Encapsulation for MPLS Performance Measurement with the Alternate-Marking Method

Encapsulation for MPLS Performance Measurement with the Alternate-Marking Method China Mobile

Beijing China chengweiqiang@chinamobile.com

ZTE Corp.

Nanjing China xiao.min2@zte.com.cn

Huawei

Beijing China zhoutianran@huawei.com

FiberHome

Wuhan China djy@fiberhome.com

Broadcom

United States of America yoav.peleg@broadcom.com

RTG mpls Flow-ID Label Indicator Flow-ID Label This document defines the encapsulation for MPLS performance measurement with the Alternate-Marking Method, which performs flow-based packet loss, delay, and jitter measurements on MPLS traffic.

Introduction describes a performance measurement method, which can be used to measure packet loss, delay, and jitter on data traffic. Since this method is based on marking consecutive batches of packets, it is referred to as the Alternate-Marking Method. outlines key considerations for developing a solution for MPLS flow identification, intended for use in performance monitoring of MPLS flows. This document defines the encapsulation for MPLS performance measurement with the Alternate-Marking Method, which performs flow-based packet loss, delay, and jitter measurements on the MPLS traffic. The encapsulation defined in this document supports performance monitoring at the intermediate nodes and MPLS flow identification at both transport and service layers. Note that, at the time of writing, there is ongoing work on MPLS Network Actions (MNAs) . The MPLS performance measurement with the Alternate-Marking Method can also be achieved by MNA encapsulation. In addition, MNA will provide a broader use-case applicability. That means the MNA encapsulation is expected to provide a more advanced solution. The MPLS Working Group has agreed that this document will be made Historic when that solution is published as an RFC.

Conventions Used in This Document

Abbreviations

ACL:: Access Control List
BoS:: Bottom of Stack
cSPL:: Composite Special Purpose Label, the combination of the Extension Label (value 15) and an Extended Special Purpose Label
DSCP:: Differentiated Services Code Point
ELC:: Entropy Label Capability
ERLD:: Entropy Readable Label Depth
eSPL:: Extended Special Purpose Label, a special-purpose label that is placed in the label stack after the Extension Label (value 15)
FL:: Flow-ID Label
FLC:: Flow-ID Label Capability
FLI:: Flow-ID Label Indicator
FRLD:: Flow-ID Readable Label Depth
IPFIX:: IP Flow Information Export
LSP:: Label Switched Path
LSR:: Label Switching Router
MPLS:: Multi-Protocol Label Switching
NMS:: Network Management System
PHP:: Penultimate Hop Popping
PM:: Performance Measurement
PW:: Pseudowire
SFL:: Synonymous Flow Label
SID:: Segment ID
SR:: Segment Routing
TC:: Traffic Class
TTL:: Time to Live
VC:: Virtual Channel
VPN:: Virtual Private Network
XL:: Extension Label

Requirements Language The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 when, and only when, they appear in all capitals, as shown here.

Flow-Based PM Encapsulation in MPLS This document defines the Flow-based MPLS performance measurement encapsulation with the Alternate-Marking Method, as shown in .

Flow-based PM Encapsulation in MPLS The Flow-ID Label Indicator (FLI) is an Extended Special Purpose Label (eSPL), which is combined with the Extension Label (XL, value 15) to form a Composite Special Purpose Label (cSPL), as defined in . The FLI is defined in this document as value 18. The Traffic Class (TC) and Time To Live (TTL) fields of the XL and FLI MUST use the same values of the label immediately preceding the XL. The Bottom of the Stack (BoS) bit for the XL and FLI MUST be zero. If any XL or FLI processed by a node has the BoS bit set, the node MUST discard the packet and MAY log an error. The Flow-ID Label (FL) is used as an MPLS flow identification . Its value MUST be unique within the administrative domain. The FL values MAY be allocated by an external NMS or controller based on the measurement object instances (such as LSP or PW). There is a one-to-one mapping between a Flow-ID and a flow. The specific method on how to allocate the FL values is described in . The FL, preceded by a cSPL, can be placed either at the bottom or in the middle, but not at the top, of the MPLS label stack, and it MAY appear multiple times within a label stack. of this document provides several examples to illustrate the application of FL in a label stack. The TTL for the FL MUST be zero to ensure that it is not used inadvertently for forwarding. The BoS bit for the FL depends on whether the FL is placed at the bottom of the MPLS label stack, i.e., the BoS bit for the FL is set only when the FL is placed at the bottom of the MPLS label stack. Besides the flow identification, a color-marking field is also necessary for the Alternate-Marking Method. To color the MPLS traffic and to distinguish between hop-by-hop measurement and edge-to-edge measurement, the TC for the FL is defined as follows:

L(oss) bit is used for coloring the MPLS packets for loss measurement. Setting the bit means color 1, and unsetting the bit means color 0.
D(elay) bit is used for coloring the MPLS packets for delay/jitter measurement. Setting the bit means color for delay measurement.
T(ype) bit is used to indicate the measurement type. When the T bit is set to 1, that means edge-to-edge performance measurement. When the T bit is set to 0, that means hop-by-hop performance measurement.

Considering the FL is not used as a forwarding label, the repurposing of the TC for the FL is feasible and viable.

Examples for Applying Flow-ID Label in a Label Stack Three examples of different layouts of the FL (4 octets) are illustrated as follows. Note that more examples may exist.

Layout of the Flow-ID Label when Applied to MPLS Transport

Applying Flow-ID to MPLS Transport With penultimate hop popping (PHP ), the top label is "popped at the penultimate LSR of the LSP, rather than at the LSP Egress". The final bullet of of the present document requires that "[t]he processing node MUST pop the XL, FLI, and FL from the MPLS label stack when it needs to pop the preceding forwarding label", which implies that the penultimate Label Switching Router (LSR) needs to follow the requirement of in order to support this specification. If this is done, the egress LSR is excluded from the performance measurement. Therefore, when this specification is in use, PHP should be disabled, unless the penultimate LSR is known to have the necessary support and unless it's acceptable to exclude the egress LSR. Also note that in other examples of applying Flow-ID to MPLS transport, one LSP label can be substituted by multiple SID labels in the case of using SR Policy, and the combination of cSPL and FL can be placed between SID labels, as specified in .

Layout of the Flow-ID Label when Applied to MPLS Service

Applying Flow-ID to MPLS Service Note that in this case, the application label can be an MPLS PW label, MPLS Ethernet VPN label, or MPLS IP VPN label, and it is also called a VC label as defined in .

Layout of the Flow-ID Label when Applied to both MPLS Transport and MPLS Service

Applying Flow-ID to both MPLS Transport and MPLS Service Note that for this example, the two FL values appearing in a label stack must be different. In other words, the FL applied to the MPLS transport and the FL applied to the MPLS service must be different. Also, note that the two FL values are independent of each other. For example, two packets can belong to the same VPN flow but different LSP flows, or two packets can belong to different VPN flows but the same LSP flow.

Procedures of Encapsulation, Look-Up, and Decapsulation The procedures for FL encapsulation, look-up, and decapsulation are summarized as follows:

The MPLS ingress node inserts the XL, FLI, and FL into the MPLS label stack. At the same time, the ingress node sets the FL value, the two color-marking bits, and the T bit, as defined in .
If edge-to-edge measurement is applied, i.e., the T bit is set to 1, then only the MPLS ingress/egress node is the processing node; otherwise, all the MPLS nodes along the LSP are the processing nodes. The processing node looks up the FL with the help of the XL and FLI, and exports the collected data (such as the Flow-ID, block counters, and timestamps) to an external NMS/controller, referring to the Alternate-Marking Method. Section 6 of describes protocols for collected data export; the details on how to export the collected data are outside the scope of this document. Note that while looking up the FL, the transit node needs to inspect beyond the label at the top of the label stack used to make forwarding decisions.
The processing node MUST pop the XL, FLI, and FL from the MPLS label stack when it needs to pop the preceding forwarding label. The egress node MUST pop the whole MPLS label stack. This document doesn't introduce any new process to the decapsulated packet.

Procedures of Flow-ID Allocation There are at least two ways of allocating Flow-ID. One way is to allocate Flow-ID by a manual trigger from the network operator, and the other way is to allocate Flow-ID by an automatic trigger from the ingress node. Details are as follows:

In the case of a manual trigger, the network operator manually inputs the characteristics (e.g., IP five tuples and IP DSCP) of the measured flow; then the NMS/controller generates one or two Flow-IDs based on the input from the network operator and provisions the ingress node with the characteristics of the measured flow and the corresponding allocated Flow-ID(s).
In the case of an automatic trigger, the ingress node identifies the flow entering the measured path and exports the characteristics of the identified flow to the NMS/controller by IPFIX ; then the NMS/controller generates one or two Flow-IDs based on the characteristics exported from the ingress node and provisions the ingress node with the characteristics of the identified flow and the corresponding allocated Flow-ID(s).

The policy preconfigured at the NMS/controller decides whether one Flow-ID or two Flow-IDs are generated. If the performance measurement on the MPLS service is enabled, then one Flow-ID applied to the MPLS service is generated. If the performance measurement on the MPLS transport is enabled, then one Flow-ID applied to the MPLS transport is generated. If both of them are enabled, then two Flow-IDs are respectively applied to the MPLS service and the MPLS transport are generated. In this case, a transit node needs to look up both of the two Flow-IDs by default. However, this behavior can be changed through configuration, such as by setting it to look up only the Flow-ID applied to the MPLS transport. Whether using the two methods mentioned above or other methods to allocate Flow-ID, the NMS/controller MUST ensure that every generated Flow-ID is unique within the administrative domain and MUST NOT have any value in the reserved label space (0-15) . Specifically, the statement of "Flow-ID is unique" means that the values of Flow-ID are distinct and non-redundant for any flow at any given time within an administrative domain, such that no two flows share the same Flow-ID. This uniqueness ensures that each flow can be individually identified, tracked, and differentiated from others for accurate performance monitoring and management.

FLC and FRLD Considerations Analogous to the Entropy Label Capability (ELC) defined in and the Entropy Readable Label Depth (ERLD) defined in , the Flow-ID Label Capability (FLC) and the Flow-ID Readable Label Depth (FRLD) are defined in this document. Both FLC and FRLD have similar semantics with the ELC and ERLD to a router, except that the Flow-ID is used in its flow identification function while the Entropy is used in its load-balancing function. The ingress node MUST insert each FL at an appropriate depth, which ensures the node to which the FL is exposed has the FLC. The ingress node SHOULD insert each FL within an appropriate FRLD, which is the minimum FRLD of all the on-path nodes that need to read and use the FL in question. How the ingress node knows the FLC and FRLD of all the on-path nodes is outside the scope of this document. When the SR paths are used for transport, the label stack grows as the number of on-path segments increases. If the number of on-path segments is high, that may become a challenge for the FL to be placed within an appropriate FRLD. To overcome this potential challenge, an implementation MAY allow the ingress node to place FL between SID labels. This means that multiple identical FLs at different depths MAY be interleaved with SID labels. When this occurs, sophisticated network planning may be needed, which is beyond the scope of this document.

Equal-Cost Multipath Considerations Analogous to what's described in , under conditions of equal-cost multipath, the introduction of the FL may lead to the same problem that is caused by the Synonymous Flow Label (SFL) . The two solutions proposed for SFL also apply here. Specifically, adding FL to an existing flow may cause that flow to take a different path. If the operator expects to resolve this problem, they can choose to apply entropy labels or add FL to all flows.

Security Considerations As specified in , "for security reasons, the Alternate-Marking Method MUST only be applied to controlled domains." This requirement applies when the MPLS performance measurement with Alternate-Marking Method is taken into account, which means the MPLS encapsulation and related procedures defined in this document MUST only be applied to controlled domains; otherwise, the potential attacks discussed in may be applied to the deployed MPLS networks. As specified in , the value of an FL MUST be unique within the administrative domain. In other words, the administrative domain is the scope of an FL. The method for achieving multi-domain performance measurement with the same FL is outside the scope of this document. The FL MUST NOT be signaled and distributed outside the administrative domain. Improper configuration that allows the FL to be passed from one administrative domain to another would result in Flow-ID conflicts. To prevent packets carrying FLs from leaking from one domain to another, domain boundary nodes MUST deploy policies (e.g., ACL) to filter out these packets. Specifically, at the sending edge, the domain boundary node MUST filter out the packets that carry the FLI and are sent to other domains. At the receiving edge, the domain boundary node MUST drop the packets that carry the FLI and are from other domains. Note that packet leakage is neither breaching privacy nor a source of DoS.

IANA Considerations IANA has assigned the following value in the "Extended Special-Purpose MPLS Label Values" registry within the "Special-Purpose Multiprotocol Label Switching (MPLS) Label Values" registry group: New Extended Special-Purpose MPLS Label Value for Flow-ID Label Indicator

Value	Description	Reference
18	Flow-ID Label Indicator (FLI)	RFC 9714

References Normative References Informative References

Acknowledgements The authors acknowledge , , , , , , , , , , , , , , , , , , , , , , and for their careful review and very helpful comments. They also acknowledge and for their insightful MPLS-RT review and constructive comments. Additionally, the authors thank for the English grammar review.

Contributors China Mobile

wangminxue@chinamobile.com

China Mobile

yewen@chinamobile.com