]> Cisco Systems, Inc.

Canada rgandhi@cisco.com

Cisco Systems, Inc.

cfilsfil@cisco.com

Bell Canada

daniel.voyer@bell.ca

Huawei

mach.chen@huawei.com

Colt

Bart.Janssens@colt.net

Nokia

footer.foote@nokia.com

SPRING Working Group Segment Routing (SR) leverages the source routing paradigm. SR is applicable to both Multiprotocol Label Switching (SR-MPLS) and IPv6 (SRv6) data planes. This document describes procedures for Performance Measurement in SR networks using the mechanisms defined in RFC 8762 (Simple Two-Way Active Measurement Protocol (STAMP)) and its optional extensions defined in RFC 8972 and further augmented in draft-ietf-ippm-stamp-srpm. The procedure described is applicable to SR-MPLS and SRv6 data planes and is used for both links and end-to-end SR paths including SR Policies.

Segment Routing (SR) leverages the source routing paradigm and greatly simplifies network operations for Software Defined Networks (SDNs). SR is applicable to both Multiprotocol Label Switching (SR-MPLS) and IPv6 (SRv6) data planes . SR takes advantage of the Equal-Cost Multipaths (ECMPs) between source and transit nodes, between transit nodes and between transit and destination nodes. SR Policies as defined in are used to steer traffic through a specific, user-defined paths using a stack of Segments. Built-in SR Performance Measurement (PM) is one of the essential requirements to provide Service Level Agreements (SLAs). The Simple Two-way Active Measurement Protocol (STAMP) provides capabilities for the measurement of various performance metrics in IP networks without the use of a control channel to pre-signal session parameters. defines optional extensions for STAMP. augments that framework to define STAMP extensions for SR networks. This document describes procedures for Performance Measurement in SR networks using the mechanisms defined in STAMP and its optional extensions defined in and further augmented in . The procedure described is applicable to SR-MPLS and SRv6 data planes and is used for both links and end-to-end SR paths including SR Policies .

The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in when, and only when, they appear in all capitals, as shown here.

BSID: Binding Segment ID. DM: Delay Measurement. ECMP: Equal Cost Multi-Path. HL: Hop Limit. HMAC: Hashed Message Authentication Code. LM: Loss Measurement. MPLS: Multiprotocol Label Switching. NTP: Network Time Protocol. OWAMP: One-Way Active Measurement Protocol. PM: Performance Measurement. PSID: Path Segment Identifier. PTP: Precision Time Protocol. SHA: Secure Hash Algorithm. SID: Segment ID. SL: Segment List. SR: Segment Routing. SRH: Segment Routing Header. SR-MPLS: Segment Routing with MPLS data plane. SRv6: Segment Routing with IPv6 data plane. SSID: STAMP Session Identifier. STAMP: Simple Two-way Active Measurement Protocol. TC: Traffic Class. TTL: Time To Live.

In the Reference Topology shown below, the STAMP Session-Sender R1 initiates a STAMP test packet and the STAMP Session-Reflector R3 transmits a reply test packet. The reply test packet may be transmitted to the STAMP Session-Sender R1 on the same path (same set of links and nodes) or a different path in the reverse direction from the path taken towards the Session-Reflector. The nodes R1 and R3 may be connected via a link or an SR path . The link may be a physical interface, virtual link, or Link Aggregation Group (LAG) , or LAG member link. The SR path may be an SR Policy on node R1 (called head-end) with destination to node R3 (called tail-end).

| | | R1 |=====================| R3 | | |<- - - - - - - - - - | | +-------+ Reply Test Packet +-------+ \ / T4 T3 STAMP Session-Sender STAMP Session-Reflector Reference Topology ]]>

For performance measurement in SR networks, the STAMP Session-Sender and Session-Reflector test packets defined in are used. The STAMP test packets require to be encapsulated to be transmitted on a desired path under measurement. The base STAMP test packets can be encapsulated using IP/UDP header and may use Destination UDP port 862 . In this document, the STAMP packets using IP/UDP header are considered for SR networks. The STAMP test packets are used in one-way, two-way (i.e. round-trip) and loopback measurement modes. Note that one-way and round-trip are referred to in and are further described in this document because of the introduction of loopback measurement mode in SR networks. The procedures defined in this document are also used to infer packet loss in SR networks. The STAMP test packets are transmitted on the same path as the data traffic flow under measurement to measure the delay and packet loss experienced by the data traffic flow. Typically, the STAMP test packets are transmitted along an IP path between a Session-Sender and a Session-Reflector to measure delay and packet loss along that IP path. Matching the forward and reverse direction paths for STAMP test packets, even for directly connected nodes is not guaranteed. It may be desired in SR networks that the same path (same set of links and nodes) between the Session-Sender and Session-Reflector be used for the STAMP test packets in both directions. This is achieved by using the optional STAMP extensions for SR-MPLS and SRv6 networks specified in . The STAMP Session-Reflector uses the return path parameters for the reply test packet from the received STAMP test packet, as described in . This way signaling and maintaining dynamic SR network state for the STAMP sessions on the Session-Reflector are avoided. The optional STAMP extensions defined in are used for direct measurement packet loss in SR networks.

An example of a STAMP reference model with some of the typical measurement parameters including the Destination UDP port for STAMP test session is shown in the following Figure 1:

A Destination UDP port number is selected as described in . The same Destination UDP port can be used for STAMP test sessions for link and end-to-end SR paths. In this case, the Destination UDP port does not distinguish between link or end-to-end SR path measurements. Example of the Timestamp Format is Precision Time Protocol 64-bit truncated (PTPv2) and Network Time Protocol (NTP). By default, the Session-Reflector replies in kind to the timestamp format received in the received Session-Sender test packet, as indicated by the "Z" field in the Error Estimate field as described in . The Session-Reflector mode can be Stateful or Stateless as defined in . Example of Delay Measurement Mode is one-way, two-way (i.e. round-trip) and loopback mode as described in this document. Example of Packet Loss Type can be round-trip, near-end (forward) and far-end (backward) packet loss as defined in . When using the authenticated mode for the STAMP test sessions, the matching Authentication Type (e.g. HMAC-SHA-256) and Key-chain are user-configured on STAMP Session-Sender and STAMP Session-Reflector . The controller shown in the example reference model is not intended for the dynamic signaling of the SR parameters for STAMP test sessions between the STAMP Session-Sender and STAMP Session-Reflector. Note that the YANG data model defined in can be used to provision the STAMP Session-Sender and STAMP Session-Reflector.

The content of an example STAMP Session-Sender test packet using an UDP header is shown in Figure 2. The payload contains the STAMP Session-Sender test packet defined in .

The STAMP Session-Sender test packet as shown in Figure 2 is transmitted over the link under delay measurement. The local and remote IP addresses of the link are used as Source and Destination Addresses, respectively. For IPv6 links, the link local addresses can be used in the IPv6 header. The Session-Sender may use the local Address Resolution Protocol (ARP) table, Neighbor Solicitation or other bootstrap method to find the IP address for the links and refresh. SR encapsulation (e.g. adjacency SID of the link) can be added for transmitting the STAMP test packets for links.

The delay measurement for end-to-end SR path in an SR network is applicable to both end-to-end SR-MPLS and SRv6 paths including SR Policies. The STAMP Session-Sender (the head-end of the SR Policy) IPv4 or IPv6 address MUST be used as the Source Address in the IP header of the test packet. The STAMP Session-Reflector (the SR Policy endpoint) IPv4 or IPv6 address MUST be used as the Destination Address in the IP header of the test packet. In the case of Color-Only Destination Steering, with IPv4 endpoint of 0.0.0.0 or IPv6 endpoint of ::0 , the loopback address from the range 127/8 for IPv4, or the loopback address ::1/128 for IPv6 is used as the Session-Reflector Address, respectively.

An SR-MPLS Policy may contain a number of Segment Lists (SLs). A STAMP Session-Sender test packet MUST be transmitted for each Segment List of the SR-MPLS Policy. The content of an example STAMP Session-Sender test packet for an end-to-end SR-MPLS Policy is shown in Figure 3.

The Segment List can be empty in case of a single-hop SR-MPLS Policy with Implicit NULL label. The Path Segment Identifier (PSID) of an SR-MPLS Policy can be carried in the MPLS header as shown in Figure 3, and can be used for direct measurement as described in Section 6, titled "Direct Measurement for Links and SR Paths".

An SRv6 Policy may contain a number of Segment Lists. A STAMP Session-Sender test packet MUST be transmitted for each Segment List of the SRv6 Policy. An SRv6 Policy can contain an SRv6 Segment Routing Header (SRH) carrying a Segment List as described in . The content of an example STAMP Session-Sender test packet for an end-to-end SRv6 Policy is shown in Figure 4. The SRv6 network programming is described in . The procedure defined for Upper-Layer Header processing for SRv6 End SIDs in Section 4.1.1 in is used to process the IPv6/UDP header in the received test packets on the Session-Reflector.

. . . +---------------------------------------------------------------+ | UDP Header | . Source Port = As chosen by Session-Sender . . Destination Port = User-configured Destination Port | 862 . . . +---------------------------------------------------------------+ | Payload = Test Packet as specified in Section 3 of RFC 8972 | . in Figure 1 and Figure 3 . . . +---------------------------------------------------------------+ ]]>

The Segment List (SL) may be empty and no SRH may be carried. The Path Segment Identifier (PSID) of the SRV6 Policy can be carried in the SRH as shown in Figure 4 and can be used for direct measurement as described in Section 6, titled "Direct Measurement for Links and SR Paths".

The STAMP Session-Reflector reply test packet uses the IP/UDP information from the received test packet as shown in Figure 5.

In one-way delay measurement mode, a reply test packet as shown in Figure 5 is transmitted by the STAMP Session-Reflector, for both links and end-to-end SR Policies. The reply test packet may be transmitted on the same path or a different path in the reverse direction. The STAMP Session-Sender address may not be reachable via IP route from the STAMP Session-Reflector. The STAMP Session-Sender in this case MUST send its reachability path information to the STAMP Session-Reflector using the Return Path TLV defined in . In this mode, as per Reference Topology, all timestamps T1, T2, T3, and T4 are collected by the test packets. However, only timestamps T1 and T2 are used to measure one-way delay as (T2 - T1). The one-way delay measurement mode requires the clock on the Session-Sender and Session-Reflector to be synchronized.

In two-way (i.e. round-trip) delay measurement mode, a reply test packet as shown in Figure 5 is transmitted by the STAMP Session-Reflector on the same path in the reverse direction, e.g. on the reverse direction link or associated reverse SR path . For two-way delay measurement mode for links, the STAMP Session-Reflector transmits the reply test packet on the same link where the test packet is received. The STAMP Session-Sender can request in the test packet to the STAMP Session-Reflector to transmit the reply test packet back on the same link using the Control Code Sub-TLV in the Return Path TLV defined in . For two-way delay measurement mode for end-to-end SR paths, the STAMP Session-Reflector transmits the reply test packet on a specific reverse path. The STAMP Session-Sender can request in the test packet to the STAMP Session-Reflector to transmit the reply test packet back on a given reverse path using a Segment List sub-TLV in the Return Path TLV defined in . In this mode, as per Reference Topology, all timestamps T1, T2, T3, and T4 are collected by the test packets. All four timestamps are used to measure two-way delay as ((T4 - T1) - (T3 - T2)). When clock synchronization on the Session-Sender and Session-Reflector nodes is not possible, the one-way delay can be derived using two-way delay divided by two.

The content of an example STAMP Session-Reflector reply test packet transmitted on the same path as the data traffic flow under measurement for two-way delay measurement of an end-to-end SR-MPLS Policy is shown in Figure 6.

The content of an example STAMP Session-Reflector reply test packet transmitted on the same path as the data traffic flow under measurement for two-way delay measurement of an end-to-end SRv6 Policy with SRH is shown in Figure 7. The procedure defined for Upper-Layer Header processing for SRv6 End SIDs in Section 4.1.1 in is used to process the IPv6/UDP header in the received reply test packets on the Session-Sender.

. . . +---------------------------------------------------------------+ | UDP Header | . Source Port = As chosen by Session-Reflector . . Destination Port = Source Port from Received Test Packet . . . +---------------------------------------------------------------+ | Payload = Test Packet as specified in Section 3 of RFC 8972 | . in Figure 2 and Figure 4 . . . +---------------------------------------------------------------+ ]]>

The STAMP Session-Sender test packets are transmitted in loopback mode to measure loopback delay of a bidirectional circular path. In this mode, the received Session-Sender test packets are not punted out of the fast path in forwarding (i.e. to slow path or control-plane) at the STAMP Session-Reflector. In other words, the Session-Reflector does not process them and generate Session-Reflector test packets. This is a new measurement mode, not defined by STAMP process . The STAMP Session-Sender MUST set the Destination UDP port to the UDP port it uses to receive the reply STAMP test packets. Since the Session-Reflector does not support the STAMP process, the loopback function simply makes the necessary changes to the encapsulation including IP and UDP headers to return the test packet to the Session-Sender. The typical Session-Reflector test packet is not used in this mode. The loopback function simply returns the received Session-Sender test packet to the Session-Sender without STAMP modifications defined in . In case of SR-MPLS paths, the SR-MPLS header can contain the MPLS label stack of the forward path or both forward and the reverse paths. The IP header of the STAMP Session-Sender test packets MUST set the Destination Address equal to the STAMP Session-Sender address and the Source Address equal to the STAMP Session-Reflector address. In case of SRv6 paths, the SRH can contain the Segment List of the forward path or both forward and the reverse paths. In the former case, an inner IPv6 header (after SRH and before UDP header) MUST be added that contains the Destination Address equal to the STAMP Session-Sender address and the Source Address equal to the STAMP Session-Reflector address. The Session-Sender may use the SSID field in the received reply test packet or local configuration to identify its test session using the loopback mode. In the received Session-Sender test packet at the Session-Sender, the 'Session-Sender Sequence Number', 'Session-Sender Timestamp', 'Session-Sender Error Estimate', and 'Session-Sender TTL' fields are not present in this mode. In this mode, as per Reference Topology, the test packet received back at the Session-Sender retrieves the timestamp T1 from the test packet and adds the received timestamp T4 locally. Both these timestamps are used to measure the loopback delay as (T4 - T1). The one-way delay can be derived using the loopback delay divided by two. In loopback mode, the loopback delay includes the processing delay on the Session-Reflector. The Session-Reflector processing delay component includes only the time required to loop the test packet from the incoming interface to the outgoing interface in forwarding plane.

The Point-to-Multipoint (P2MP) SR path that originates from a root node terminates on multiple destinations called leaf nodes (e.g. P2MP SR Policy ). The procedures for delay and loss measurement described in this document for end-to-end P2P SR Policies are also equally applicable to the P2MP SR Policies. The procedure for one-way measurement is defined as following: The STAMP Session-Sender root node transmits test packets using the Tree-SID defined in for the P2MP SR-MPLS Policy as shown in Figure 8. The STAMP Session-Sender test packets may contain the replication SID as defined in . The Destination Address MUST be set to the loopback address from the range 127/8 for IPv4, or the loopback address ::1/128 for IPv6. Each STAMP Session-Reflector leaf node MUST transmit its node address in the Source Address of the reply test packets shown in Figure 5. This allows the STAMP Session-Sender root node to identify the STAMP Session-Reflector leaf nodes of the P2MP SR Policy. The P2MP root node measures the delay for each P2MP leaf node individually.

The considerations for two-way mode for P2MP SR Policy (e.g. for co-routed bidirectional SR-MPLS path) are outside the scope of this document.

The processing rules described in this section are applicable to the STAMP test packets for links and end-to-end SR paths including SR Policies.

The TTL field in the IPv4 and MPLS headers of the STAMP Session-Sender and STAMP Session-Reflector test packet is set to 255 as per Generalized TTL Security Mechanism (GTSM) .

The Hop Limit (HL) field in the IPv6 and SRH headers of the STAMP Session-Sender and STAMP Session-Reflector test packet is set to 255 as per Generalized TTL Security Mechanism (GTSM) .

The Router Alert IP option (RAO) is not set in the STAMP test packets for links and end-to-end SR paths.

For IPv4 test packets, where the hardware is not capable of re-computing the UDP checksum or adding checksum complement , the Session-Sender may set the UDP checksum value to 0 . For IPv6 test packets, where the hardware is not capable of re-computing the UDP checksum or adding checksum complement , the Session-Sender and Session-Reflector may use the procedure defined in for the UDP checksum.

The procedure described in Section 4 for delay measurement using STAMP test packets can be used to detect (test) packet loss for links and end-to-end SR paths. The Sequence Number field in the STAMP test packet is used as described in Section 4 "Theory of Operation" where Stateful and Stateless Session-Reflector operations are defined , to detect round-trip, near-end (forward) and far-end (backward) packet loss. In the case of the loopback mode introduced in this document, only the round-trip packet loss is applicable. This method can be used for inferred packet loss measurement, however, it provides only approximate view of the data packet loss.

The STAMP "Direct Measurement" TLV (Type 5) defined in can be used in SR networks for data packet loss measurement. The STAMP test packets with this TLV are transmitted using the procedures described in Section 4 to collect the transmit and receive counters of the data flow for the links and end-to-end SR paths. The PSID carried in the received data packet for the traffic flow under measurement can be used to measure receive data packets (for receive traffic counter) for an end-to-end SR path on the STAMP Session-Reflector. The PSID in the received Session-Sender test packet header can be used to associate the receive traffic counter on the Session-Reflector for the end-to-end SR path. The STAMP "Direct Measurement" TLV (Type 5) lacks the support to identify the Block Number of the Direct Measurement traffic counters, which is required for Alternate-Marking Method for accurate data packet loss metric.

The STAMP test session state allows to know if the performance measurement test is active. The threshold-based notification may not be generated if the delay values do not change significantly. For an unambiguous monitoring, the controller needs to distinguish the cases whether the performance measurement is active, or delay values are not changing to cross threshold. The STAMP test session state initially is declared active when one or more reply test packets are received at the STAMP Session-Sender. The STAMP test session state is declared idle (or failed) when consecutive N number of reply test packets are not received at the STAMP Session-Sender, where N is locally provisioned value.

An SR Policy can have ECMPs between the source and transit nodes, between transit nodes and between transit and destination nodes. Usage of Anycast SID by an SR Policy can result in ECMP paths via transit nodes part of that Anycast group. The test packets SHOULD be transmitted to traverse different ECMP paths to measure end-to-end delay of an SR Policy. Forwarding plane has various hashing functions available to forward packets on specific ECMP paths. The mechanisms described in and for handling ECMPs are also applicable to the delay measurement. For SR-MPLS Policy, sweeping of MPLS entropy label values can be used in Session-Sender test packets and Session-Reflector test packets to take advantage of the hashing function in forwarding plane to influence the ECMP path taken by them. In IPv4 header of the STAMP Session-Sender test packets, sweeping of Session-Reflector Address from the range 127/8 can be used to exercise ECMP paths. In this case, both the forward and the return paths MUST be SR-MPLS paths when using the loopback mode. As specified in , Flow Label field in the outer IPv6 header can also be used for sweeping to exercise different IPv6 ECMP paths. The "Destination Node Address" TLV MUST be carried in the STAMP Session-Sender test packet to identify the intended Session-Reflector, when using IPv4 Session-Reflector Address from 127/8 range for a P2P SR Policy, when the STAMP test packet is encapsulated by a tunneling protocol or an MPLS Segment List.

The performance measurement is intended for deployment in well-managed private and service provider networks. As such, it assumes that a node involved in a measurement operation has previously verified the integrity of the path and the identity of the far-end STAMP Session-Reflector. If desired, attacks can be mitigated by performing basic validation and sanity checks, at the STAMP Session-Sender, of the counter or timestamp fields in received measurement reply test packets. The minimal state associated with these protocols also limits the extent of measurement disruption that can be caused by a corrupt or invalid packet to a single test cycle. Use of HMAC-SHA-256 in the authenticated mode protects the data integrity of the test packets. SRv6 has HMAC protection authentication defined for SRH . Hence, test packets for SRv6 may not need authentication mode. Cryptographic measures may be enhanced by the correct configuration of access-control lists and firewalls. The security considerations specified in and also apply to the procedures described in this document. The Security Considerations specified in are also equally applicable to the procedures defined in this document. When using the procedures defined in [RFC6936], the security considerations specified in also apply.

This document does not require any IANA action.

&RFC0768; &RFC2119; &RFC8174; &RFC8762; &RFC8972; &I-D.ietf-ippm-stamp-srpm; IEEE &RFC2113; &RFC4291; &RFC5082; &RFC5884; &RFC6437; &RFC6790; &RFC6936; &RFC7404; &RFC7820; &RFC8029; &RFC8321; &RFC8085; &RFC8402; &RFC8754; &RFC8986; &I-D.ietf-spring-segment-routing-policy; &I-D.ietf-spring-sr-replication-segment; &I-D.ietf-pim-sr-p2mp-policy; &I-D.ietf-spring-mpls-path-segment; &I-D.ietf-spring-srv6-path-segment; &I-D.ietf-pce-sr-bidir-path; &I-D.ietf-ippm-stamp-yang; IEEE Std. 802.1AX

The authors would like to thank Thierry Couture for the discussions on the use-cases for Performance Measurement in Segment Routing. The authors would also like to thank Greg Mirsky, Gyan Mishra, Xie Jingrong, and Mike Koldychev for reviewing this document and providing useful comments and suggestions. Patrick Khordoc and Radu Valceanu have helped improve the mechanisms described in this document.