Internet-Draft | Client Signals YANG Model | August 2024 |
Zheng, et al. | Expires 2 February 2025 | [Page] |
A transport network is a server-layer network to provide connectivity services to its client. The topology and tunnel information in the transport layer has already been defined by generic Traffic- engineered models and technology-specific models (e.g., OTN, WSON). However, how the client signals are accessing to the network has not been described. These information is necessary to both client and provider.¶
This draft describes how the client signals are carried over transport network and defines YANG data models which are required during configuration procedure. More specifically, several client signal (of transport network) models including ETH, STM-n, FC and so on, are defined in this draft.¶
This note is to be removed before publishing as an RFC.¶
The latest revision of this draft can be found at https://ietf-ccamp-wg.github.io/draft-ietf-ccamp-client-signal-yang/draft-ietf-ccamp-client-signal-yang.html. Status information for this document may be found at https://datatracker.ietf.org/doc/draft-ietf-ccamp-client-signal-yang/.¶
Discussion of this document takes place on the Common Control and Measurement Plane Working Group mailing list (mailto:[email protected]), which is archived at https://mailarchive.ietf.org/arch/browse/ccamp/. Subscribe at https://www.ietf.org/mailman/listinfo/ccamp/.¶
Source for this draft and an issue tracker can be found at https://github.com/ietf-ccamp-wg/draft-ietf-ccamp-client-signal-yang.¶
This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.¶
Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet-Drafts is at https://datatracker.ietf.org/drafts/current/.¶
Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress."¶
This Internet-Draft will expire on 2 February 2025.¶
Copyright (c) 2024 IETF Trust and the persons identified as the document authors. All rights reserved.¶
This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Revised BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Revised BSD License.¶
A transport network is a server-layer network designed to provide connectivity services for a client-layer network to carry the client traffic transparently across the server-layer network resources. Currently the topology and tunnel models have been defined for transport networks, including [I-D.ietf-ccamp-otn-topo-yang] and [I-D.ietf-ccamp-otn-tunnel-model], providing server-layer topology abstraction and tunnel configuration between PEs. However, there is a missing piece for configuring how the PEs should map the client- layer traffic, received from the CE, over the server-layer tunnels: this gap is expected to be solved in this document.¶
This document defines a data model of all transport network client signals, using YANG language defined in [RFC7950]. The model can be used by applications exposing to a transport network controller via a RESTconf interface. Furthermore, it can be used by an application for the following purposes (but not limited to):¶
To request/update an end-to-end service by driving a new tunnel to be set up to support this service;¶
To request/update an end-to-end service by using an existing tunnel;¶
To receive notification with regard to the information change of the given service;¶
The YANG modules defined in this document conforms to the Network Management Datastore Architecture (NMDA) defined in [RFC8342].¶
In this document, names of data nodes and other data model objects are prefixed using the standard prefix associated with the corresponding YANG imported modules, including [RFC6991], [RFC8294] and [I-D.ietf-ccamp-otn-tunnel-model], which are shown as follow.¶
Prefix | YANG module | Reference |
---|---|---|
yang | ietf-yang-types | [RFC6991] |
rt-types | ietf-routing-types | [RFC8294] |
te-types | ietf-te-types | [RFCYYYY] |
l1-types | ietf-layer1-types | [RFCZZZZ] |
nw | ietf-network | [RFC8345] |
nt | ietf-network-topology | [RFC8345] |
te | ietf-te | [RFCKKKK] |
clnt-types | ietf-trans-client-svc-types | RFC XXXX |
etht-types | ietf-eth-tran-types | RFC XXXX |
clnt-svc | ietf-trans-client-service | RFC XXXX |
etht-svc | ietf-eth-tran-service | RFC XXXX |
RFC Editor Note: Please replace XXXX with the number assigned to the RFC once this draft becomes an RFC. Please replace YYYY with the RFC numbers assigned to [I-D.ietf-teas-rfc8776-update]. Please replace ZZZZ with the RFC numbers assigned to [I-D.ietf-ccamp-layer1-types]. Please replace KKKK with the RFC numbers assigned to [I-D.ietf-teas-yang-te].¶
A simplified graphical representation of the data model is used in this document. The meaning of the symbols in the YANG data tree presented later in this document is defined in [RFC8340]. They are provided below for reference.¶
Brackets "[" and "]" enclose list keys.¶
Abbreviations before data node names: "rw" means configuration (read-write) and "ro" state data (read-only).¶
Symbols after data node names: "?" means an optional node, "!" means a presence container, and "*" denotes a list and leaf-list.¶
Parentheses enclose choice and case nodes, and case nodes are also marked with a colon (":").¶
Ellipsis ("...") stands for contents of subtrees that are not shown.¶
A global view of a multi-domain service can be described as the Figure 1. The customer is usually responsible to configure the CE nodes and to request to the provider the service intent, from the CE nodes perspective, while the provider is responsible to configure the whole network (including the PE nodes) to support the customer service intent. Generally speaking, the network configurations required to support a customer service can be split into two different groups: CE-PE and PE-PE. The CE-PE configuration deals with the client layer one-hop access link, while PE-PE configuration deals with the server layer tunnel. In Figure 1 we mark the intermediate nodes as 'P', which has same switching capability of PE but just not the 'end-point'. In this example, the link P-P and PE-P are a server-layer intra-domain or inter-domain link.¶
According to the responsibilities of each controller in [RFC8453], the controllers have different views of the service request and network configuration. The duty of CNC is to give the MDSC a description of the customer service intent: candidate YANG models include L1CSM [I-D.ietf-ccamp-l1csm-yang], L2SM [RFC8466] and L3SM [RFC8299], which are classified as customer service models, according to [RFC8309]. These models provide necessary attributes to describe the customer service intent from the customer/CE perspective, and do not provide any specific network configuration. These models also implies that the customer service description can be considered in a separate manner rather than integratig with network configurations, which also enable the controllers to abstract/virtualize the network resource to make them visible to the customer and also easier to manage. In other words, the network knowledge is not necessary at CNC and CMI, which is seen in an abstracted form as shown in Figure 2.¶
The functionalities of MDSC have been described in [RFC8453], which include the customer mapping/translation and multi-domain coordination. By receiving the request from CNC, MDSC need to understand what network configuration can support the customer service intent and turn to the corresponding PNCs for configuration. The service request is therefore decomposed by MDSC into a few network configurations and forwarded to one or multiple PNCs respectively in single-domain and multi-domain scenario. In general, the MDSC has the view of both PE and CE nodes and of some abstract information regarding the P nodes, as shown in Figure 3. It is worth noting that this MDSC view is different with Figure 1 at the intra- domain link. Usually these details are hidden, for scalability purposes, and therefore the MDSC has only an abstract view of each domain internal topology.¶
PNC is the controller that configure the physical devices, based on the network configuration received from the MDSC. Each PNC has the detailed view of its own domain, the example of view from PNC in domain 1 is shown in Figure 4. The PNC has all the detailed topology information on PE and P nodes and on the intra-domain links. The PNC configures the tunnel/tunnel segment within its domain based on the network configuration provided by the MDSC. The PNC also configures the network part of the CE-PE access links as well as the mapping of the client-layer traffic and the server-layer tunnels, based on the network configuration provided by the MDSC. The interaction between PNC and MDSC for the client-layer network configuration is accomplished by the models defined in this draft.¶
Existing TE and technology-specific models, such as topology models and tunnel models, support the network configuration among PEs and Ps. The customer service models, such as L1CSM, L2SM and L3SM, focus on describing the attributes among CEs. However, there is a missing piece on how to configure the CE-PE session. The models defined in this document provide the configuration on CE-PE when the provider server-layer network is TE-based technology.¶
In the example of OTN as the server-layer transport network, a full list of G-PID was summarized in [RFC7139], which can be divided into a few categories. The G-PID signals can be categorized into transparent and non-transparent. Examples of transparent signals may include Ethernetphysical interfaces, FC, STM-n and so on. In this approach the OTN devices is not aware of the client signal type, and this information is only necessary among the controllers. Once the OTN tunnel is set up, there is no switching requested on the client layer, and therefore only signal mapping is needed, without a client tunnel set up. The models that supporting the configuration of transparent signals are defined in Section 4.2. The other category would be non-transparent, such as Carrier Ethernet and MPLS-TP, with a switching request on the client layer. Once the OTN tunnel is set up, a corresponding tunnel in the client layer has to be set up to carry services. The models that supporting the configuration of transparent signals are defined in Section 4.1.¶
It is also worth noting that some client signal can be carried over multiple types of networks. For example, the Ethernet services can be carried over either OTN or Ethernet TE tunnels (over optical or microwave networks). The model specified in this document allows the support from networks with different technologies. The list of identities for client signals are defined in [I-D.ietf-ccamp-layer1-types].¶
Name: The identifier of the client signal service can be configured or retrieved by the name information. The naming of this attribute is based on the requirements of unifying with TE tunnel models. It can be readable or unreadable. If it is readable, to make it unique, it is needed to specify a unified structure for all the PNC systems. It is hard to do that for the reason of a lot of customized requirements from different Operators. In the ICT technology, UUID format could be a good option. Title: The title of client signal service can be configured with its readable name. Description: More detail information of client signal service, such as some administrative information or location information. User-label: An alias of client signal service which is used to tag based on the maintenance requirement.¶
module: ietf-trans-client-service +--rw client-svc +--rw client-svc-instances* [client-svc-name] +--rw client-svc-name string +--rw client-svc-title? string +--rw user-label? string +--rw client-svc-descr? string ................................... module: ietf-eth-tran-service +--rw etht-svc-instances* [etht-svc-name] +--rw etht-svc-name string +--rw etht-svc-title? string +--rw user-label? string +--rw etht-svc-descr? string ...................................¶
In [I-D.ietf-teas-rfc8776-update], it is recognized that if the used topology resource adopts TE modeling, there could be two identifier systems, one is the network modeling system and the other one from TE topology modeling. The access ports of client signal service or end points of Ethernet service also adopt the TE topology resources. So two identifiers are both fine to specified in the configuration requests.¶
module: ietf-trans-client-service +--rw client-svc +--rw client-svc-instances* [client-svc-name] +--rw src-access-ports | +--rw access-node-id? te-types:te-node-id | +--rw access-node-uri? nw:node-id | +--rw access-ltp-id? te-types:te-tp-id | +--rw access-ltp-uri? nt:tp-id | +--rw client-signal? identityref +--rw dst-access-ports +--rw access-node-id? te-types:te-node-id +--rw access-node-uri? nw:node-id +--rw access-ltp-id? te-types:te-tp-id +--rw access-ltp-uri? nt:tp-id +--rw client-signal? identityref ................................... module: ietf-eth-tran-service +--rw etht-svc +--rw etht-svc-instances* [etht-svc-name] +--rw etht-svc-end-points* [etht-svc-end-point-name] +--rw etht-svc-access-points* [access-point-id] +--rw access-point-id string +--rw access-node-id? te-types:te-node-id +--rw access-node-uri? nw:node-id +--rw access-ltp-id? te-types:te-tp-id +--rw access-ltp-uri? nt:tp-id +--rw access-role? identityref ...................................¶
Some PM data are quite critical for client signal service users' experience, e.g. latency value. For the Operators, they can configure a service-oriented threshold for the PM data, to make the maintenance engineer recognize the network issues in time.¶
module: ietf-trans-client-service +--rw client-svc +--rw client-svc-instances* [client-svc-name] +--rw alarm-shreshold +--rw latency-threshold? uint32 ................................... module: ietf-eth-tran-service +--rw etht-svc +--rw etht-svc-instances* [etht-svc-name] +--rw alarm-shreshold +--rw latency-threshold? uint32 ...................................¶
States have been defined to retrieve the status of the delivered services. A few generic states defined in [I-D.ietf-teas-rfc8776-update] are reused in this document. These states include the operational state and the provisioning state.¶
module: ietf-trans-client-service +--rw client-svc +--rw client-svc-instances* [client-svc-name] +--ro operational-state? identityref +--ro provisioning-state? identityref ................................... module: ietf-eth-tran-service +--rw etht-svc +--rw etht-svc-instances* [etht-svc-name] +--ro state +--ro operational-state? identityref +--ro provisioning-state? identityref ...................................¶
A few other parameters are defined for the management of the services. Given the complicated labor division, the creation, update and maintainance may have different responsible systems. So the log of the service would be helpful and should be easy to retrived in a standard way as well. These information include, but not restricted to the following:¶
When the service is created and has been last updated, these are specified in the creation-time and last-updated-time.¶
Who created the service and who made the last update to the service, these are specified in the created-by and last-updated- by.¶
The owner of the service is specifed as owned-by, which would be useful when the service is delegated by or to a specific system. The identifier of the system is used to represent such information.¶
module: ietf-trans-client-service +--rw client-svc +--rw client-svc-instances* [client-svc-name] +--ro creation-time? yang:date-and-time +--ro last-updated-time? yang:date-and-time +--ro created-by? string +--ro last-updated-by? string +--ro owned-by? string ................................... module: ietf-eth-tran-service +--rw etht-svc +--rw etht-svc-instances* [etht-svc-name] +--ro state +--ro creation-time? yang:date-and-time +--ro last-updated-time? yang:date-and-time +--ro created-by? string +--ro last-updated-by? string +--ro owned-by? string ...................................¶
For the Operators, there are also some requirements to provide some PM data to the service users. Especially some PM data are quite critical for users' experience.¶
For example, one of the great advantage of client signal service is it can provide deterministic low latency. Some industry users, e.g. stock trading company, have great demand of stable short latency service to ensure their high-frequency fast trading operations can be finished quickly and definitely. In addition to provision a low- latency client signal service, Operators can also provide some other differentiated services for these kinds of service users. For example latency visibility can be provided to the users, to let them monitor the latency values.¶
The latency measurement mechanism can be referenced from sub-clause 7.10.1 of [ITU-T_G.875]. And PNC can provide a measurement latency after service is provisioned if the devices support this measurement mechanism. Otherwise providing an estimated value could be a workaround solution.¶
module: ietf-trans-client-service +--rw client-svc +--rw client-svc-instances* [client-svc-name] +--ro pm-state +--ro latency? uint32 ................................... module: ietf-eth-tran-service +--rw etht-svc +--rw etht-svc-instances* [etht-svc-name] +--ro state +--ro pm-state +--ro latency? uint32 ...................................¶
Once the processing of client signal service configuration, including creation, updating, deletion, meets with some internal problems, such as no enough available bandwidth, or wrong client signal mapping .etc., the problems should be reported to the client system to make the issues read by the maintenance engineers. Since these error messages are more technologies-specific, especially sometimes accurate location information is needed in these messages, it is hard to standardize these error messages by RESTCONF protocol errors. For the client systems, they would like to get these messages both from notifications and retrieval. So this document defines an error container in the YANG data model to let the PNC to provide these messages.¶
module: ietf-trans-client-service +--rw client-svc +--rw client-svc-instances* [client-svc-name] +--ro error-info +--ro error-code? uint16 +--ro error-description? string +--ro error-timestamp? yang:date-and-time ................................... module: ietf-eth-tran-service +--rw etht-svc +--rw etht-svc-instances* [etht-svc-name] +--rw etht-svc-end-points* [etht-svc-end-point-name] +--ro state +--ro error-info +--ro error-code? uint16 +--ro error-description? string +--ro error-timestamp? yang:date-and-time ...................................¶
This module imports typedefs and modules from [RFC6991], [RFC8294], [I-D.ietf-teas-rfc8776-update].¶
This module references a few documents including [RFC2697], [RFC2698], [RFC4115], [IEEE_802.1ad], [IEEE_802.1q] and [MEF10].¶
This module imports typedefs and modules from [RFC6991], [I-D.ietf-ccamp-otn-tunnel-model], [I-D.ietf-teas-rfc8776-update].¶
This module defines the types for other client signal types.¶
[Note to the RFC Editor - remove this section before publication, as well as remove the reference to RFC [RFC7942].]¶
This section records the status of known implementations of the protocol defined by this specification at the time of posting of this Internet-Draft, and is based on a proposal described in [RFC7942]. The description of implementations in this section is intended to assist the IETF in its decision processes in progressing drafts to RFCs. Please note that the listing of any individual implementation here does not imply endorsement by the IETF. Furthermore, no effort has been spent to verify the information presented here that was supplied by IETF contributors. This is not intended as, and must not be construed to be, a catalog of available implementations or their features. Readers are advised to note that other implementations may exist.¶
According to [RFC7942], "this will allow reviewers and working groups to assign due consideration to documents that have the benefit of running code, which may serve as evidence of valuable experimentation and feedback that have made the implemented protocols more mature. It is up to the individual working groups to use this information as they see fit".¶
The implementation of the CCVPN (Cross Domain and Cross Layer VPN) use-case on ONAP follows the ACTN [RFC8453] architecture. In the design of CCVPN, ONAP presumes the responsibility of the MDSC, and third party network domain controllers the PNCs. Consequently, the ETH Service YANG model is used as the MPI between ONAP and the domain controllers.¶
Organization: China Mobile, Huawei Technologies, etc.¶
Implementation: ONAP CCVPN uses the ETH Service YANG model as the ACTN MPI¶
Description: ONAP CCVPN realizes the E-LINE and E-TREE service on a multi-domain network. Both of the services are modeled on ONAP by the ETH Service YANG model, and the model instances (e.g., JSON objects) are sent between ONAP and the domain controllers. Refer to the following CCVPN wiki for more information: https://wiki.onap.org/display/DW/ CCVPN%28Cross+Domain+and+Cross+Layer+VPN%29+USE+CASE¶
Maturity Level: Prototype¶
Coverage: Partial¶
Contact: [email protected]¶
The data following the model defined in this document is exchanged via, for example, the interface between an orchestrator and a network domain controller.¶
The YANG module defined in this document can be accessed via the RESTCONF protocol defined in [RFC8040], or maybe via the NETCONF protocol [RFC6241].¶
There are a number of data nodes defined in the YANG module which are writable/creatable/deletable (i.e., config true, which is the default). These data nodes may be considered sensitive or vulnerable in some network environments. Write operations (e.g., POST) to these data nodes without proper protection can have a negative effect on network operations.¶
It is proposed that IANA should assign new URIs from the "IETF XML Registry" [RFC3688] as follows:¶
URI: urn:ietf:params:xml:ns:yang:ietf-eth-tran-service Registrant Contact: The IESG XML: N/A; the requested URI is an XML namespace. URI: urn:ietf:params:xml:ns:yang:ietf-trans-client-service Registrant Contact: The IESG XML: N/A; the requested URI is an XML namespace. URI: urn:ietf:params:xml:ns:yang:ietf-eth-tran-types Registrant Contact: The IESG XML: N/A; the requested URI is an XML namespace. URI: urn:ietf:params:xml:ns:yang:ietf-trans-client-svc-types Registrant Contact: The IESG XML: N/A; the requested URI is an XML namespace.¶
This document registers following YANG modules in the YANG Module Names registry [RFC6020].¶
name: ietf-eth-tran-service namespace: urn:ietf:params:xml:ns:yang:ietf-eth-tran-service prefix: ethtsvc reference: RFC XXXX: A YANG Data Model for Transport Network Client Signals name: ietf-eth-tran-types namespace: urn:ietf:params:xml:ns:yang:ietf-eth-tran-types prefix: etht-types reference: RFC XXXX: A YANG Data Model for Transport Network Client Signals name: ietf-trans-client-service namespace: urn:ietf:params:xml:ns:yang:ietf-trans-client-service prefix: clntsvc reference: RFC XXXX: A YANG Data Model for Transport Network Client Signals name: ietf-trans-client-svc-types namespace: urn:ietf:params:xml:ns:yang:ietf-trans-client-svc-types prefix: clntsvc-types reference: RFC XXXX: A YANG Data Model for Transport Network Client Signals¶
We would like to acknowledge the contribution from Francesco Lazzeri to the initial versions of this document, before it has been adopted by CCAMP WG.¶
We would like to thank Igor Bryskin and Daniel King for their comments and discussions.¶