Requirements for MPLS Network Actions

1.1. Terminology

• Network Action: An operation to be performed on a packet or as a consequence of a packet being processed by a router. A network action may affect router state, packet forwarding, or it may affect the packet in some other way.¶
• Network Action Indication (NAI): An indication in the packet that a certain network action is to be performed.¶
• Ancillary Data (AD): Data in an MPLS packet associated with a given Network Action that may be used as input to the processing of the Network Action or results from the processing of the Network Action.¶
• In-Stack Data: Ancillary data carried within the MPLS label stack.¶
• Post-Stack Data: Ancillary data carried in a packet between the bottom of the MPLS label stack and the first octet of the user payload. This document does not prescribe whether post-stack data precedes or follows any other protocol structure such as a control word or associated channel header (ACH).¶
• Scope: The set of nodes that should perform a given action.¶

1.2. Background

The MPLS architecture is specified in [RFC3031] and provides a mechanism for forwarding packets through a network without requiring any analysis of the packet payload's network layer header by intermediate nodes (Label Switching Routers - LSRs). Formally, inspection may only occur at network ingress (the Label edge router - LER) where the packet is assigned to a forwarding equivalence class (FEC).¶

MPLS uses switching based on a label pushed on the packet to achieve efficient forwarding and traffic engineering of flows associated with the FEC. While originally used for IP traffic, MPLS has been extended to support point-to-point, point-to-multipoint and multipoint-to-multipoint layer 2 and layer 3 services. An overview of the development of MPLS is provided in [I-D.bryant-mpls-dev-primer].¶

A number of applications have emerged which require LSRs to make forwarding or other processing decisions based on inspection of the network layer header, or some other ancillary information in the protocol stack encapsulated deeper in the packet. An early example of this was generation of a hash of the payload header to be used for load balancing over Equal Cost Multipath (ECMP) or Link Aggregation Group (LAG) next hops. This is based on an assumption that the network layer protocol is IP. MPLS was extended to avoid the need for LSRs to perform this operation if load balancing was needed based on the payload and instead use only the MPLS label stack, using the Entropy Label / Entropy Label Indicator [RFC6790] which are inserted at the LER. Other applications where the intermediate LSRs may need to inspect and process a packet on an LSP include OAM, which can make use of mechanisms such the Router Alert Label [RFC3032] or the Generic Associated Channel Label (GAL) [RFC5586] to indicate that an intercepted packet should be processed locally. See [I-D.bryant-mpls-dev-primer] for detailed list of such applications.¶

There have been a number of new proposals for how network actions and associated ancillary data is to be carried in MPLS and how its presence is indicated to the LSR or egress LER, for example In-situ OAM [I-D.gandhi-mpls-ioam-sr] and Service Function Chaining (SFC) [RFC7665]. A summary of these proposals is contained in [I-D.bryant-mpls-dev-primer], and an overview of use cases is provided in [I-D.ietf-mpls-mna-usecases]. [I-D.song-mpls-extension-header] discusses some of the issues with these proposals (note that this document draws on the requirements and issues without endorsing a specific solution from [I-D.song-mpls-extension-header]):¶

These solutions rely on either the built-in next-protocol
indicator in the header or the knowledge of the format and size of
the header to access the following packet data.  The node is
required to be able to parse the new header, which is unrealistic
in an incremental deployment environment.

A piecemeal solution often assumes the new header is the only
extra header and its location in the packet is fixed by default.
It is impossible or difficult to support multiple new headers in
one packet due to the conflicted assumption.  An example of this
is that the GAL/G-ACH mechanism assumes that if the GAL is
present, only a single G-ACH header follows.

New use cases therefore require the definition of extensions to the MPLS architecture and label stack operations that can be used across these use cases in order to minimise implementation complexity and promote interoperability and extensibility.¶

3.1. General Requirements

1. MPLS combines extensibility, flexibility and efficiency by using control plane context combined with a simple data plane mechanism to allow the network to make forwarding decisions about a packet. Any solution MUST maintain these properties of MPLS.¶
2. Any MNA solutions to these requirements MUST NOT restrict the generality of MPLS architecture [RFC3031], [RFC3032] and [RFC5331].¶
3. If extensions to the MPLS data plane are required, they MUST NOT be inconsistent with the MPLS architecture [RFC3031], [RFC3032] and [RFC5331].¶
4. MNA solutions meeting the requirements set out in this document MUST be able to coexist with and MUST NOT obsolete existing MPLS mechanisms.¶
5. Subject to the constraints in these requirements a solution MAY carry MNA information in-stack, post-stack or both in-stack and post-stack.¶
6. The design of any MNA solution SHOULD be such that an LSR is able to efficiently parse the label stack.¶
7. Any MNA solution specification MUST discuss the ECMP consequences of the design.¶
8. Consistent with MPLS best practise, MNA solutions MUST NOT increase the size of the MPLS label stack more than is necessary.¶
9. MNA solutions that increase the size of the MPLS label stack in a way that is not controlled by the ingress LER MUST discuss the consequences.¶
10. The design of any MNA solution MUST NOT expose confidential information [RFC6973] [RFC3552] to the LSRs.¶
11. Any MNA solution specification MUST document any changes to the existing MPLS data plane security model that it introduces.¶
12. An MNA solution MUST allow NAI-carrying and non-NAI-carrying packets to coexist on the same LSP.¶

3.2. Requirements on the MPLS Network Action Sub-Stack Indicator

1. An MNA solution MUST define how a node determines whether Network Action Indicators (NAIs) are present in the packet.¶
2. If NAIs are required, an MNA solution MUST specify where they are to be placed in the packet i.e. in-stack or post-stack.¶
3. An MNA solution MUST respect the principle that Special Purpose Labels are the mechanism of last resort and therefore must minimise the number of new SPLs that are allocated.¶

3.3. Requirements on Network Action Indicators

1. Insertion, parsing, processing and disposition of NAIs SHOULD make use of existing MPLS data plane operations.¶
2. An NAI MUST NOT be delivered to a node that is not capable of processing the indicated network action in a way that is acceptable to the imposing LER.¶
3. An MNA solution MUST enable a node inserting or modifying NAIs to determine if the far-end LER can accept and process a packet containing a given NAI.¶
4. The NAI design MUST support scoping of network actions.¶
5. A given NAI specification MUST specify if the scope is end-to-end, hop-by-hop, or directed at one or more selected nodes.¶
6. If an MNA solution allows more than one scope, it MUST provide a mechanism to specify the precedence of the scopes.¶
7. An MNA solution SHOULD support NAIs for both P2P and P2MP paths, but any specific NAI MAY only be supported for one or the other.¶
8. An MNA solution defining data plane mechanisms for NAIs MUST be consistent across different control plane protocol types.¶
9. An MNA solution MUST allow the in-use control / management planes to determine the ability of downstream LSRs/LERs to accept/process a given NAI.¶
10. An MNA solution SHOULD allow indicators for multiple network actions in the same packet.¶
11. An MNA solution MUST support the processing of a subset of the NAIs on a packet.¶
12. NAIs SHOULD only be inserted at LERs, and MAY be processed at LSRs and LERs.¶
13. If a network action needs to insert an NAI with in-stack ancillary data at an LSR on an LSP, then the new network action indicator and any required ancillary data MUST be pushed onto the MPLS label stack.¶
14. If a network action needs to insert an NAI with below stack ancillary data at an LSR on an LSP, then the MNA solution specification MUST specify how this is achieved in all circumstances and MUST be consistent with {RFC3031}.¶
15. MPLS network action specifications MUST specify if in-stack or post-stack ancillary data can be rewritten by an LSR.¶
16. An MPLS network action specification MUST specify whether ancillary data is required and whether it is in-stack and/or post-stack.¶
17. Network action indicators MUST be allocated through the IANA process specified in the MNA solution specification.¶
18. Is it RECOMENDED that an MPLS network action specification supports network actions for private use [RFC8126].¶
19. A node removing an NAI MUST NOT leave the MPLS label stack in such a way that downstream nodes are unable to determine the presence of ancillary data remaining in the packet.¶

3.4. Requirements on Ancillary Data

1. Solutions for in-stack ancillary data MUST be able to coexist with and MUST NOT obsolete existing MPLS mechanisms. Such solutions MUST be described in a standards track RFC.¶
2. MNA solutions MUST take care to limit the quantity of in-stack ancillary data to the minimum amount required.¶
3. A common preamble for ancillary data MUST be defined so that a node receiving the ancillary data can determine whether to process, ignore, skip over or discard it according to network or local policies.¶
4. A standardised container MUST be defined for in-stack ancillary data based on the MPLS LSE.¶
5. Any MNA solution specification MUST describe whether it can coexist with existing post-stack data mechanisms e.g. control words and G-ACH, and if so how this coexistence operates.¶
6. An MNA solution MUST allow an LER inserting ancillary data to determine that each node that needs to process the ancillary data can read the required distance into the packet at that node, for example [RFC9088].¶
7. In order to prevent unnecessary scanning of the packet, care needs to be taken in the location of any post stack ancillary data, for example it SHOULD be located as close to the bottom of the label stack as possible.¶
8. Ancillary data MAY be associated with control or maintenance information for traffic carried by an LSP, and/or it MAY be associated with the user traffic itself.¶
9. For scoped ancillary data, any MNA solution MUST allow an LER inserting NAIs whose network actions make use of that ancillary data to determine if the NAI and ancillary data will be processed by LSRs within the scope along the path. Such a solution MAY need to determine if LSRs along the path can process a specific type of AD implied by the NAI at the depth in the stack that it will be presented to the LSR.¶
10. MNA solution specifications MUST specify if the ancillary data needs to be processed as a part of the immediate forwarding operation and whether packet mis-ordering is allowed to occur as a result of the time taken to process the ancillary data. Ed. We think this applies to both NAs and ancillary data and should be generalised.¶
11. A solution MUST be provided to verify the authenticity of ancillary data processed to LSRs [RFC3552].¶
12. The design of the ancillary data MUST NOT expose confidential information [RFC6973] [RFC3552] to the LSRs.¶
13. A mechanism MUST exist to notify an egress LER of the presence of ancillary data so that it can dispose of it appropriately.¶
14. An egress LER MUST NOT forward a packet with ancillary data to a node that is not expecting the ancillary data to be present.¶
15. The size of the ancillary data carried post-stack end-to-end in a packet is a matter for agreement between the ingress and egress PEs, and is not part of these requirements.¶