mismatch between the sequence number checksum of the LSP data unit and the sequence number checksum computed from the link state information.

Abstract: A first device may receive a packet that includes information identifying a path through a network. The first device may configure a header of the packet to include a first set of identifiers that identifies the path and the first device via which the packet was received. The first device may configure the header of the packet to include a second set of identifiers that identifies a set of devices associated with the path. The set of devices may be associated with providing the packet via a network. The first device may determine whether a counter associated with the first set of identifiers has been initialized. The first device may modify a value of the counter to record a metric. The first device may provide the packet to a second device. The first device may perform an action related to the packet or based on the value of the counter.

Abstract: Echo or traceroute functionality is supported in a path spanning multiple autonomous systems (ASes) having segment routing (SR) enabled, the path including an ingress node and an egress node, by: (a) obtaining a return label stack to reach the ingress node from either (A) the egress node, or (B) a transit node in the path; (b) obtaining a label stack to reach, from the ingress node, either (A) the egress node, or (B) the transit node; (c) generating a request message including the return label stack; and (d) sending the request message towards either (A) the egress node, or (B) the transit node using the label stack.

Abstract: Techniques are disclosed for a link state routing protocol adjacency state machine. The adjacency state machine ensures that first and second logical links using different networking protocols are established on a single physical link between two network devices prior to indicating adjacency between the network devices. In some examples, the adjacency state machine determines that both the first and second links are active in response to determining that hello messages are generated by both network devices for both links. In some examples, the adjacency state machine determines that both the first and second logical links are active upon expiration of a predetermined time corresponding to a time required for a duplicate address detection (DAD) operation to complete. In some examples, the first and second logical links use Internet Protocol version 4 (IPv4) and Internet Protocol version 6 (IPv6), respectively.

Abstract: An improved traceroute mechanism for use in a label-switched path (LSP) is provided by (a) receiving, by a device in the LSP, an echo request message, wherein the echo request includes a label stack having a least one label, and wherein each of the at least one label has an associated time-to-live (TTL) value; (b) responsive to receiving the echo request, determining by the device, whether or not the device is a penultimate hop popping (PHP) device for the outermost label of the label stack; and (c) responsive to determining that the device is the PHP device for the outermost label of the label stack, (1) generating an echo reply message corresponding to the echo request message, wherein the echo reply message is encoded to indicate that the device is the PHP device for the outermost label of the label stack, and (2) sending the echo reply message back towards a source of the echo request message.

Abstract: In general, techniques are described by which to provide a scaled end-to-end view of link metrics to integrate multiple non-uniform Interior Gateway Protocol (“IGP”) domains. For example, an Accumulated Interior Gateway Protocol (“AIGP”) attribute, a non-transitive BGP attribute, which includes a link metric assigned to a link within a first IGP domain, is scaled to conform to a metric scale of the second IGP domain. The AIGP attribute may also add link metric assigned to a link within the second IGP domain and may add static metrics of non-IGP links connecting the IGP domains. An IGP domain may set its IGP to the scaled AIGP attribute such that the link metric may include a uniformly scaled end-to-end view of link metrics across the IGP domains. Additionally, a sham-link is assigned a metric value in accordance with the scaling techniques.

Abstract: Techniques are disclosed for a link state routing protocol adjacency state machine. The adjacency state machine ensures that first and second logical links using different networking protocols are established on a single physical link between two network devices prior to indicating adjacency between the network devices. In some examples, the adjacency state machine determines that both the first and second links are active in response to determining that hello messages are generated by both network devices for both links. In some examples, the adjacency state machine determines that both the first and second logical links are active upon expiration of a predetermined time corresponding to a time required for a duplicate address detection (DAD) operation to complete. In some examples, the first and second logical links use Internet Protocol version 4 (IPv4) and Internet Protocol version 6 (IPv6), respectively.

Abstract: An improved traceroute mechanism for use in a label-switched path (LSP) is provided by (a) receiving, by a device in the LSP, an echo request message, wherein the echo request includes a label stack having a least one label, and wherein each of the at least one label has an associated time-to-live (TTL) value; (b) responsive to receiving the echo request, determining by the device, whether or not the device is a penultimate hop popping (PHP) device for the outermost label of the label stack; and (c) responsive to determining that the device is the PHP device for the outermost label of the label stack, (1) generating an echo reply message corresponding to the echo request message, wherein the echo reply message is encoded to indicate that the device is the PHP device for the outermost label of the label stack, and (2) sending the echo reply message back towards a source of the echo request message.

Abstract: A first device may receive a packet that includes information identifying a path through a network. The first device may configure a header of the packet to include a first set of identifiers that identifies the path and the first device via which the packet was received. The first device may configure the header of the packet to include a second set of identifiers that identifies a set of devices associated with the path. The set of devices may be associated with providing the packet via a network. The first device may determine whether a counter associated with the first set of identifiers has been initialized. The first device may modify a value of the counter to record a metric. The first device may provide the packet to a second device. The first device may perform an action related to the packet or based on the value of the counter.

Abstract: The use and processing of update messages (e.g., BGP UPDATEs) that bind (e.g., MPLS) labels to address prefixes is improved such that labels are used more efficiently, and/or such that such update messages can be processed more efficiently. A distance vector control signaling protocol (e.g., BGP) peer device receives a control plane message (e.g., BGP Update) from a downstream peer device, the control plane message including (1) a network address of the downstream device as a next hop value, (2) a prefix value, and (3) at least one label associated with the prefix value. Responsive to receiving the control plane message, the peer device generates a new control plane message including (1) a network address of the peer device as a next hop value, (2) the prefix value from the control plane message, and (3) a label stack including (i) the at least one label from the control plane message, and (ii) a local label associated with the peer device.