idnits 2.17.1 draft-ietf-forces-model-15.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- ** It looks like you're using RFC 3978 boilerplate. You should update this to the boilerplate described in the IETF Trust License Policy document (see https://trustee.ietf.org/license-info), which is required now. -- Found old boilerplate from RFC 3978, Section 5.1 on line 16. -- Found old boilerplate from RFC 3978, Section 5.5, updated by RFC 4748 on line 5982. -- Found old boilerplate from RFC 3979, Section 5, paragraph 1 on line 5993. -- Found old boilerplate from RFC 3979, Section 5, paragraph 2 on line 6000. -- Found old boilerplate from RFC 3979, Section 5, paragraph 3 on line 6006. Checking nits according to https://www.ietf.org/id-info/1id-guidelines.txt: ---------------------------------------------------------------------------- No issues found here. Checking nits according to https://www.ietf.org/id-info/checklist : ---------------------------------------------------------------------------- ** The abstract seems to contain references ([2], [4]), which it shouldn't. Please replace those with straight textual mentions of the documents in question. == There are 1 instance of lines with non-RFC6890-compliant IPv4 addresses in the document. If these are example addresses, they should be changed. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust Copyright Line does not match the current year == The document seems to lack the recommended RFC 2119 boilerplate, even if it appears to use RFC 2119 keywords. (The document does seem to have the reference to RFC 2119 which the ID-Checklist requires). -- The document seems to lack a disclaimer for pre-RFC5378 work, but may have content which was first submitted before 10 November 2008. If you have contacted all the original authors and they are all willing to grant the BCP78 rights to the IETF Trust, then this is fine, and you can ignore this comment. If not, you may need to add the pre-RFC5378 disclaimer. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (September 11, 2008) is 5705 days in the past. Is this intentional? Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) -- Looks like a reference, but probably isn't: 'N' on line 3887 -- Looks like a reference, but probably isn't: '1-9' on line 4070 -- Looks like a reference, but probably isn't: '0-9' on line 4070 -- No information found for draft-ietf - is the name correct? -- Possible downref: Normative reference to a draft: ref. '2' Summary: 2 errors (**), 0 flaws (~~), 3 warnings (==), 12 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Working Group: ForCES J. Halpern 3 Internet-Draft Self 4 Intended status: Standards Track J. Hadi Salim 5 Expires: March 15, 2009 Znyx Networks 6 September 11, 2008 8 ForCES Forwarding Element Model 9 draft-ietf-forces-model-15.txt 11 Status of this Memo 13 By submitting this Internet-Draft, each author represents that any 14 applicable patent or other IPR claims of which he or she is aware 15 have been or will be disclosed, and any of which he or she becomes 16 aware will be disclosed, in accordance with Section 6 of BCP 79. 18 Internet-Drafts are working documents of the Internet Engineering 19 Task Force (IETF), its areas, and its working groups. Note that 20 other groups may also distribute working documents as Internet- 21 Drafts. 23 Internet-Drafts are draft documents valid for a maximum of six months 24 and may be updated, replaced, or obsoleted by other documents at any 25 time. It is inappropriate to use Internet-Drafts as reference 26 material or to cite them other than as "work in progress." 28 The list of current Internet-Drafts can be accessed at 29 http://www.ietf.org/ietf/1id-abstracts.txt. 31 The list of Internet-Draft Shadow Directories can be accessed at 32 http://www.ietf.org/shadow.html. 34 This Internet-Draft will expire on March 15, 2009. 36 Comments are solicited and should be addressed to the working group's 37 mailing list at forces@peach.ease.lsoft.com and/or the author(s). 39 Abstract 41 This document defines the forwarding element (FE) model used in the 42 Forwarding and Control Element Separation (ForCES) protocol [2]. The 43 model represents the capabilities, state and configuration of 44 forwarding elements within the context of the ForCES protocol, so 45 that control elements (CEs) can control the FEs accordingly. More 46 specifically, the model describes the logical functions that are 47 present in an FE, what capabilities these functions support, and how 48 these functions are or can be interconnected. This FE model is 49 intended to satisfy the model requirements specified in the ForCES 50 requirements document, RFC3654 [4]. 52 Table of Contents 54 1. Definitions . . . . . . . . . . . . . . . . . . . . . . . . . 5 55 2. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 7 56 2.1. Requirements on the FE model . . . . . . . . . . . . . . 7 57 2.2. The FE Model in Relation to FE Implementations . . . . . 8 58 2.3. The FE Model in Relation to the ForCES Protocol . . . . . 8 59 2.4. Modeling Language for the FE Model . . . . . . . . . . . 9 60 2.5. Document Structure . . . . . . . . . . . . . . . . . . . 9 61 3. ForCES Model Concepts . . . . . . . . . . . . . . . . . . . . 10 62 3.1. ForCES Capability Model and State Model . . . . . . . . . 11 63 3.1.1. FE Capability Model and State Model . . . . . . . . . 12 64 3.1.2. Relating LFB and FE Capability and State Model . . . 13 65 3.2. Logical Functional Block (LFB) Modeling . . . . . . . . . 14 66 3.2.1. LFB Outputs . . . . . . . . . . . . . . . . . . . . . 17 67 3.2.2. LFB Inputs . . . . . . . . . . . . . . . . . . . . . 20 68 3.2.3. Packet Type . . . . . . . . . . . . . . . . . . . . . 23 69 3.2.4. Metadata . . . . . . . . . . . . . . . . . . . . . . 24 70 3.2.5. LFB Events . . . . . . . . . . . . . . . . . . . . . 26 71 3.2.6. Component Properties . . . . . . . . . . . . . . . . 28 72 3.2.7. LFB Versioning . . . . . . . . . . . . . . . . . . . 28 73 3.2.8. LFB Inheritance . . . . . . . . . . . . . . . . . . . 29 74 3.3. ForCES Model Addressing . . . . . . . . . . . . . . . . . 30 75 3.3.1. Addressing LFB Components: Paths and Keys . . . . . . 31 76 3.4. FE Datapath Modeling . . . . . . . . . . . . . . . . . . 32 77 3.4.1. Alternative Approaches for Modeling FE Datapaths . . 32 78 3.4.2. Configuring the LFB Topology . . . . . . . . . . . . 36 79 4. Model and Schema for LFB Classes . . . . . . . . . . . . . . 40 80 4.1. Namespace . . . . . . . . . . . . . . . . . . . . . . . . 41 81 4.2. Element . . . . . . . . . . . . . . . . . . 41 82 4.3. Element . . . . . . . . . . . . . . . . . . . . . 43 83 4.4. Element for Frame Type Declarations . . . . . 44 84 4.5. Element for Data Type Definitions . . . . 44 85 4.5.1. Element for Renaming Existing Data Types . 48 86 4.5.2. Element for Deriving New Atomic Types . . . 48 87 4.5.3. Element to Define Arrays . . . . . . . . . . 49 88 4.5.4. Element to Define Structures . . . . . . . . 53 89 4.5.5. Element to Define Union Types . . . . . . . . 55 90 4.5.6. Element . . . . . . . . . . . . . . . . . . . 55 91 4.5.7. Augmentations . . . . . . . . . . . . . . . . . . . . 56 92 4.6. Element for Metadata Definitions . . . . . 57 93 4.7. Element for LFB Class Definitions . . . . 58 94 4.7.1. Element to Express LFB Inheritance . . 61 95 4.7.2. Element to Define LFB Inputs . . . . . . 61 96 4.7.3. Element to Define LFB Outputs . . . . . 64 97 4.7.4. Element to Define LFB Operational 98 Components . . . . . . . . . . . . . . . . . . . . . 66 99 4.7.5. Element to Define LFB Capability 100 Components . . . . . . . . . . . . . . . . . . . . . 69 101 4.7.6. Element for LFB Notification Generation . . 70 102 4.7.7. Element for LFB Operational 103 Specification . . . . . . . . . . . . . . . . . . . . 77 104 4.8. Properties . . . . . . . . . . . . . . . . . . . . . . . 77 105 4.8.1. Basic Properties . . . . . . . . . . . . . . . . . . 78 106 4.8.2. Array Properties . . . . . . . . . . . . . . . . . . 80 107 4.8.3. String Properties . . . . . . . . . . . . . . . . . . 80 108 4.8.4. Octetstring Properties . . . . . . . . . . . . . . . 81 109 4.8.5. Event Properties . . . . . . . . . . . . . . . . . . 82 110 4.8.6. Alias Properties . . . . . . . . . . . . . . . . . . 85 111 4.9. XML Schema for LFB Class Library Documents . . . . . . . 86 112 5. FE Components and Capabilities . . . . . . . . . . . . . . . 97 113 5.1. XML for FEObject Class definition . . . . . . . . . . . . 98 114 5.2. FE Capabilities . . . . . . . . . . . . . . . . . . . . . 104 115 5.2.1. ModifiableLFBTopology . . . . . . . . . . . . . . . . 105 116 5.2.2. SupportedLFBs and SupportedLFBType . . . . . . . . . 105 117 5.3. FE Components . . . . . . . . . . . . . . . . . . . . . . 108 118 5.3.1. FEState . . . . . . . . . . . . . . . . . . . . . . . 108 119 5.3.2. LFBSelectors and LFBSelectorType . . . . . . . . . . 108 120 5.3.3. LFBTopology and LFBLinkType . . . . . . . . . . . . . 109 121 5.3.4. FENeighbors and FEConfiguredNeighborType . . . . . . 109 122 6. Satisfying the Requirements on FE Model . . . . . . . . . . . 110 123 7. Using the FE model in the ForCES Protocol . . . . . . . . . . 111 124 7.1. FE Topology Query . . . . . . . . . . . . . . . . . . . . 113 125 7.2. FE Capability Declarations . . . . . . . . . . . . . . . 114 126 7.3. LFB Topology and Topology Configurability Query . . . . . 114 127 7.4. LFB Capability Declarations . . . . . . . . . . . . . . . 114 128 7.5. State Query of LFB Components . . . . . . . . . . . . . . 116 129 7.6. LFB Component Manipulation . . . . . . . . . . . . . . . 116 130 7.7. LFB Topology Re-configuration . . . . . . . . . . . . . . 116 131 8. Example LFB Definition . . . . . . . . . . . . . . . . . . . 117 132 8.1. Data Handling . . . . . . . . . . . . . . . . . . . . . . 124 133 8.1.1. Setting up a DLCI . . . . . . . . . . . . . . . . . . 125 134 8.1.2. Error Handling . . . . . . . . . . . . . . . . . . . 125 135 8.2. LFB Components . . . . . . . . . . . . . . . . . . . . . 126 136 8.3. Capabilities . . . . . . . . . . . . . . . . . . . . . . 126 137 8.4. Events . . . . . . . . . . . . . . . . . . . . . . . . . 127 138 9. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 128 139 9.1. URN Namespace Registration . . . . . . . . . . . . . . . 128 140 9.2. LFB Class Names and LFB Class Identifiers . . . . . . . . 128 141 10. Authors Emeritus . . . . . . . . . . . . . . . . . . . . . . 129 142 11. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 130 143 12. Security Considerations . . . . . . . . . . . . . . . . . . . 130 144 13. References . . . . . . . . . . . . . . . . . . . . . . . . . 130 145 13.1. Normative References . . . . . . . . . . . . . . . . . . 130 146 13.2. Informative References . . . . . . . . . . . . . . . . . 130 147 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 131 148 Intellectual Property and Copyright Statements . . . . . . . . . 132 150 1. Definitions 152 The use of compliance terminology (MUST, SHOULD, MAY) is used in 153 accordance with RFC2119 [1]. Such terminology is used in describing 154 the required behavior of ForCES forwarding elements or control 155 elements in supporting or manipulating information described in this 156 model. 158 Terminology associated with the ForCES requirements is defined in 159 RFC3654 [4] and is not copied here. The following list of 160 terminology relevant to the FE model is defined in this section. 162 FE Model -- The FE model is designed to model the logical processing 163 functions of an FE. The FE model proposed in this document includes 164 three components: the modeling of individual logical functional 165 blocks (LFB model), the logical interconnection between LFBs (LFB 166 topology) and the FE level attributes, including FE capabilities. 167 The FE model provides the basis to define the information elements 168 exchanged between the CE and the FE in the ForCES Protocol [2]. 170 Datapath -- A conceptual path taken by packets within the forwarding 171 plane inside an FE. Note that more than one datapath can exist 172 within an FE. 174 LFB (Logical Functional Block) Class (or type) -- A template that 175 represents a fine-grained, logically separable aspect of FE 176 processing. Most LFBs relate to packet processing in the data path. 177 LFB classes are the basic building blocks of the FE model. 179 LFB Instance -- As a packet flows through an FE along a datapath, it 180 flows through one or multiple LFB instances, where each LFB is an 181 instance of a specific LFB class. Multiple instances of the same LFB 182 class can be present in an FE's datapath. Note that we often refer 183 to LFBs without distinguishing between an LFB class and LFB instance 184 when we believe the implied reference is obvious for the given 185 context. 187 LFB Model -- The LFB model describes the content and structures in an 188 LFB, plus the associated data definition. XML is used to provide a 189 formal definition of the necessary structures for the modeling. Four 190 types of information are defined in the LFB model. The core part of 191 the LFB model is the LFB class definitions; the other three types of 192 information define constructs associated with and used by the class 193 definition. These are reusable data types, supported frame (packet) 194 formats, and metadata. 196 Element -- Element is generally used in this document in accordance 197 with the XML usage of the term. It refers to an XML tagged part of 198 an XML document. For a precise definition, please see the full set 199 of XML specifications from the W3C. This term is included in this 200 list for completeness because the ForCES formal model uses XML. 202 Attribute -- Attribute is used in the ForCES formal modelling in 203 accordance with standard XML usage of the term. i.e, to provide 204 attribute information include in an XML tag. 206 LFB Metadata -- Metadata is used to communicate per-packet state from 207 one LFB to another, but is not sent across the network. The FE model 208 defines how such metadata is identified, produced and consumed by the 209 LFBs, but not how the per-packet state is implemented within actual 210 hardware. Metadata is sent between the FE and the CE on redirect 211 packets. 213 ForCES Component -- a ForCES Component is a well-defined, uniquely 214 identifiable and addressable ForCES model building block. A 215 component has a 32-bit ID, name, type and an optional synopsis 216 description. These are often referred to simply as components. 218 LFB Component -- A ForCES component that defines the Operational 219 parameters of the LFBs that must be visible to the CEs. 221 Structure Component -- A ForCES component that is part of a complex 222 data structure to be used in LFB data definitions. The individual 223 parts which make up a structured set of data are referred to as 224 Structure Components. These can themselves be of any valid data 225 type, including tables and structures. 227 Property -- ForCES components have properties associated with them, 228 such as readability. Other examples include lengths for variable 229 sized components. These properties are acessed by the CE for reading 230 (or, where appropriate, writing.) Details on the ForCES properties 231 are found in section 4.8. 233 LFB Topology -- A representation of the logical interconnection and 234 the placement of LFB instances along the datapath within one FE. 235 Sometimes this representation is called intra-FE topology, to be 236 distinguished from inter-FE topology. LFB topology is outside of the 237 LFB model, but is part of the FE model. 239 FE Topology -- A representation of how multiple FEs within a single 240 NE (Network Element) are interconnected. Sometimes this is called 241 inter-FE topology, to be distinguished from intra-FE topology (i.e., 242 LFB topology). An individual FE might not have the global knowledge 243 of the full FE topology, but the local view of its connectivity with 244 other FEs is considered to be part of the FE model. The FE topology 245 is discovered by the ForCES base protocol or by some other means. 247 Inter-FE Topology -- See FE Topology. 249 Intra-FE Topology -- See LFB Topology. 251 LFB class library -- A set of LFB classes that has been identified as 252 the most common functions found in most FEs and hence should be 253 defined first by the ForCES Working Group. 255 2. Introduction 257 RFC3746 [5] specifies a framework by which control elements (CEs) can 258 configure and manage one or more separate forwarding elements (FEs) 259 within a networking element (NE) using the ForCES protocol. The 260 ForCES architecture allows Forwarding Elements of varying 261 functionality to participate in a ForCES network element. The 262 implication of this varying functionality is that CEs can make only 263 minimal assumptions about the functionality provided by FEs in an NE. 264 Before CEs can configure and control the forwarding behavior of FEs, 265 CEs need to query and discover the capabilities and states of their 266 FEs. RFC3654 [4] mandates that the capabilities, states and 267 configuration information be expressed in the form of an FE model. 269 RFC3444 [8] observed that information models (IMs) and data models 270 (DMs) are different because they serve different purposes. "The main 271 purpose of an IM is to model managed objects at a conceptual level, 272 independent of any specific implementations or protocols used". 273 "DMs, conversely, are defined at a lower level of abstraction and 274 include many details. They are intended for implementors and include 275 protocol-specific constructs." Sometimes it is difficult to draw a 276 clear line between the two. The FE model described in this document 277 is primarily an information model, but also includes some aspects of 278 a data model, such as explicit definitions of the LFB class schema 279 and FE schema. It is expected that this FE model will be used as the 280 basis to define the payload for information exchange between the CE 281 and FE in the ForCES protocol. 283 2.1. Requirements on the FE model 285 RFC3654 [4]defines requirements that must be satisfied by a ForCES FE 286 model. To summarize, an FE model must define: 288 o Logically separable and distinct packet forwarding operations in 289 an FE datapath (logical functional blocks or LFBs); 291 o The possible topological relationships (and hence the sequence of 292 packet forwarding operations) between the various LFBs; 294 o The possible operational capabilities (e.g., capacity limits, 295 constraints, optional features, granularity of configuration) of 296 each type of LFB; 298 o The possible configurable parameters (e.g., components) of each 299 type of LFB; 301 o Metadata that may be exchanged between LFBs. 303 2.2. The FE Model in Relation to FE Implementations 305 The FE model proposed here is based on an abstraction using distinct 306 logical functional blocks (LFBs), which are interconnected in a 307 directed graph, and receive, process, modify, and transmit packets 308 along with metadata. The FE model is designed, and any defined LFB 309 classes should be designed, such that different implementations of 310 the forwarding datapath can be logically mapped onto the model with 311 the functionality and sequence of operations correctly captured. 312 However, the model is not intended to directly address how a 313 particular implementation maps to an LFB topology. It is left to the 314 forwarding plane vendors to define how the FE functionality is 315 represented using the FE model. Our goal is to design the FE model 316 such that it is flexible enough to accommodate most common 317 implementations. 319 The LFB topology model for a particular datapath implementation must 320 correctly capture the sequence of operations on the packet. Metadata 321 generation by certain LFBs MUST always precede any use of that 322 metadata by subsequent LFBs in the topology graph; this is required 323 for logically consistent operation. Further, modification of packet 324 fields that are subsequently used as inputs for further processing 325 MUST occur in the order specified in the model for that particular 326 implementation to ensure correctness. 328 2.3. The FE Model in Relation to the ForCES Protocol 330 The ForCES base Protocol [2] is used by the CEs and FEs to maintain 331 the communication channel between the CEs and FEs. The ForCES 332 protocol may be used to query and discover the intra-FE topology. 333 The details of a particular datapath implementation inside an FE, 334 including the LFB topology, along with the operational capabilities 335 and attributes of each individual LFB, are conveyed to the CE within 336 information elements in the ForCES protocol. The model of an LFB 337 class should define all of the information that needs to be exchanged 338 between an FE and a CE for the proper configuration and management of 339 that LFB. 341 Specifying the various payloads of the ForCES messages in a 342 systematic fashion is difficult without a formal definition of the 343 objects being configured and managed (the FE and the LFBs within). 344 The FE Model document defines a set of classes and components for 345 describing and manipulating the state of the LFBs within an FE. 346 These class definitions themselves will generally not appear in the 347 ForCES protocol. Rather, ForCES protocol operations will reference 348 classes defined in this model, including relevant components and the 349 defined operations. 351 Section 7 provides more detailed discussion on how the FE model 352 should be used by the ForCES protocol. 354 2.4. Modeling Language for the FE Model 356 Even though not absolutely required, it is beneficial to use a formal 357 data modeling language to represent the conceptual FE model described 358 in this document. Use of a formal language can help to enforce 359 consistency and logical compatibility among LFBs. A full 360 specification will be written using such a data modeling language. 361 The formal definition of the LFB classes may facilitate the eventual 362 automation of some of the code generation process and the functional 363 validation of arbitrary LFB topologies. These class definitions form 364 the LFB Library. Documents which describe LFB Classes are therefore 365 referred to as LFB Library documents. 367 Human readability was the most important factor considered when 368 selecting the specification language, whereas encoding, decoding and 369 transmission performance was not a selection factor. The encoding 370 method for over the wire transport is not dependent on the 371 specification language chosen and is outside the scope of this 372 document and up to the ForCES protocol to define. 374 XML is chosen as the specification language in this document, because 375 XML has the advantage of being both human and machine readable with 376 widely available tools support. This document uses an XML Schema to 377 define the structure of the LFB Library documents, as defined in [9] 378 and [10] and [11]. While these LFB Class definitions are not sent in 379 the ForCES protocol, these definitions comply with the 380 recommendations in RFC3470 [9] on the use of XML in IETF protocols. 382 2.5. Document Structure 384 Section 3 provides a conceptual overview of the FE model, laying the 385 foundation for the more detailed discussion and specifications in the 386 sections that follow. Section 4 and Section 5 constitute the core of 387 the FE model, detailing the two major aspects of the FE model: a 388 general LFB model and a definition of the FE Object LFB, with its 389 components, including FE capabilities and LFB topology information. 391 Section 6 directly addresses the model requirements imposed by the 392 ForCES requirements defined in RFC3654 [4] while Section 7 explains 393 how the FE model should be used in the ForCES protocol. 395 3. ForCES Model Concepts 397 Some of the important ForCES concepts used throughout this document 398 are introduced in this section. These include the capability and 399 state abstraction, the FE and LFB model construction, and the unique 400 addressing of the different model structures. Details of these 401 aspects are described in Section 4 and Section 5. The intent of this 402 section is to discuss these concepts at the high level and lay the 403 foundation for the detailed description in the following sections. 405 The ForCES FE model includes both a capability and a state 406 abstraction. 408 o The FE/LFB capability model describes the capabilities and 409 capacities of an FE/LFB by specifying the variation in functions 410 supported and any limitations. Capacity describes the limits of 411 specific components (an example would be a table size limit). 413 o The state model describes the current state of the FE/LFB, that 414 is, the instantaneous values or operational behavior of the FE/ 415 LFB. 417 Section 3.1 explains the difference between a capability model and a 418 state model, and describes how the two can be combined in the FE 419 model. 421 The ForCES model construction laid out in this document allows an FE 422 to provide information about its structure for operation. This can 423 be thought of as FE level information and information about the 424 individual instances of LFBs provided by the FE. 426 o The ForCES model includes the constructions for defining the class 427 of logical function blocks (LFBS) that an FE may support. These 428 classes are defined in this and other documents. The definition 429 of such a class provides the information content for monitoring 430 and controlling instances of the LFB class for ForCES purposes. 431 Each LFB model class formally defines the operational LFB 432 components, LFB capabilities, and LFB events. Essentially, 433 Section 3.2 introduces the concept of LFBs as the basic functional 434 building blocks in the ForCES model. 436 o The FE model also provides the construction necessary to monitor 437 and control the FE as a whole for ForCES purposes. For 438 consistency of operation and simplicity, this information is 439 represented as an LFB, the FE Object LFB class and a singular LFB 440 instance of that class, defined using the LFB model. The FE 441 Object class defines the components to provide information at the 442 FE level, particularly the capabilities of the FE at a coarse 443 level, i.e., not all possible capabilities nor all details about 444 the capabilities of the FE. Part of the FE level information is 445 the LFB topology, which expresses the logical inter-connection 446 between the LFB instances along the datapath(s) within the FE. 447 Section 3.3 discusses the LFB topology. The FE Object also 448 includes information about what LFB classes the FE can support. 450 The ForCES model allows for unique identification of the different 451 constructs it defines. This includes identification of the LFB 452 classes, and of LFB instances within those classes, as well as 453 identification of components within those instances. 455 The ForCES Protocol [2] encapsulates target address(es) to eventually 456 get to a fine-grained entity being referenced by the CE. The 457 addressing hierarchy is broken into the following: 459 o An FE is uniquely identified by a 32 bit FEID. 461 o Each Class of LFB is uniquely identified by a 32 bit LFB ClassID. 462 The LFB ClassIDs are global within the Network Element and may be 463 issued by IANA. 465 o Within an FE, there can be multiple instances of each LFB class. 466 Each LFB Class instance is identified by a 32 bit identifier which 467 is unique within a particular LFB class on that FE. 469 o All the components within an LFB instance are further defined 470 using 32 bit identifiers. 472 Refer to Section 3.3 for more details on addressing. 474 3.1. ForCES Capability Model and State Model 476 Capability and state modelling applies to both the FE and LFB 477 abstraction. 479 Figure 1 shows the concepts of FE state, capabilities and 480 configuration in the context of CE-FE communication via the ForCES 481 protocol. 483 +-------+ +-------+ 484 | | FE capabilities: what it can/cannot do. | | 485 | |<-----------------------------------------| | 486 | | | | 487 | CE | FE state: what it is now. | FE | 488 | |<-----------------------------------------| | 489 | | | | 490 | | FE configuration: what it should be. | | 491 | |----------------------------------------->| | 492 +-------+ +-------+ 494 Figure 1: Illustration of FE capabilities, state and configuration 495 exchange in the context of CE-FE communication via ForCES. 497 3.1.1. FE Capability Model and State Model 499 Conceptually, the FE capability model tells the CE which states are 500 allowed on an FE, with capacity information indicating certain 501 quantitative limits or constraints. Thus, the CE has general 502 knowledge about configurations that are applicable to a particular 503 FE. 505 3.1.1.1. FE Capability Model 507 The FE capability model may be used to describe an FE at a coarse 508 level. For example, an FE might be defined as follows: 510 o the FE can handle IPv4 and IPv6 forwarding; 512 o the FE can perform classification based on the following fields: 513 source IP address, destination IP address, source port number, 514 destination port number, etc.; 516 o the FE can perform metering; 518 o the FE can handle up to N queues (capacity); 520 o the FE can add and remove encapsulating headers of types including 521 IPsec, GRE, L2TP. 523 While one could try to build an object model to fully represent the 524 FE capabilities, other efforts found this approach to be a 525 significant undertaking. The main difficulty arises in describing 526 detailed limits, such as the maximum number of classifiers, queues, 527 buffer pools, and meters that the FE can provide. We believe that a 528 good balance between simplicity and flexibility can be achieved for 529 the FE model by combining coarse level capability reporting with an 530 error reporting mechanism. That is, if the CE attempts to instruct 531 the FE to set up some specific behavior it cannot support, the FE 532 will return an error indicating the problem. Examples of similar 533 approaches include DiffServ PIB RFC3317 [6] and Framework PIB RFC3318 534 [7]. 536 3.1.1.2. FE State Model 538 The FE state model presents the snapshot view of the FE to the CE. 539 For example, using an FE state model, an FE might be described to its 540 corresponding CE as the following: 542 o on a given port, the packets are classified using a given 543 classification filter; 545 o the given classifier results in packets being metered in a certain 546 way and then marked in a certain way; 548 o the packets coming from specific markers are delivered into a 549 shared queue for handling, while other packets are delivered to a 550 different queue; 552 o a specific scheduler with specific behavior and parameters will 553 service these collected queues. 555 3.1.1.3. LFB Capability and State Model 557 Both LFB Capability and State information are defined formally using 558 the LFB modelling XML schema. 560 Capability information at the LFB level is an integral part of the 561 LFB model and provides for powerful semantics. For example, when 562 certain features of an LFB class are optional, the CE needs to be 563 able to determine whether those optional features are supported by a 564 given LFB instance. The schema for the definition of LFB classes 565 provides a means for identifying such components. 567 State information is defined formally using LFB component constructs. 569 3.1.2. Relating LFB and FE Capability and State Model 571 Capability information at the FE level describes the LFB classes that 572 the FE can instantiate, the number of instances of each that can be 573 created, the topological (linkage) limitations between these LFB 574 instances, etc. Section 5 defines the FE level components including 575 capability information. Since all information is represented as 576 LFBs, this is provided by a single instance of the FE Object LFB 577 Class. By using a single instance with a known LFB Class and a known 578 instance identification, the ForCES protocol can allow a CE to access 579 this information whenever it needs to, including while the CE is 580 establishing the control of the FE. 582 Once the FE capability is described to the CE, the FE state 583 information can be represented at two levels. The first level is the 584 logically separable and distinct packet processing functions, called 585 LFBs. The second level of information describes how these individual 586 LFBs are ordered and placed along the datapath to deliver a complete 587 forwarding plane service. The interconnection and ordering of the 588 LFBs is called LFB Topology. Section 3.2 discusses high level 589 concepts around LFBs, whereas Section 3.3 discusses LFB topology 590 issues. This topology information is represented as components of 591 the FE Object LFB instance, to allow the CE to fetch and manipulate 592 this. 594 3.2. Logical Functional Block (LFB) Modeling 596 Each LFB performs a well-defined action or computation on the packets 597 passing through it. Upon completion of its prescribed function, 598 either the packets are modified in certain ways (e.g., decapsulator, 599 marker), or some results are generated and stored, often in the form 600 of metadata (e.g., classifier). Each LFB typically performs a single 601 action. Classifiers, shapers and meters are all examples of such 602 LFBs. Modeling LFBs at such a fine granularity allows us to use a 603 small number of LFBs to express the higher-order FE functions (such 604 as an IPv4 forwarder) precisely, which in turn can describe more 605 complex networking functions and vendor implementations of software 606 and hardware. These fine grained LFBs will be defined in detail in 607 one or more documents to be published separately, using the material 608 in this model. 610 It is also the case that LFBs may exist in order to provide a set of 611 components for control of FE operation by the CE (i.e., a locus of 612 control), without tying that control to specific packets or specific 613 parts of the data path. An example of such an LFB is the FE Object 614 which provides the CE with information about the FE as a whole, and 615 allows the FE to control some aspects of the FE, such as the datapath 616 itself. Such LFBs will not have the packet oriented properties 617 described in this section. 619 In general, multiple LFBs are contained in one FE, as shown in 620 Figure 2, and all the LFBs share the same ForCES protocol (Fp) 621 termination point that implements the ForCES protocol logic and 622 maintains the communication channel to and from the CE. 624 +-----------+ 625 | CE | 626 +-----------+ 627 ^ 628 | Fp reference point 629 | 630 +--------------------------|-----------------------------------+ 631 | FE | | 632 | v | 633 | +----------------------------------------------------------+ | 634 | | ForCES protocol | | 635 | | termination point | | 636 | +----------------------------------------------------------+ | 637 | ^ ^ | 638 | : : Internal control | 639 | : : | 640 | +---:----------+ +---:----------| | 641 | | :LFB1 | | : LFB2 | | 642 | =====>| v |============>| v |======>...| 643 | Inputs| +----------+ |Outputs | +----------+ | | 644 | (P,M) | |Components| |(P',M') | |Components| |(P",M") | 645 | | +----------+ | | +----------+ | | 646 | +--------------+ +--------------+ | 647 | | 648 +--------------------------------------------------------------+ 650 Figure 2: Generic LFB Diagram 652 An LFB, as shown in Figure 2, may have inputs, outputs and components 653 that can be queried and manipulated by the CE via an Fp reference 654 point (defined in RFC3746 [5]) and the ForCES protocol termination 655 point. The horizontal axis is in the forwarding plane for connecting 656 the inputs and outputs of LFBs within the same FE. P (with marks to 657 indicate modification) indicates a data packet, while M (with marks 658 to indicate modification) indicates the metadata associated with a 659 packet. The vertical axis between the CE and the FE denotes the Fp 660 reference point where bidirectional communication between the CE and 661 FE occurs: the CE to FE communication is for configuration, control, 662 and packet injection, while FE to CE communication is used for packet 663 redirection to the control plane, reporting of monitoring and 664 accounting information, reporting of errors, etc. Note that the 665 interaction between the CE and the LFB is only abstract and indirect. 666 The result of such an interaction is for the CE to manipulate the 667 components of the LFB instances. 669 An LFB can have one or more inputs. Each input takes a pair of a 670 packet and its associated metadata. Depending upon the LFB input 671 port definition, the packet or the metadata may be allowed to be 672 empty (or equivalently to not be provided.) When input arrives at an 673 LFB, either the packet or its associated metadata must be non-empty 674 or there is effectively no input. (LFB operation generally may be 675 triggered by input arrival, by timers, or by other system state. It 676 is only in the case where the goal is to have input drive operation 677 that the input must be non-empty.) 679 The LFB processes the input, and produces one or more outputs, each 680 of which is a pair of a packet and its associated metadata. Again, 681 depending upon the LFB output port definition, either the packet or 682 the metadata may be allowed to be empty (or equivalently to be 683 absent.) Metadata attached to packets on output may be metadata that 684 was received, or may be information about the packet processing that 685 may be used by later LFBs in the FEs packet processing. 687 A namespace is used to associate a unique name and ID with each LFB 688 class. The namespace MUST be extensible so that a new LFB class can 689 be added later to accommodate future innovation in the forwarding 690 plane. 692 LFB operation is specified in the model to allow the CE to understand 693 the behavior of the forwarding datapath. For instance, the CE needs 694 to understand at what point in the datapath the IPv4 header TTL is 695 decremented by the FE. That is, the CE needs to know if a control 696 packet could be delivered to it either before or after this point in 697 the datapath. In addition, the CE needs to understand where and what 698 type of header modifications (e.g., tunnel header append or strip) 699 are performed by the FEs. Further, the CE works to verify that the 700 various LFBs along a datapath within an FE are compatible to link 701 together. Connecting incompatible LFB instances will produce a non- 702 working data path. So the model is designed to provide sufficient 703 information for the CE to make this determination. 705 Selecting the right granularity for describing the functions of the 706 LFBs is an important aspect of this model. There is value to vendors 707 if the operation of LFB classes can be expressed in sufficient detail 708 so that physical devices implementing different LFB functions can be 709 integrated easily into an FE design. However, the model, and the 710 associated library of LFBs, must not be so detailed and so specific 711 as to significantly constrain implementations. Therefore, a semi- 712 formal specification is needed; that is, a text description of the 713 LFB operation (human readable), but sufficiently specific and 714 unambiguous to allow conformance testing and efficient design, so 715 that interoperability between different CEs and FEs can be achieved. 717 The LFB class model specifies information such as: 719 o number of inputs and outputs (and whether they are configurable) 721 o metadata read/consumed from inputs; 723 o metadata produced at the outputs; 725 o packet type(s) accepted at the inputs and emitted at the outputs; 727 o packet content modifications (including encapsulation or 728 decapsulation); 730 o packet routing criteria (when multiple outputs on an LFB are 731 present); 733 o packet timing modifications; 735 o packet flow ordering modifications; 737 o LFB capability information components; 739 o events that can be detected by the LFB, with notification to the 740 CE; 742 o LFB operational components; 744 o etc. 746 Section 4 of this document provides a detailed discussion of the LFB 747 model with a formal specification of LFB class schema. The rest of 748 Section 3.2 only intends to provide a conceptual overview of some 749 important issues in LFB modeling, without covering all the specific 750 details. 752 3.2.1. LFB Outputs 754 An LFB output is a conceptual port on an LFB that can send 755 information to another LFB. The information sent on that port is a 756 pair of a packet and associated metadata, one of which may be empty. 757 (If both were empty, there would be no output.) 759 A single LFB output can be connected to only one LFB input. This is 760 required to make the packet flow through the LFB topology 761 unambiguous. 763 Some LFBs will have a single output, as depicted in Figure 3.a. 765 +---------------+ +-----------------+ 766 | | | | 767 | | | OUT +--> 768 ... OUT +--> ... | 769 | | | EXCEPTIONOUT +--> 770 | | | | 771 +---------------+ +-----------------+ 773 a. One output b. Two distinct outputs 775 +---------------+ +-----------------+ 776 | | | EXCEPTIONOUT +--> 777 | OUT:1 +--> | | 778 ... OUT:2 +--> ... OUT:1 +--> 779 | ... +... | OUT:2 +--> 780 | OUT:n +--> | ... +... 781 +---------------+ | OUT:n +--> 782 +-----------------+ 784 c. One output group d. One output and one output group 786 Figure 3: Examples of LFBs with various output combinations. 788 To accommodate a non-trivial LFB topology, multiple LFB outputs are 789 needed so that an LFB class can fork the datapath. Two mechanisms 790 are provided for forking: multiple singleton outputs and output 791 groups, which can be combined in the same LFB class. 793 Multiple separate singleton outputs are defined in an LFB class to 794 model a pre-determined number of semantically different outputs. 795 That is, the LFB class definition MUST include the number of outputs, 796 implying the number of outputs is known when the LFB class is 797 defined. Additional singleton outputs cannot be created at LFB 798 instantiation time, nor can they be created on the fly after the LFB 799 is instantiated. 801 For example, an IPv4 LPM (Longest-Prefix-Matching) LFB may have one 802 output (OUT) to send those packets for which the LPM look-up was 803 successful, passing a META_ROUTEID as metadata; and have another 804 output (EXCEPTIONOUT) for sending exception packets when the LPM 805 look-up failed. This example is depicted in Figure 3.b. Packets 806 emitted by these two outputs not only require different downstream 807 treatment, but they are a result of two different conditions in the 808 LFB and each output carries different metadata. This concept assumes 809 the number of distinct outputs is known when the LFB class is 810 defined. For each singleton output, the LFB class definition defines 811 the types of frames (packets) and metadata the output emits. 813 An output group, on the other hand, is used to model the case where a 814 flow of similar packets with an identical set of permitted metadata 815 needs to be split into multiple paths. In this case, the number of 816 such paths is not known when the LFB class is defined because it is 817 not an inherent property of the LFB class. An output group consists 818 of a number of outputs, called the output instances of the group, 819 where all output instances share the same frame (packet) and metadata 820 emission definitions (see Figure 3.c). Each output instance can 821 connect to a different downstream LFB, just as if they were separate 822 singleton outputs, but the number of output instances can differ 823 between LFB instances of the same LFB class. The class definition 824 may include a lower and/or an upper limit on the number of outputs. 825 In addition, for configurable FEs, the FE capability information may 826 define further limits on the number of instances in specific output 827 groups for certain LFBs. The actual number of output instances in a 828 group is an component of the LFB instance, which is read-only for 829 static topologies, and read-write for dynamic topologies. The output 830 instances in a group are numbered sequentially, from 0 to N-1, and 831 are addressable from within the LFB. To use Output Port groups, the 832 LFB has to have a built-in mechanism to select one specific output 833 instance for each packet. This mechanism is described in the textual 834 definition of the class and is typically configurable via some 835 attributes of the LFB. 837 For example, consider a redirector LFB, whose sole purpose is to 838 direct packets to one of N downstream paths based on one of the 839 metadata associated with each arriving packet. Such an LFB is fairly 840 versatile and can be used in many different places in a topology. 841 For example, given LFBs which record the type of packet in a 842 FRAMETYPE metadatum, or a packet rate class in a COLOR metadatum, one 843 may uses these metadata for branching. A redirector can be used to 844 divide the data path into an IPv4 and an IPv6 path based on a 845 FRAMETYPE metadatum (N=2), or to fork into rate specific paths after 846 metering using the COLOR metadatum (red, yellow, green; N=3), etc. 848 Using an output group in the above LFB class provides the desired 849 flexibility to adapt each instance of this class to the required 850 operation. The metadata to be used as a selector for the output 851 instance is a property of the LFB. For each packet, the value of the 852 specified metadata may be used as a direct index to the output 853 instance. Alternatively, the LFB may have a configurable selector 854 table that maps a metadatum value to output instance. 856 Note that other LFBs may also use the output group concept to build 857 in similar adaptive forking capability. For example, a classifier 858 LFB with one input and N outputs can be defined easily by using the 859 output group concept. Alternatively, a classifier LFB with one 860 singleton output in combination with an explicit N-output re- 861 director LFB models the same processing behavior. The decision of 862 whether to use the output group model for a certain LFB class is left 863 to the LFB class designers. 865 The model allows the output group to be combined with other singleton 866 output(s) in the same class, as demonstrated in Figure 3.d. The LFB 867 here has two types of outputs, OUT, for normal packet output, and 868 EXCEPTIONOUT for packets that triggered some exception. The normal 869 OUT has multiple instances, thus, it is an output group. 871 In summary, the LFB class may define one output, multiple singleton 872 outputs, one or more output groups, or a combination thereof. 873 Multiple singleton outputs should be used when the LFB must provide 874 for forking the datapath and at least one of the following conditions 875 hold: 877 o the number of downstream directions is inherent from the 878 definition of the class and hence fixed; 880 o the frame type and set of permitted metadata emitted on any of the 881 outputs are different from what is emitted on the other outputs 882 (i.e., they cannot share their frametype and permitted metadata 883 definitions). 885 An output group is appropriate when the LFB must provide for forking 886 the datapath and at least one of the following conditions hold: 888 o the number of downstream directions is not known when the LFB 889 class is defined; 891 o the frame type and set of metadata emitted on these outputs are 892 sufficiently similar or, ideally, identical, such they can share 893 the same output definition. 895 3.2.2. LFB Inputs 897 An LFB input is a conceptual port on an LFB on which the LFB can 898 receive information from other LFBs. The information is typically a 899 pair of a packet and its associated metadata. Either the packet, or 900 the metadata, may for some LFBs and some situations be empty. They 901 can not both be empty, as then there is no input. 903 For LFB instances that receive packets from more than one other LFB 904 instance (fan-in) there are three ways to model fan-in, all supported 905 by the LFB model and can all be combined in the same LFB: 907 o Implicit multiplexing via a single input 909 o Explicit multiplexing via multiple singleton inputs 911 o Explicit multiplexing via a group of inputs (input group) 913 The simplest form of multiplexing uses a singleton input 914 (Figure 4.a). Most LFBs will have only one singleton input. 915 Multiplexing into a single input is possible because the model allows 916 more than one LFB output to connect to the same LFB input. This 917 property applies to any LFB input without any special provisions in 918 the LFB class. Multiplexing into a single input is applicable when 919 the packets from the upstream LFBs are similar in frametype and 920 accompanying metadata, and require similar processing. Note that 921 this model does not address how potential contention is handled when 922 multiple packets arrive simultaneously. If contention handling needs 923 to be explicitly modeled, one of the other two modeling solutions 924 must be used. 926 The second method to model fan-in uses individually defined singleton 927 inputs (Figure 4.b). This model is meant for situations where the 928 LFB needs to handle distinct types of packet streams, requiring 929 input-specific handling inside the LFB, and where the number of such 930 distinct cases is known when the LFB class is defined. For example, 931 an LFB which can perform both Layer 2 decapsulation (to Layer 3) and 932 Layer 3 encapsulation (to Layer 2) may have two inputs, one for 933 receiving Layer 2 frames for decapsulation, and one for receiving 934 Layer 3 frames for encapsulation. This LFB type expects different 935 frames (L2 vs. L3) at its inputs, each with different sets of 936 metadata, and would thus apply different processing on frames 937 arriving at these inputs. This model is capable of explicitly 938 addressing packet contention by defining how the LFB class handles 939 the contending packets. 941 +--------------+ +------------------------+ 942 | LFB X +---+ | | 943 +--------------+ | | | 944 | | | 945 +--------------+ v | | 946 | LFB Y +---+-->|input Meter LFB | 947 +--------------+ ^ | | 948 | | | 949 +--------------+ | | | 950 | LFB Z |---+ | | 951 +--------------+ +------------------------+ 953 (a) An LFB connects with multiple upstream LFBs via a single input. 955 +--------------+ +------------------------+ 956 | LFB X +---+ | | 957 +--------------+ +-->|layer2 | 958 +--------------+ | | 959 | LFB Y +------>|layer3 LFB | 960 +--------------+ +------------------------+ 962 (b) An LFB connects with multiple upstream LFBs via two separate 963 singleton inputs. 965 +--------------+ +------------------------+ 966 | Queue LFB #1 +---+ | | 967 +--------------+ | | | 968 | | | 969 +--------------+ +-->|in:0 \ | 970 | Queue LFB #2 +------>|in:1 | input group | 971 +--------------+ |... | | 972 +-->|in:N-1 / | 973 ... | | | 974 +--------------+ | | | 975 | Queue LFB #N |---+ | Scheduler LFB | 976 +--------------+ +------------------------+ 978 (c) A Scheduler LFB uses an input group to differentiate which queue 979 LFB packets are coming from. 981 Figure 4: Examples of LFBs with various input combinations. 983 The third method to model fan-in uses the concept of an input group. 984 The concept is similar to the output group introduced in the previous 985 section and is depicted in Figure 4.c. An input group consists of a 986 number of input instances, all sharing the properties (same frame and 987 metadata expectations). The input instances are numbered from 0 to 988 N-1. From the outside, these inputs appear as normal inputs, i.e., 989 any compatible upstream LFB can connect its output to one of these 990 inputs. When a packet is presented to the LFB at a particular input 991 instance, the index of the input where the packet arrived is known to 992 the LFB and this information may be used in the internal processing. 993 For example, the input index can be used as a table selector, or as 994 an explicit precedence selector to resolve contention. As with 995 output groups, the number of input instances in an input group is not 996 defined in the LFB class. However, the class definition may include 997 restrictions on the range of possible values. In addition, if an FE 998 supports configurable topologies, it may impose further limitations 999 on the number of instances for particular port group(s) of a 1000 particular LFB class. Within these limitations, different instances 1001 of the same class may have a different number of input instances. 1002 The number of actual input instances in the group is a component 1003 defined in the LFB class, which is read-only for static topologies, 1004 and is read-write for configurable topologies. 1006 As an example for the input group, consider the Scheduler LFB 1007 depicted in Figure 4.c. Such an LFB receives packets from a number 1008 of Queue LFBs via a number of input instances, and uses the input 1009 index information to control contention resolution and scheduling. 1011 In summary, the LFB class may define one input, multiple singleton 1012 inputs, one or more input groups, or a combination thereof. Any 1013 input allows for implicit multiplexing of similar packet streams via 1014 connecting multiple outputs to the same input. Explicit multiple 1015 singleton inputs are useful when either the contention handling must 1016 be handled explicitly, or when the LFB class must receive and process 1017 a known number of distinct types of packet streams. An input group 1018 is suitable when contention handling must be modeled explicitly, but 1019 the number of inputs is not inherent from the class (and hence is not 1020 known when the class is defined), or when it is critical for LFB 1021 operation to know exactly on which input the packet was received. 1023 3.2.3. Packet Type 1025 When LFB classes are defined, the input and output packet formats 1026 (e.g., IPv4, IPv6, Ethernet, etc.) MUST be specified. These are the 1027 types of packets that a given LFB input is capable of receiving and 1028 processing, or that a given LFB output is capable of producing. This 1029 model requires that distinct packet types be uniquely labeled with a 1030 symbolic name and/or ID. 1032 Note that each LFB has a set of packet types that it operates on, but 1033 does not care whether the underlying implementation is passing a 1034 greater portion of the packets. For example, an IPv4 LFB might only 1035 operate on IPv4 packets, but the underlying implementation may or may 1036 not be stripping the L2 header before handing it over. Whether such 1037 processing is happening or not is opaque to the CE. 1039 3.2.4. Metadata 1041 Metadata is state that is passed from one LFB to another alongside a 1042 packet. The metadata passed with the packet assists subsequent LFBs 1043 to process that packet. 1045 The ForCES model defines metadata as precise atomic definitions in 1046 the form of label, value pairs. 1048 The ForCES model provides to the authors of LFB classes a way to 1049 formally define how to achieve metadata creation, modification, 1050 reading, as well as consumption (deletion). 1052 Inter-FE metadata, i.e, metadata crossing FEs, while it is likely to 1053 be semantically similar to this metadata, is out of scope for this 1054 document. 1056 Section 4 has informal details on metadata. 1058 3.2.4.1. Metadata Lifecycle Within the ForCES Model 1060 Each metadatum is modeled as a pair, where the label 1061 identifies the type of information, (e.g., "color"), and its value 1062 holds the actual information (e.g., "red"). The label here is shown 1063 as a textual label, but for protocol processing it is associated with 1064 a unique numeric value (identifier). 1066 To ensure inter-operability between LFBs, the LFB class specification 1067 must define what metadata the LFB class "reads" or "consumes" on its 1068 input(s) and what metadata it "produces" on its output(s). For 1069 maximum extensibility, this definition should neither specify which 1070 LFBs the metadata is expected to come from for a consumer LFB, nor 1071 which LFBs are expected to consume metadata for a given producer LFB. 1073 3.2.4.2. Metadata Production and Consumption 1075 For a given metadatum on a given packet path, there MUST be at least 1076 one producer LFB that creates that metadatum and SHOULD be at least 1077 one consumer LFB that needs that metadatum. 1079 In the ForCES model, the producer and consumer LFBs of a metadatum 1080 are not required to be adjacent. In addition, there may be multiple 1081 producers and consumers for the same metadatum. When a packet path 1082 involves multiple producers of the same metadatum, then subsequent 1083 producers overwrite that metadatum value. 1085 The metadata that is produced by an LFB is specified by the LFB class 1086 definition on a per-output-port-group basis. A producer may always 1087 generate the metadata on the port group, or may generate it only 1088 under certain conditions. We call the former "unconditional" 1089 metadata, whereas the latter is a "conditional" metadata. For 1090 example, deep packet inspection LFB might produce several pieces of 1091 metadata about the packet. The first metadatum might be the IP 1092 protocol (TCP, UDP, SCTP, ...) being carried, and two additional 1093 metadata items might be the source and destination port number. 1094 These additional metadata items are conditional on the value of the 1095 first metadatum (IP carried protocol) as they are only produced for 1096 protocols which use port numbers. In the case of conditional 1097 metadata, it should be possible to determine from the definition of 1098 the LFB when "conditional" metadata is produced. The consumer 1099 behavior of an LFB, that is, the metadata that the LFB needs for its 1100 operation, is defined in the LFB class definition on a per-input- 1101 port-group basis. An input port group may "require" a given 1102 metadatum, or may treat it as "optional" information. In the latter 1103 case, the LFB class definition MUST explicitly define what happens if 1104 any optional metadata is not provided. One approach is to specify a 1105 default value for each optional metadatum, and assume that the 1106 default value is used for any metadata which is not provided with the 1107 packet. 1109 When specifying the metadata tags, some harmonization effort must be 1110 made so that the producer LFB class uses the same tag as its intended 1111 consumer(s). 1113 3.2.4.3. LFB Operations on Metadata 1115 When the packet is processed by an LFB (i.e., between the time it is 1116 received and forwarded by the LFB), the LFB may perform read, write, 1117 and/or consume operations on any active metadata associated with the 1118 packet. If the LFB is considered to be a black box, one of the 1119 following operations is performed on each active metadatum. 1121 * IGNORE: ignores and forwards the metadatum 1123 * READ: reads and forwards the metadatum 1124 * READ/RE-WRITE: reads, over-writes and forwards the metadatum 1126 * WRITE: writes and forwards the metadatum (can also be used to 1127 create new metadata) 1129 * READ-AND-CONSUME: reads and consumes the metadatum 1131 * CONSUME consumes metadatum without reading 1133 The last two operations terminate the life-cycle of the metadatum, 1134 meaning that the metadatum is not forwarded with the packet when the 1135 packet is sent to the next LFB. 1137 In the ForCES model, a new metadatum is generated by an LFB when the 1138 LFB applies a WRITE operation to a metadatum type that was not 1139 present when the packet was received by the LFB. Such implicit 1140 creation may be unintentional by the LFB, that is, the LFB may apply 1141 the WRITE operation without knowing or caring if the given metadatum 1142 existed or not. If it existed, the metadatum gets over-written; if 1143 it did not exist, the metadatum is created. 1145 For LFBs that insert packets into the model, WRITE is the only 1146 meaningful metadata operation. 1148 For LFBs that remove the packet from the model, they may either READ- 1149 AND-CONSUME (read) or CONSUME (ignore) each active metadatum 1150 associated with the packet. 1152 3.2.5. LFB Events 1154 During operation, various conditions may occur that can be detected 1155 by LFBs. Examples range from link failure or restart to timer 1156 expiration in special purpose LFBs. The CE may wish to be notified 1157 of the occurrence of such events. The description of how such 1158 messages are sent, and their format, is part of the Forwarding and 1159 Control Element Separation (ForCES) protocol [2] document. 1160 Indicating how such conditions are understood is part of the job of 1161 this model. 1163 Events are declared in the LFB class definition. The LFB event 1164 declaration constitutes: 1166 o a unique 32 bit identifier. 1168 o An LFB component which is used to trigger the event. This entity 1169 is known as the event target. 1171 o A condition that will happen to the event target that will result 1172 in a generation of an event to the CE. Examples of a condition 1173 include something getting created, deleted, config change, etc. 1175 o What should be reported to the CE by the FE if the declared 1176 condition is met. 1178 The declaration of an event within an LFB class essentially defines 1179 what part of the LFB component(s) need to be monitored for events, 1180 what condition on the LFB monitored LFB component an FE should detect 1181 to trigger such an event, and what to report to the CE when the event 1182 is triggered. 1184 While events may be declared by the LFB class definition, runtime 1185 activity is controlled using built-in event properties using LFB 1186 component Properties (discussed in Section 3.2.6). A CE subscribes 1187 to the events on an LFB class instance by setting an event property 1188 for subscription. Each event has a subscription property which is by 1189 default off. A CE wishing to receive a specific event needs to turn 1190 on the subscription property at runtime. 1192 Event properties also provide semantics for runtime event filtering. 1193 A CE may set an event property to further suppress events to which it 1194 has already subscribed. The LFB model defines such filters to 1195 include threshold values, hysteresis, time intervals, number of 1196 events, etc. 1198 The contents of reports with events are designed to allow for the 1199 common, closely related information that the CE can be strongly 1200 expected to need to react to the event. It is not intended to carry 1201 information that the CE already has, nor large volumes of 1202 information, nor information related in complex fashions. 1204 From a conceptual point of view, at runtime, event processing is 1205 split into: 1207 1. detection of something happening to the (declared during LFB 1208 class definition) event target. Processing the next step happens 1209 if the CE subscribed (at runtime) to the event. 1211 2. checking of the (declared during LFB class definition) condition 1212 on the LFB event target. If the condition is met, proceed with 1213 the next step. 1215 3. checking (runtime set) event filters if they exist to see if the 1216 event should be reported or suppressed. If the event is to be 1217 reported proceed to the next step. 1219 4. Submitting of the declared report to the CE. 1221 Section 4.7.6 discusses events in more details. 1223 3.2.6. Component Properties 1225 LFBs and structures are made up of Components, containing the 1226 information that the CE needs to see and/or change about the 1227 functioning of the LFB. These Components, as described in detail in 1228 Section 4.7, may be basic values, complex structures (containing 1229 multiple Components themselves, each of which can be values, 1230 structures, or tables), or tables (which contain values, structures 1231 or tables). Components may be defined such that their appearence in 1232 LFB instances is optional. Components may be readable or writable at 1233 the discretion of the FE implementation. The CE needs to know these 1234 properties. Additionally, certain kinds of Components (arrays / 1235 tables, aliases, and events) have additional property information 1236 that the CE may need to read or write. This model defines the 1237 structure of the property information for all defined data types. 1239 Section 4.8 describes properties in more details. 1241 3.2.7. LFB Versioning 1243 LFB class versioning is a method to enable incremental evolution of 1244 LFB classes. In general, an FE is not allowed to contain an LFB 1245 instance for more than one version of a particular class. 1246 Inheritance (discussed next in Section 3.2.8) has special rules. If 1247 an FE datapath model containing an LFB instance of a particular class 1248 C also simultaneously contains an LFB instance of a class C' 1249 inherited from class C; C could have a different version than C'. 1251 LFB class versioning is supported by requiring a version string in 1252 the class definition. CEs may support multiple versions of a 1253 particular LFB class to provide backward compatibility, but FEs MUST 1254 NOT support more than one version of a particular class. 1256 Versioning is not restricted to making backwards compatible changes. 1257 It is specifically expected to be used to make changes that cannot be 1258 represented by inheritance. Often this will be to correct errors, 1259 and hence may not be backwards compatible. It may also be used to 1260 remove components which are not considered useful (particularly if 1261 they were previously mandatory, and hence were an implementation 1262 impediment.) 1264 3.2.8. LFB Inheritance 1266 LFB class inheritance is supported in the FE model as a method to 1267 define new LFB classes. This also allows FE vendors to add vendor- 1268 specific extensions to standardized LFBs. An LFB class specification 1269 MUST specify the base class and version number it inherits from (the 1270 default is the base LFB class). Multiple inheritance is not allowed, 1271 however, to avoid unnecessary complexity. 1273 Inheritance should be used only when there is significant reuse of 1274 the base LFB class definition. A separate LFB class should be 1275 defined if little or no reuse is possible between the derived and the 1276 base LFB class. 1278 An interesting issue related to class inheritance is backward 1279 compatibility between a descendant and an ancestor class. Consider 1280 the following hypothetical scenario where a standardized LFB class 1281 "L1" exists. Vendor A builds an FE that implements LFB "L1" and 1282 vendor B builds a CE that can recognize and operate on LFB "L1". 1283 Suppose that a new LFB class, "L2", is defined based on the existing 1284 "L1" class by extending its capabilities incrementally. Let us 1285 examine the FE backward compatibility issue by considering what would 1286 happen if vendor B upgrades its FE from "L1" to "L2" and vendor C's 1287 CE is not changed. The old L1-based CE can interoperate with the new 1288 L2-based FE if the derived LFB class "L2" is indeed backward 1289 compatible with the base class "L1". 1291 The reverse scenario is a much less problematic case, i.e., when CE 1292 vendor B upgrades to the new LFB class "L2", but the FE is not 1293 upgraded. Note that as long as the CE is capable of working with 1294 older LFB classes, this problem does not affect the model; hence we 1295 will use the term "backward compatibility" to refer to the first 1296 scenario concerning FE backward compatibility. 1298 Backward compatibility can be designed into the inheritance model by 1299 constraining LFB inheritance to require the derived class be a 1300 functional superset of the base class (i.e. the derived class can 1301 only add functions to the base class, but not remove functions). 1302 Additionally, the following mechanisms are required to support FE 1303 backward compatibility: 1305 1. When detecting an LFB instance of an LFB type that is unknown to 1306 the CE, the CE MUST be able to query the base class of such an 1307 LFB from the FE. 1309 2. The LFB instance on the FE SHOULD support a backward 1310 compatibility mode (meaning the LFB instance reverts itself back 1311 to the base class instance), and the CE SHOULD be able to 1312 configure the LFB to run in such a mode. 1314 3.3. ForCES Model Addressing 1316 Figure 5 demonstrates the abstraction of the different ForCES model 1317 entities. The ForCES protocol provides the mechanism to uniquely 1318 identify any of the LFB Class instance components. 1320 FE Address = FE01 1321 +--------------------------------------------------------------+ 1322 | | 1323 | +--------------+ +--------------+ | 1324 | | LFB ClassID 1| |LFB ClassID 91| | 1325 | | InstanceID 3 |============>|InstanceID 3 |======>... | 1326 | | +----------+ | | +----------+ | | 1327 | | |Components| | | |Components| | | 1328 | | +----------+ | | +----------+ | | 1329 | +--------------+ +--------------+ | 1330 | | 1331 +--------------------------------------------------------------+ 1333 Figure 5: FE Entity Hierarchy 1335 At the top of the addressing hierachy is the FE identifier. In the 1336 example above, the 32-bit FE identifier is illustrated with the 1337 mnemonic FE01. The next 32-bit entity selector is the LFB ClassID. 1338 In the illustration above, two LFB classes with identifiers 1 and 91 1339 are demonstrated. The example above further illustrates one instance 1340 of each of the two classes. The scope of the 32-bit LFB class 1341 instance identifier is valid only within the LFB class. To emphasize 1342 that point, each of class 1 and 91 has an instance of 3. 1344 Using the described addressing scheme, a message could be sent to 1345 address FE01, LFB ClassID 1, LFB InstanceID 3, utilizing the ForCES 1346 protocol. However, to be effective, such a message would have to 1347 target entities within an LFB. These entities could be carrying 1348 state, capability, etc. These are further illustrated in Figure 6 1349 below. 1351 LFB Class ID 1,InstanceID 3 Components 1352 +-------------------------------------+ 1353 | | 1354 | LFB ComponentID 1 | 1355 | +----------------------+ | 1356 | | | | 1357 | +----------------------+ | 1358 | | 1359 | LFB ComponentID 31 | 1360 | +----------------------+ | 1361 | | | | 1362 | +----------------------+ | 1363 | | 1364 | LFB ComponentID 51 | 1365 | +----------------------+ | 1366 | | LFB ComponentID 89 | | 1367 | | +-----------------+ | | 1368 | | | | | | 1369 | | +-----------------+ | | 1370 | +----------------------+ | 1371 | | 1372 | | 1373 +-------------------------------------+ 1375 Figure 6: LFB Hierarchy 1377 Figure 6 zooms into the components carried by LFB Class ID 1, LFB 1378 InstanceID 3 from Figure 5. 1380 The example shows three components with 32-bit component identifiers 1381 1, 31, and 51. LFB ComponentID 51 is a complex structure 1382 encapsulating within it an entity with LFB ComponentID 89. LFB 1383 ComponentID 89 could be a complex structure itself but is restricted 1384 in the example for the sake of clarity. 1386 3.3.1. Addressing LFB Components: Paths and Keys 1388 As mentioned above, LFB components could be complex structures, such 1389 as a table, or even more complex structures such as a table whose 1390 cells are further tables, etc. The ForCES model XML schema 1391 (Section 4) allows for uniquely identifying anything with such 1392 complexity, utilizing the concept of dot-annotated static paths and 1393 content addressing of paths as derived from keys. As an example, if 1394 the LFB Component 51 were a structure, then the path to LFB 1395 ComponentID 89 above will be 51.89. 1397 LFB ComponentID 51 might represent a table (an array). In that case, 1398 to select the LFB Component with ID 89 from within the 7th entry of 1399 the table, one would use the path 51.7.89. In addition to supporting 1400 explicit table element selection by including an index in the dotted 1401 path, the model supports identifying table elements by their 1402 contents. This is referred to as using keys, or key indexing. So, 1403 as a further example, if ComponentID 51 was a table which was key 1404 index-able, then a key describing content could also be passed by the 1405 CE, along with path 51 to select the table, and followed by the path 1406 89 to select the table structure element, which upon computation by 1407 the FE would resolve to the LFB ComponentID 89 within the specified 1408 table entry. 1410 3.4. FE Datapath Modeling 1412 Packets coming into the FE from ingress ports generally flow through 1413 one or more LFBs before leaving out of the egress ports. How an FE 1414 treats a packet depends on many factors, such as type of the packet 1415 (e.g., IPv4, IPv6, or MPLS), header values, time of arrival, etc. 1416 The result of LFB processing may have an impact on how the packet is 1417 to be treated in downstream LFBs. This differentiation of packet 1418 treatment downstream can be conceptualized as having alternative 1419 datapaths in the FE. For example, the result of a 6-tuple 1420 classification performed by a classifier LFB could control which rate 1421 meter is applied to the packet by a rate meter LFB in a later stage 1422 in the datapath. 1424 LFB topology is a directed graph representation of the logical 1425 datapaths within an FE, with the nodes representing the LFB instances 1426 and the directed link depicting the packet flow direction from one 1427 LFB to the next. Section 3.4.1 discusses how the FE datapaths can be 1428 modeled as LFB topology; while Section 3.4.2 focuses on issues 1429 related to LFB topology reconfiguration. 1431 3.4.1. Alternative Approaches for Modeling FE Datapaths 1433 There are two basic ways to express the differentiation in packet 1434 treatment within an FE, one represents the datapath directly and 1435 graphically (topological approach) and the other utilizes metadata 1436 (the encoded state approach). 1438 o Topological Approach 1440 Using this approach, differential packet treatment is expressed by 1441 splitting the LFB topology into alternative paths. In other words, 1442 if the result of an LFB operation controls how the packet is further 1443 processed, then such an LFB will have separate output ports, one for 1444 each alternative treatment, connected to separate sub-graphs, each 1445 expressing the respective treatment downstream. 1447 o Encoded State Approach 1449 An alternate way of expressing differential treatment is by using 1450 metadata. The result of the operation of an LFB can be encoded in a 1451 metadatum, which is passed along with the packet to downstream LFBs. 1452 A downstream LFB, in turn, can use the metadata and its value (e.g., 1453 as an index into some table) to determine how to treat the packet. 1455 Theoretically, either approach could substitute for the other, so one 1456 could consider using a single pure approach to describe all datapaths 1457 in an FE. However, neither model by itself results in the best 1458 representation for all practically relevant cases. For a given FE 1459 with certain logical datapaths, applying the two different modeling 1460 approaches will result in very different looking LFB topology graphs. 1461 A model using only the topological approach may require a very large 1462 graph with many links or paths, and nodes (i.e., LFB instances) to 1463 express all alternative datapaths. On the other hand, a model using 1464 only the encoded state model would be restricted to a string of LFBs, 1465 which is not an intuitive way to describe different datapaths (such 1466 as MPLS and IPv4). Therefore, a mix of these two approaches will 1467 likely be used for a practical model. In fact, as we illustrate 1468 below, the two approaches can be mixed even within the same LFB. 1470 Using a simple example of a classifier with N classification outputs 1471 followed by other LFBs, Figure 7.a shows what the LFB topology looks 1472 like when using the pure topological approach. Each output from the 1473 classifier goes to one of the N LFBs where no metadata is needed. 1474 The topological approach is simple, straightforward and graphically 1475 intuitive. However, if N is large and the N nodes following the 1476 classifier (LFB#1, LFB#2, ..., LFB#N) all belong to the same LFB type 1477 (e.g., meter), but each has its own independent components, the 1478 encoded state approach gives a much simpler topology representation, 1479 as shown in Figure 7.b. The encoded state approach requires that a 1480 table of N rows of meter components is provided in the Meter node 1481 itself, with each row representing the attributes for one meter 1482 instance. A metadatum M is also needed to pass along with the packet 1483 P from the classifier to the meter, so that the meter can use M as a 1484 look-up key (index) to find the corresponding row of the attributes 1485 that should be used for any particular packet P. 1487 What if those N nodes (LFB#1, LFB#2, ..., LFB#N) are not of the same 1488 type? For example, if LFB#1 is a queue while the rest are all 1489 meters, what is the best way to represent such datapaths? While it 1490 is still possible to use either the pure topological approach or the 1491 pure encoded state approach, the natural combination of the two 1492 appears to be the best option. Figure 7.c depicts two different 1493 functional datapaths using the topological approach while leaving the 1494 N-1 meter instances distinguished by metadata only, as shown in 1495 Figure 7.c. 1497 +----------+ 1498 P | LFB#1 | 1499 +--------->|(Compon-1)| 1500 +-------------+ | +----------+ 1501 | 1|------+ P +----------+ 1502 | 2|---------------->| LFB#2 | 1503 | classifier 3| |(Compon-2)| 1504 | ...|... +----------+ 1505 | N|------+ ... 1506 +-------------+ | P +----------+ 1507 +--------->| LFB#N | 1508 |(Compon-N)| 1509 +----------+ 1511 (a) Using pure topological approach 1513 +-------------+ +-------------+ 1514 | 1| | Meter | 1515 | 2| (P, M) | (Compon-1) | 1516 | 3|---------------->| (Compon-2) | 1517 | ...| | ... | 1518 | N| | (Compon-N) | 1519 +-------------+ +-------------+ 1521 (b) Using pure encoded state approach to represent the LFB 1522 topology in 5(a), if LFB#1, LFB#2, ..., and LFB#N are of the 1523 same type (e.g., meter). 1525 +-------------+ 1526 +-------------+ (P, M) | queue | 1527 | 1|------------->| (Compon-1) | 1528 | 2| +-------------+ 1529 | 3| (P, M) +-------------+ 1530 | ...|------------->| Meter | 1531 | N| | (Compon-2) | 1532 +-------------+ | ... | 1533 | (Compon-N) | 1534 +-------------+ 1536 (c) Using a combination of the two, if LFB#1, LFB#2, ..., and 1537 LFB#N are of different types (e.g., queue and meter). 1539 Figure 7: An example of how to model FE datapaths 1541 From this example, we demonstrate that each approach has a distinct 1542 advantage depending on the situation. Using the encoded state 1543 approach, fewer connections are typically needed between a fan-out 1544 node and its next LFB instances of the same type because each packet 1545 carries metadata the following nodes can interpret and hence invoke a 1546 different packet treatment. For those cases, a pure topological 1547 approach forces one to build elaborate graphs with many more 1548 connections and often results in an unwieldy graph. On the other 1549 hand, a topological approach is the most intuitive for representing 1550 functionally different datapaths. 1552 For complex topologies, a combination of the two is the most 1553 flexible. A general design guideline is provided to indicate which 1554 approach is best used for a particular situation. The topological 1555 approach should primarily be used when the packet datapath forks to 1556 distinct LFB classes (not just distinct parameterizations of the same 1557 LFB class), and when the fan-outs do not require changes, such as 1558 adding/removing LFB outputs, or require only very infrequent changes. 1559 Configuration information that needs to change frequently should be 1560 expressed by using the internal attributes of one or more LFBs (and 1561 hence using the encoded state approach). 1563 +---------------------------------------------+ 1564 | | 1565 +----------+ V +----------+ +------+ | 1566 | | | | |if IP-in-IP| | | 1567 ---->| ingress |->+----->|classifier|---------->|Decap.|---->---+ 1568 | ports | | |---+ | | 1569 +----------+ +----------+ |others +------+ 1570 | 1571 V 1572 (a) The LFB topology with a logical loop 1574 +-------+ +-----------+ +------+ +-----------+ 1575 | | | |if IP-in-IP | | | | 1576 --->|ingress|-->|classifier1|----------->|Decap.|-->+classifier2|-> 1577 | ports | | |----+ | | | | 1578 +-------+ +-----------+ |others +------+ +-----------+ 1579 | 1580 V 1581 (b)The LFB topology without the loop utilizing two independent 1582 classifier instances. 1584 Figure 8: An LFB topology example. 1586 It is important to point out that the LFB topology described here is 1587 the logical topology, not the physical topology of how the FE 1588 hardware is actually laid out. Nevertheless, the actual 1589 implementation may still influence how the functionality is mapped to 1590 the LFB topology. Figure 8 shows one simple FE example. In this 1591 example, an IP-in-IP packet from an IPSec application like VPN may go 1592 to the classifier first and have the classification done based on the 1593 outer IP header; upon being classified as an IP-in-IP packet, the 1594 packet is then sent to a decapsulator to strip off the outer IP 1595 header, followed by a classifier again to perform classification on 1596 the inner IP header. If the same classifier hardware or software is 1597 used for both outer and inner IP header classification with the same 1598 set of filtering rules, a logical loop is naturally present in the 1599 LFB topology, as shown in Figure 8.a. However, if the classification 1600 is implemented by two different pieces of hardware or software with 1601 different filters (i.e., one set of filters for the outer IP header 1602 and another set for the inner IP header), then it is more natural to 1603 model them as two different instances of classifier LFB, as shown in 1604 Figure 8.b. 1606 3.4.2. Configuring the LFB Topology 1608 While there is little doubt that an individual LFB must be 1609 configurable, the configurability question is more complicated for 1610 LFB topology. Since the LFB topology is really the graphic 1611 representation of the datapaths within an FE, configuring the LFB 1612 topology means dynamically changing the datapaths, including changing 1613 the LFBs along the datapaths on an FE (e.g., creating/instantiating, 1614 updating or deleting LFBs) and setting up or deleting 1615 interconnections between outputs of upstream LFBs to inputs of 1616 downstream LFBs. 1618 Why would the datapaths on an FE ever change dynamically? The 1619 datapaths on an FE are set up by the CE to provide certain data plane 1620 services (e.g., DiffServ, VPN, etc.) to the Network Element's (NE) 1621 customers. The purpose of reconfiguring the datapaths is to enable 1622 the CE to customize the services the NE is delivering at run time. 1623 The CE needs to change the datapaths when the service requirements 1624 change, such as adding a new customer or when an existing customer 1625 changes their service. However, note that not all datapath changes 1626 result in changes in the LFB topology graph. Changes in the graph 1627 are dependent on the approach used to map the datapaths into LFB 1628 topology. As discussed in Section 3.4.1, the topological approach 1629 and encoded state approach can result in very different looking LFB 1630 topologies for the same datapaths. In general, an LFB topology based 1631 on a pure topological approach is likely to experience more frequent 1632 topology reconfiguration than one based on an encoded state approach. 1633 However, even an LFB topology based entirely on an encoded state 1634 approach may have to change the topology at times, for example, to 1635 bypass some LFBs or insert new LFBs. Since a mix of these two 1636 approaches is used to model the datapaths, LFB topology 1637 reconfiguration is considered an important aspect of the FE model. 1639 We want to point out that allowing a configurable LFB topology in the 1640 FE model does not mandate that all FEs are required to have this 1641 capability. Even if an FE supports configurable LFB topology, the FE 1642 may impose limitations on what can actually be configured. 1643 Performance-optimized hardware implementations may have zero or very 1644 limited configurability, while FE implementations running on network 1645 processors may provide more flexibility and configurability. It is 1646 entirely up to the FE designers to decide whether or not the FE 1647 actually implements reconfiguration and if so, how much. Whether a 1648 simple runtime switch is used to enable or disable (i.e., bypass) 1649 certain LFBs, or more flexible software reconfiguration is used, is 1650 an implementation detail internal to the FE and outside of the scope 1651 of FE model. In either case, the CE(s) MUST be able to learn the 1652 FE's configuration capabilities. Therefore, the FE model MUST 1653 provide a mechanism for describing the LFB topology configuration 1654 capabilities of an FE. These capabilities may include (see Section 5 1655 for full details): 1657 o Which LFB classes the FE can instantiate 1659 o The maximum number of instances of the same LFB class that can be 1660 created 1662 o Any topological limitations, for example: 1664 * The maximum number of instances of the same class or any class 1665 that can be created on any given branch of the graph 1667 * Ordering restrictions on LFBs (e.g., any instance of LFB class 1668 A must be always downstream of any instance of LFB class B). 1670 The CE needs some programming help in order to cope with the range of 1671 complexity. In other words, even when the CE is allowed to configure 1672 LFB topology for the FE, the CE is not expected to be able to 1673 interpret an arbitrary LFB topology and determine which specific 1674 service or application (e.g. VPN, DiffServ, etc.) is supported by 1675 the FE. However, once the CE understands the coarse capability of an 1676 FE, the CE MUST configure the LFB topology to implement the network 1677 service the NE is supposed to provide. Thus, the mapping the CE has 1678 to understand is from the high level NE service to a specific LFB 1679 topology, not the other way around. The CE is not expected to have 1680 the ultimate intelligence to translate any high level service policy 1681 into the configuration data for the FEs. However, it is conceivable 1682 that within a given network service domain, a certain amount of 1683 intelligence can be programmed into the CE to give the CE a general 1684 understanding of the LFBs involved to allow the translation from a 1685 high level service policy to the low level FE configuration to be 1686 done automatically. Note that this is considered an implementation 1687 issue internal to the control plane and outside the scope of the FE 1688 model. Therefore, it is not discussed any further in this draft. 1690 +----------+ +-----------+ 1691 ---->| Ingress |---->|classifier |--------------+ 1692 | | |chip | | 1693 +----------+ +-----------+ | 1694 v 1695 +-------------------------------------------+ 1696 +--------+ | Network Processor | 1697 <----| Egress | | +------+ +------+ +-------+ | 1698 +--------+ | |Meter | |Marker| |Dropper| | 1699 ^ | +------+ +------+ +-------+ | 1700 | | | 1701 +----------+-------+ | 1702 | | | 1703 | +---------+ +---------+ +------+ +---------+ | 1704 | |Forwarder|<------|Scheduler|<--|Queue | |Counter | | 1705 | +---------+ +---------+ +------+ +---------+ | 1706 +--------------------------------------------------------------+ 1708 Figure 9: The Capability of an FE as reported to the CE 1710 Figure 9 shows an example where a QoS-enabled router has several line 1711 cards that have a few ingress ports and egress ports, a specialized 1712 classification chip, and a network processor containing codes for FE 1713 blocks like meter, marker, dropper, counter, queue, scheduler, and 1714 IPv4 forwarder. Some of the LFB topology is already fixed and has to 1715 remain static due to the physical layout of the line cards. For 1716 example, all of the ingress ports might be hardwired into the 1717 classification chip so all packets flow from the ingress port into 1718 the classification engine. On the other hand, the LFBs on the 1719 network processor and their execution order are programmable. 1720 However, certain capacity limits and linkage constraints could exist 1721 between these LFBs. Examples of the capacity limits might be: 1723 o 8 meters 1725 o 16 queues in one FE 1727 o the scheduler can handle at most up to 16 queues 1728 o The linkage constraints might dictate that: 1730 * the classification engine may be followed by: 1732 + a meter 1734 + marker 1736 + dropper 1738 + counter 1740 + queue or IPv4 forwarder, but not a scheduler 1742 * queues can only be followed by a scheduler 1744 * a scheduler must be followed by the IPv4 forwarder 1746 * the last LFB in the datapath before going into the egress ports 1747 must be the IPv4 forwarder 1749 +-----+ +-------+ +---+ 1750 | A|--->|Queue1 |--------------------->| | 1751 ------>| | +-------+ | | +---+ 1752 | | | | | | 1753 | | +-------+ +-------+ | | | | 1754 | B|--->|Meter1 |----->|Queue2 |------>| |->| | 1755 | | | | +-------+ | | | | 1756 | | | |--+ | | | | 1757 +-----+ +-------+ | +-------+ | | +---+ 1758 classifier +-->|Dropper| | | IPv4 1759 +-------+ +---+ Fwd. 1760 Scheduler 1762 Figure 10: An LFB topology as configured by the CE and accepted by 1763 the FE 1765 Once the FE reports these capabilities and capacity limits to the CE, 1766 it is now up to the CE to translate the QoS policy into a desirable 1767 configuration for the FE. Figure 9 depicts the FE capability while 1768 Figure 10 and Figure 11 depict two different topologies that the CE 1769 may request the FE to configure. Note that Figure 11 is not fully 1770 drawn, as inter-LFB links are included to suggest potential 1771 complexity, without drawing in the endpoints of all such links. 1773 Queue1 1774 +---+ +--+ 1775 | A|------------------->| |--+ 1776 +->| | | | | 1777 | | B|--+ +--+ +--+ +--+ | 1778 | +---+ | | | | | | 1779 | Meter1 +->| |-->| | | 1780 | | | | | | 1781 | +--+ +--+ | Ipv4 1782 | Counter1 Dropper1 Queue2| +--+ Fwd. 1783 +---+ | +--+ +--->|A | +-+ 1784 | A|---+ | |------>|B | | | 1785 ------>| B|------------------------------>| | +-->|C |->| |-> 1786 | C|---+ +--+ | +>|D | | | 1787 | D|-+ | | | +--+ +-+ 1788 +---+ | | +---+ Queue3 | |Scheduler 1789 Classifier1 | | | A|------------> +--+ | | 1790 | +->| | | |-+ | 1791 | | B|--+ +--+ +-------->| | | 1792 | +---+ | | | | +--+ | 1793 | Meter2 +->| |-+ | 1794 | | | | 1795 | +--+ Queue4 | 1796 | Marker1 +--+ | 1797 +---------------------------->| |---+ 1798 | | 1799 +--+ 1801 Figure 11: Another LFB topology as configured by the CE and accepted 1802 by the FE 1804 Note that both the ingress and egress are omitted in Figure 10 and 1805 Figure 11 to simplify the representation. The topology in Figure 11 1806 is considerably more complex than Figure 10 but both are feasible 1807 within the FE capabilities, and so the FE should accept either 1808 configuration request from the CE. 1810 4. Model and Schema for LFB Classes 1812 The main goal of the FE model is to provide an abstract, generic, 1813 modular, implementation-independent representation of the FEs. This 1814 is facilitated using the concept of LFBs, which are instantiated from 1815 LFB classes. LFB classes and associated definitions will be provided 1816 in a collection of XML documents. The collection of these XML 1817 documents is called a LFB class library, and each document is called 1818 an LFB class library document (or library document, for short). Each 1819 of the library documents MUST conform to the schema presented in this 1820 section. The root element of the library document is the 1821 element. 1823 It is not expected that library documents will be exchanged between 1824 FEs and CEs "over-the-wire". But the model will serve as an 1825 important reference for the design and development of the CEs 1826 (software) and FEs (mostly the software part). It will also serve as 1827 a design input when specifying the ForCES protocol elements for CE-FE 1828 communication. 1830 The following sections describe the portions of an LFBLibrary XML 1831 Document. The descriptions primarily provide the necessary semantic 1832 information to understand the meaning and uses of the XML elements. 1833 The XML Schema below provides the final definition on what elements 1834 are permitted, and their base syntax. Unfortunately, due to the 1835 limitations of english and XML, there are constraints described in 1836 the semantic sections which are not fully captured in the XML Schema, 1837 so both sets of information need to be used to build a compliant 1838 library document. 1840 4.1. Namespace 1842 A namespace is needed to uniquely identify the LFB type in the LFB 1843 class library. The reference to the namespace definition is 1844 contained in Section 9, IANA Considerations. 1846 4.2. Element 1848 The element serves as a root element of all library 1849 documents. A library document contains a sequence of top level 1850 elements. The following is a list of all the elements which can 1851 occur directly in the element. If they occur, they must 1852 occur in the order listed. 1854 o providing a text description of the purpose of the 1855 library document. 1857 o for loading information from other library documents. 1859 o for the frame declarations; 1861 o for defining common data types; 1863 o for defining metadata, and 1865 o for defining LFB classes. 1867 Each element is optional. One library document may contain only 1868 metadata definitions, another may contain only LFB class definitions, 1869 yet another may contain all of the above. 1871 A library document can import other library documents if it needs to 1872 refer to definitions contained in the included document. This 1873 concept is similar to the "#include" directive in C. Importing is 1874 expressed by the use of elements, which must precede all the 1875 above elements in the document. For unique referencing, each 1876 LFBLibrary instance document has a unique label defined in the 1877 "provide" attribute of the LFBLibrary element. Note that what this 1878 performs is a ForCES inclusion, not an XML inclusion. The semantic 1879 content of the library referenced by the element is included, 1880 not the xml content. Also, in terms of the conceptual processing of 1881 elements, the total set of documents loaded are considered to 1882 form a single document for processing. A given document is included 1883 in this set only once, even if it is referenced by elements 1884 several times, even from several different files. As the processing 1885 of LFBLibrary information is not order dependent, the order for 1886 processing loaded elements is up to the implementor, as long as the 1887 total effect is as if all of the information from all the files were 1888 available for referencing when needed. Note that such computer 1889 processing of ForCES model library documents may be helpful for 1890 various implementations, but is not required to define the libraries, 1891 or for the actual operation of the protocol itself. 1893 The following is a skeleton of a library document: 1895 1896 1899 1901 1903 1904 1905 ... 1907 1908 1909 ... 1910 1912 1913 1914 ... 1915 1917 1918 1919 ... 1920 1922 1926 1928 1929 1931 4.3. Element 1933 This element is used to refer to another LFB library document. 1934 Similar to the "#include" directive in C, this makes the objects 1935 (metadata types, data types, etc.) defined in the referred library 1936 document available for referencing in the current document. 1938 The load element MUST contain the label of the library document to be 1939 included and MAY contain a URL to specify where the library can be 1940 retrieved. The load element can be repeated unlimited times. Three 1941 examples for the elements: 1943 1944 1945 1948 4.4. Element for Frame Type Declarations 1950 Frame names are used in the LFB definition to define the types of 1951 frames the LFB expects at its input port(s) and emits at its output 1952 port(s). The optional element in the library document 1953 contains one or more elements, each declaring one frame 1954 type. 1956 Each frame definition MUST contain a unique name (NMTOKEN) and a 1957 brief synopsis. In addition, an optional detailed description MAY be 1958 provided. 1960 Uniqueness of frame types MUST be ensured among frame types defined 1961 in the same library document and in all directly or indirectly 1962 included library documents. 1964 The following example defines two frame types: 1966 1967 1968 ipv4 1969 IPv4 packet 1970 1971 This frame type refers to an IPv4 packet. 1972 1973 1974 1975 ipv6 1976 IPv6 packet 1977 1978 This frame type refers to an IPv6 packet. 1979 1980 1981 ... 1982 1984 4.5. Element for Data Type Definitions 1986 The (optional) element can be used to define commonly 1987 used data types. It contains one or more elements, 1988 each defining a data type with a unique name. Such data types can be 1989 used in several places in the library documents, including: 1991 o Defining other data types 1993 o Defining components of LFB classes 1995 This is similar to the concept of having a common header file for 1996 shared data types. 1998 Each element MUST contain a unique name (NMTOKEN), a 1999 brief synopsis, and a type definition element. The name MUST be 2000 unique among all data types defined in the same library document and 2001 in any directly or indirectly included library documents. The 2002 element MAY also include an optional longer 2003 description, For example: 2005 2006 2007 ieeemacaddr 2008 48-bit IEEE MAC address 2009 ... type definition ... 2010 2011 2012 ipv4addr 2013 IPv4 address 2014 ... type definition ... 2015 2016 ... 2017 2019 There are two kinds of data types: atomic and compound. Atomic data 2020 types are appropriate for single-value variables (e.g. integer, 2021 string, byte array). 2023 The following built-in atomic data types are provided, but additional 2024 atomic data types can be defined with the and 2025 elements: 2027 Meaning 2028 ---- ------- 2029 char 8-bit signed integer 2030 uchar 8-bit unsigned integer 2031 int16 16-bit signed integer 2032 uint16 16-bit unsigned integer 2033 int32 32-bit signed integer 2034 uint32 32-bit unsigned integer 2035 int64 64-bit signed integer 2036 uint64 64-bit unsigned integer 2037 boolean A true / false value where 2038 0 = false, 1 = true 2039 string[N] A UTF-8 string represented in at most 2040 N Octets. 2041 string A UTF-8 string without a configured 2042 storage length limit. 2043 byte[N] A byte array of N bytes 2044 octetstring[N] A buffer of N octets, which MAY 2045 contain fewer than N octets. Hence 2046 the encoded value will always have 2047 a length. 2048 float16 16-bit floating point number 2049 float32 32-bit IEEE floating point number 2050 float64 64-bit IEEE floating point number 2052 These built-in data types can be readily used to define metadata or 2053 LFB attributes, but can also be used as building blocks when defining 2054 new data types. The boolean data type is defined here because it is 2055 so common, even though it can be built by sub-ranging the uchar data 2056 type, as defined under atomic types (Section 4.5.2). 2058 Compound data types can build on atomic data types and other compound 2059 data types. Compound data types can be defined in one of four ways. 2060 They may be defined as an array of components of some compound or 2061 atomic data type. They may be a structure of named components of 2062 compound or atomic data types (c.f. C structures). They may be a 2063 union of named components of compound or atomic data types (c.f. C 2064 unions). They may also be defined as augmentations (explained in 2065 Section 4.5.7) of existing compound data types. 2067 Given that the ForCES protocol will be getting and setting component 2068 values, all atomic data types used here must be able to be conveyed 2069 in the ForCES protocol. Further, the ForCES protocol will need a 2070 mechanism to convey compound data types. However, the details of 2071 such representations are for the ForCES Protocol [2] document to 2072 define, not the model document. Strings and octetstrings must be 2073 conveyed by the protocol with their length, as they are not 2074 delimited, the value does not itself include the length, and these 2075 items are variable length. 2077 With regard to strings, this model defines a small set of 2078 restrictions and definitions on how they are structured. String and 2079 octetstring length limits can be specified in the LFB Class 2080 definitions. The component properties for string and octetstring 2081 components also contain actual lengths and length limits. This 2082 duplication of limits is to allow for implementations with smaller 2083 limits than the maximum limits specified in the LFB Class definition. 2084 In all cases, these lengths are specified in octets, not in 2085 characters. In terms of protocol operation, as long as the specified 2086 length is within the FE's supported capabilities, the FE stores the 2087 contents of a string exactly as provided by the CE, and returns those 2088 contents when requested. No canonicalization, transformations, or 2089 equivalences are performed by the FE. Components of type string (or 2090 string[n]) MAY be used to hold identifiers for correlation with 2091 components in other LFBs. In such cases, an exact octet for octet 2092 match is used. No equivalences are used by the FE or CE in 2093 performing that matching. The ForCES Protocol [2] does not perform 2094 or require validation of the content of UTF-8 strings. However, 2095 UTF-8 strings SHOULD be encoded in the shortest form to avoid 2096 potential security issues described in [12]. Any entity displaying 2097 such strings is expected to perform its own validation (for example 2098 for correct multi-byte characters, and for ensuring that the string 2099 does not end in the middle of a multi-byte sequence.) Specific LFB 2100 class definitions MAY restrict the valid contents of a string as 2101 suited to the particular usage (for example, a component that holds a 2102 DNS name would be restricted to hold only octets valid in such a 2103 name.) FEs should validate the contents of SET requests for such 2104 restricted components at the time the set is performed, just as range 2105 checks for range limited components are performed. The ForCES 2106 protocol behavior defines the normative processing for requests using 2107 that protocol. 2109 For the definition of the actual type in the element, 2110 the following elements are available: , , , 2111 , and . 2113 The predefined type alias is somewhere between the atomic and 2114 compound data types. Alias is used to allow a component inside an 2115 LFB to be an indirect reference to another component inside the same 2116 or a different LFB class or instance. The alias component behaves 2117 like a structure, one component of which has special behavior. Given 2118 that the special behavior is tied to the other parts of the 2119 structure, the compound result is treated as a predefined construct. 2121 4.5.1. Element for Renaming Existing Data Types 2123 The element refers to an existing data type by its name. 2124 The referred data type MUST be defined either in the same library 2125 document, or in one of the included library documents. If the 2126 referred data type is an atomic data type, the newly defined type 2127 will also be regarded as atomic. If the referred data type is a 2128 compound type, the new type will also be compound. Some usage 2129 examples follow: 2131 2132 short 2133 Alias to int16 2134 int16 2135 2136 2137 ieeemacaddr 2138 48-bit IEEE MAC address 2139 byte[6] 2140 2142 4.5.2. Element for Deriving New Atomic Types 2144 The element allows the definition of a new atomic type from 2145 an existing atomic type, applying range restrictions and/or providing 2146 special enumerated values. Note that the element can only 2147 use atomic types as base types, and its result MUST be another atomic 2148 type. 2150 For example, the following snippet defines a new "dscp" data type: 2152 2153 dscp 2154 Diffserv code point. 2155 2156 uchar 2157 2158 2159 2160 2161 2162 DSCP-BE 2163 Best Effort 2164 2165 ... 2166 2167 2168 2170 4.5.3. Element to Define Arrays 2172 The element can be used to create a new compound data type as 2173 an array of a compound or an atomic data type. Depending upon 2174 context, this document, and others, refer to such arrays as tables or 2175 arrays interchangeably, without semantic or syntactic implication. 2176 The type of the array entry can be specified either by referring to 2177 an existing type (using the element) or defining an unnamed 2178 type inside the element using any of the , , 2179 , or elements. 2181 The array can be "fixed-size" or "variable-size", which is specified 2182 by the "type" attribute of the element. The default is 2183 "variable-size". For variable size arrays, an optional "maxlength" 2184 attribute specifies the maximum allowed length. This attribute 2185 should be used to encode semantic limitations, not implementation 2186 limitations. The latter (support for implementation constraints) 2187 should be handled by capability components of LFB classes, and should 2188 never be included as the maxlength in a data type array which is 2189 regarded as being of unlimited size. 2191 For fixed-size arrays, a "length" attribute MUST be provided that 2192 specifies the constant size of the array. 2194 The result of this construct is always a compound type, even if the 2195 array has a fixed size of 1. 2197 Arrays MUST only be subscripted by integers, and will be presumed to 2198 start with index 0. 2200 In addition to their subscripts, arrays MAY be declared to have 2201 content keys. Such a declaration has several effects: 2203 o Any declared key can be used in the ForCES protocol to select a 2204 component for operations (for details, see the ForCES Protocol 2205 [2]). 2207 o In any instance of the array, each declared key MUST be unique 2208 within that instance. That is, no two components of an array may 2209 have the same values on all the fields which make up a key. 2211 Each key is declared with a keyID for use in the ForCES Protocol [2], 2212 where the unique key is formed by combining one or more specified key 2213 fields. To support the case where an array of an atomic type with 2214 unique values can be referenced by those values, the key field 2215 identifier MAY be "*" (i.e., the array entry is the key). If the 2216 value type of the array is a structure or an array, then the key is 2217 one or more components of the value type, each identified by name. 2219 Since the field MAY be a component of the contained structure, a 2220 component of a component of a structure, or further nested, the field 2221 name is actually a concatenated sequence of component identifiers, 2222 separated by decimal points ("."). The syntax for key field 2223 identification is given following the array examples. 2225 The following example shows the definition of a fixed size array with 2226 a pre-defined data type as the array content type: 2228 2229 dscp-mapping-table 2230 2231 A table of 64 DSCP values, used to re-map code space. 2232 2233 2234 dscp 2235 2236 2238 The following example defines a variable size array with an upper 2239 limit on its size: 2241 2242 mac-alias-table 2243 A table with up to 8 IEEE MAC addresses 2244 2245 ieeemacaddr 2246 2247 2249 The following example shows the definition of an array with a local 2250 (unnamed) content type definition: 2252 2253 classification-table 2254 2255 A table of classification rules and result opcodes. 2256 2257 2258 2259 2260 rule 2261 The rule to match 2262 classrule 2263 2264 2265 opcode 2266 The result code 2267 opcode 2268 2269 2270 2271 2273 In the above example, each entry of the array is a of two 2274 components ("rule" and "opcode"). 2276 The following example shows a table of IP Prefix information that can 2277 be accessed by a multi-field content key on the IP Address, prefix 2278 length, and information source. This means that in any instance of 2279 this table, no two entries can have the same IP address, prefix 2280 length, and information source. 2282 2283 ipPrefixInfo_table 2284 2285 A table of information about known prefixes 2286 2287 2288 2289 2290 address-prefix 2291 the prefix being described 2292 ipv4Prefix 2293 2294 2295 source 2296 2297 the protocol or process providing this information 2298 2299 uint16 2300 2301 2302 prefInfo 2303 the information we care about 2304 hypothetical-info-type 2305 2306 2307 2308 address-prefix.ipv4addr 2309 address-prefix.prefixlen 2310 source 2311 2312 2313 2315 Note that the keyField elements could also have been simply address- 2316 prefix and source, since all of the fields of address-prefix are 2317 being used. 2319 4.5.3.1. Key Field References 2321 In order to use key declarations, one must refer to components that 2322 are potentially nested inside other components in the array. If 2323 there are nested arrays, one might even use an array element as a key 2324 (but great care would be needed to ensure uniqueness.) 2326 The key is the combination of the values of each field declared in a 2327 keyField element. 2329 Therefore, the value of a keyField element MUST be a concatenated 2330 Sequence of field identifiers, separated by a "." (period) character. 2331 Whitespace is permitted and ignored. 2333 A valid string for a single field identifier within a keyField 2334 depends upon the current context. Initially, in an array key 2335 declaration, the context is the type of the array. Progressively, 2336 the context is whatever type is selected by the field identifiers 2337 processed so far in the current key field declaration. 2339 When the current context is an array, (e.g., when declaring a key for 2340 an array whose content is an array) then the only valid value for the 2341 field identifier is an explicit number. 2343 When the current context is a structure, the valid values for the 2344 field identifiers are the names of the components of the structure. 2345 In the special case of declaring a key for an array containing an 2346 atomic type, where that content is unique and is to be used as a key, 2347 the value "*" MUST be used as the single key field identifier. 2349 In reference array or structure elements, it is possible to construct 2350 keyFields that do not exist. keyField references SHOULD never 2351 reference optional structure components. For references to array 2352 elements, care must be taken to ensure that the necessary array 2353 elements exist when creating or modifying the overall array element. 2354 Failure to do so will result in FEs returning errors on the creation 2355 attempt. 2357 4.5.4. Element to Define Structures 2359 A structure is comprised of a collection of data components. Each 2360 data components has a data type (either an atomic type or an existing 2361 compound type) and is assigned a name unique within the scope of the 2362 compound data type being defined. These serve the same function as 2363 "struct" in C, etc. These components are defined using 2364 elements. A element MAY contain an optional derivation 2365 indication, a element. The structure definition MUST 2366 contain a sequence of one or more elements. 2368 The actual type of the component can be defined by referring to an 2369 existing type (using the element), or can be a locally 2370 defined (unnamed) type created by any of the , , 2371 , or elements. 2373 The element MUST include a componentID attribute. This 2374 provides the numeric ID for this component, for use by the protocol. 2375 The MUST contain a component name and a synopsis. It MAY 2376 contain a element giving a textual description of the 2377 component. The definition MAY also include a element, 2378 which indicates that the component being defined is optional. The 2379 definition MUST contain elements to define the data type of the 2380 component, as described above. 2382 For a dataTypeDef of a struct, the structure definition MAY be 2383 inherited from, and augment, a previously defined structured type. 2384 This is indicated by including the optional derivedFrom attribute in 2385 the struct declaration before the definition of the augmenting or 2386 replacing components. The augmentation (Section 4.5.7) section 2387 describes how this is done in more detail. 2389 The componentID attribute for different components in a structure (or 2390 in an LFB) MUST be distinct. They do not need to be in order, nor do 2391 they need to be sequential. For clarity of human readability, and 2392 ease of maintanence, it is usual to define at least sequential sets 2393 of values. But this is for human ease, not a model or protocol 2394 requirement. 2396 The result of this construct is always a compound type, even when the 2397 contains only one field. 2399 An example: 2401 2402 ipv4prefix 2403 2404 IPv4 prefix defined by an address and a prefix length 2405 2406 2407 2408 address 2409 Address part 2410 ipv4addr 2411 2412 2413 prefixlen 2414 Prefix length part 2415 2416 uchar 2417 2418 2419 2420 2421 2422 2423 2425 4.5.5. Element to Define Union Types 2427 Similar to the union declaration in C, this construct allows the 2428 definition of overlay types. Its format is identical to the 2429 element. 2431 The result of this construct is always a compound type, even when the 2432 union contains only one element. 2434 4.5.6. Element 2436 It is sometimes necessary to have a component in an LFB or structure 2437 refer to information (a component) in other LFBs. This can, for 2438 example, allow an ARP LFB to share the IP->MAC Address table with the 2439 local transmission LFB, without duplicating information. Similarly, 2440 it could allow a traffic measurement LFB to share information with a 2441 traffic enforcement LFB. The declaration creates the 2442 constructs for this. This construct tells the CE and FE that any 2443 manipulation of the defined data is actually manipulation of data 2444 defined to exist in some specified part of some other LFB instance. 2445 The content of an element MUST be a named type. Whatever 2446 component the alias references (which is determined by the alias 2447 component properties, as described below) that component must be of 2448 the same type as that declared for the alias. Thus, when the CE or 2449 FE dereferences the alias component, the type of the information 2450 returned is known. The type can be a base type or a derived type. 2451 The actual value referenced by an alias is known as its target. When 2452 a GET or SET operation references the alias element, the value of the 2453 target is returned or replaced. Write access to an alias element is 2454 permitted if write access to both the alias and the target are 2455 permitted. 2457 The target of a component declared by an element is 2458 determined by the information in the component's properties. Like 2459 all components, the properties include the support / read / write 2460 permission for the alias. In addition, there are several fields 2461 (components) in the alias properties which define the target of the 2462 alias. These components are the ID of the LFB class of the target, 2463 the ID of the LFB instance of the target, and a sequence of integers 2464 representing the path within the target LFB instance to the target 2465 component. The type of the target element must match the declared 2466 type of the alias. Details of the alias property structure are 2467 described in Section 4.8 of this document on properties. 2469 Note that the read / write property of the alias refers to the value. 2470 The CE can only determine if it can write the target selection 2471 properties of the alias by attempting such a write operation. 2472 (Property components do not themselves have properties.) 2474 4.5.7. Augmentations 2476 Compound types can also be defined as augmentations of existing 2477 compound types. If the existing compound type is a structure, 2478 augmentation MAY add new elements to the type. The type of an 2479 existing component MAY be replaced in the definition of an augmenting 2480 structure, but MAY only be replaced with an augmentation derived from 2481 the current type of the existing component. An existing component 2482 cannot be deleted. If the existing compound type is an array, 2483 augmentation means augmentation of the array element type. 2485 Augmentation MUST NOT be applied to unions. 2487 One consequence of this is that augmentations are backwards 2488 compatible with the compound type from which they are derived. As 2489 such, augmentations are useful in defining components for LFB 2490 subclasses with backward compatibility. In addition to adding new 2491 components to a class, the data type of an existing component MAY be 2492 replaced by an augmentation of that component, and still meet the 2493 compatibility rules for subclasses. This compatibility constraint is 2494 why augmentations can not be applied to unions. 2496 For example, consider a simple base LFB class A that has only one 2497 component (comp1) of type X. One way to derive class A1 from A can be 2498 by simply adding a second component (of any type). Another way to 2499 derive a class A2 from A can be by replacing the original component 2500 (comp1) in A of type X with one of type Y, where Y is an augmentation 2501 of X. Both classes A1 and A2 are backward compatible with class A. 2503 The syntax for augmentations is to include a element in 2504 a structure definition, indicating what structure type is being 2505 augmented. Component names and component IDs for new components 2506 within the augmentation MUST NOT be the same as those in the 2507 structure type being augmented. For those components where the data 2508 type of an existing component is being replaced with a suitable 2509 augmenting data type, the existing Component name and component ID 2510 MUST be used in the augmentation. Other than the constraint on 2511 existing elements, there is no requirement that the new component IDs 2512 be sequential with, greater than, or in any other specific 2513 relationship to the existing component IDs except different. It is 2514 expected that using values sequential within an augmentation, and 2515 distinct from the previously used values, will be a common method to 2516 enhance human readability. 2518 4.6. Element for Metadata Definitions 2520 The (optional) element in the library document 2521 contains one or more elements. Each 2522 element defines a metadatum. 2524 Each element MUST contain a unique name (NMTOKEN). 2525 Uniqueness is defined to be over all metadata defined in this library 2526 document and in all directly or indirectly included library 2527 documents. The element MUST also contain a brief 2528 synopsis, the tag value to be used for this metadata, and value type 2529 definition information. Only atomic data types can be used as value 2530 types for metadata. The element MAY contain a detailed 2531 description element. 2533 Two forms of type definitions are allowed. The first form uses the 2534 element to refer to an existing atomic data type defined in 2535 the element of the same library document or in one of 2536 the included library documents. The usage of the element 2537 is identical to how it is used in the elements, except 2538 here it can only refer to atomic types. The latter restriction is 2539 not enforced by the XML schema. 2541 The second form is an explicit type definition using the 2542 element. This element is used here in the same way as in the 2543 elements. 2545 The following example shows both usages: 2547 2548 2549 NEXTHOPID 2550 Refers to a Next Hop entry in NH LFB 2551 17 2552 int32 2553 2554 2555 CLASSID 2556 2557 Result of classification (0 means no match). 2558 2559 21 2560 2561 int32 2562 2563 2564 NOMATCH 2565 2566 Classification didn't result in match. 2567 2568 2569 2570 2571 2572 2574 4.7. Element for LFB Class Definitions 2576 The (optional) element can be used to define one or 2577 more LFB classes using elements. Each 2578 element MUST define an LFB class and include the following elements: 2580 o provides the symbolic name of the LFB class. Example: 2581 "ipv4lpm" 2583 o provides a short synopsis of the LFB class. Example: 2584 "IPv4 Longest Prefix Match Lookup LFB" 2586 o is the version indicator 2587 o is the inheritance indicator 2589 o lists the input ports and their specifications 2591 o lists the output ports and their specifications 2593 o defines the operational components of the LFB 2595 o defines the capability components of the LFB 2597 o contains the operational specification of the LFB 2599 o The LFBClassID attribute of the LFBClassDef element defines the ID 2600 for this class. These must be globally unique. 2602 o defines the events that can be generated by instances of 2603 this LFB. 2605 LFB Class Names must be unique, in order to enable other documents to 2606 reference the classes by name, and to enable human readers to 2607 understand references to class names. While a complex naming 2608 structure could be created, simplicity is preferred. As given in the 2609 IANA considerations section of this document, the IANA will maintain 2610 a registry of LFB Class names and Class identifiers, along with a 2611 reference to the document defining the class. 2613 Below is a skeleton of an example LFB class definition. Note that in 2614 order to keep from complicating the XML Schema, the order of elements 2615 in the class definition is fixed. Elements, if they appear, must 2616 appear in the order shown. 2618 2619 2620 ipv4lpm 2621 IPv4 Longest Prefix Match Lookup LFB 2622 1.0 2623 baseclass 2625 2626 ... 2627 2629 2630 ... 2631 2633 2634 ... 2635 2637 2638 ... 2639 2641 2642 ... 2643 2645 2646 This LFB represents the IPv4 longest prefix match lookup 2647 operation. 2648 The modeled behavior is as follows: 2649 Blah-blah-blah. 2650 2652 2653 ... 2654 2656 The individual components and capabilities will have componentIDs for 2657 use by the ForCES protocol. These parallel the componentIDs used in 2658 structs, and are used the same way. Component and capability 2659 componentIDs must be unique within the LFB class definition. 2661 Note that the , , and elements are 2662 required, all other elements are optional in . However, 2663 when they are present, they must occur in the above order. 2665 The componentID attribute for different items in an LFB class 2666 definition (or components in a struct) MUST be distinct. They do not 2667 need to be in order, nor do they need to be sequential. For clarity 2668 of human readability, and ease of maintanence, it is usual to define 2669 at least sequential sets of values. But this is for human ease, not 2670 a model or protocol requirement. 2672 4.7.1. Element to Express LFB Inheritance 2674 The optional element can be used to indicate that this 2675 class is a derivative of some other class. The content of this 2676 element MUST be the unique name () of another LFB class. The 2677 referred LFB class MUST be defined in the same library document or in 2678 one of the included library documents. In the absence of a 2679 the class is conceptually derived from the common, 2680 empty, base class. 2682 It is assumed that a derived class is backwards compatible with its 2683 base class. A derived class MAY add compoents to a parent class, but 2684 can not delete components. This also applies to input and output 2685 ports, events, and to capabilities. 2687 4.7.2. Element to Define LFB Inputs 2689 The optional element is used to define input ports. An 2690 LFB class MAY have zero, one, or more inputs. If the LFB class has 2691 no input ports, the element MUST be omitted. The 2692 element can contain one or more elements, 2693 one for each port or port-group. We assume that most LFBs will have 2694 exactly one input. Multiple inputs with the same input type are 2695 modeled as one input group. Input groups are defined the same way as 2696 input ports by the element, differentiated only by an 2697 optional "group" attribute. 2699 Multiple inputs with different input types should be avoided if 2700 possible (see discussion in Section 4.7.3). Some special LFBs will 2701 have no inputs at all. For example, a packet generator LFB does not 2702 need an input. 2704 Single input ports and input port groups are both defined by the 2705 element; they are differentiated by only an optional 2706 "group" attribute. 2708 The element MUST contain the following elements: 2710 o provides the symbolic name of the input. Example: "in". 2711 Note that this symbolic name must be unique only within the scope 2712 of the LFB class. 2714 o contains a brief description of the input. Example: 2715 "Normal packet input". 2717 o lists all allowed frame formats. Example: {"ipv4" 2718 and "ipv6"}. Note that this list should refer to names specified 2719 in the element of the same library document or in any 2720 included library documents. The element can also 2721 provide a list of required metadata. Example: {"classid", 2722 "vpnid"}. This list should refer to names of metadata defined in 2723 the element in the same library document or in any 2724 included library documents. For each metadatum, it must be 2725 specified whether the metadatum is required or optional. For each 2726 optional metadatum, a default value must be specified, which is 2727 used by the LFB if the metadatum is not provided with a packet. 2729 In addition, the optional "group" attribute of the 2730 element can specify if the port can behave as a port group, i.e., it 2731 is allowed to be instantiated. This is indicated by a "true" value 2732 (the default value is "false"). 2734 An example element, defining two input ports, the second 2735 one being an input port group: 2737 2738 2739 in 2740 Normal input 2741 2742 2743 ipv4 2744 ipv6 2745 2746 2747 classid 2748 vifid 2749 vrfid 2750 2751 2752 2753 2754 ... another input port ... 2755 2756 2758 For each , the frame type expectations are defined by the 2759 element using one or more elements (see example 2760 above). When multiple frame types are listed, it means that "one of 2761 these" frame types is expected. A packet of any other frame type is 2762 regarded as incompatible with this input port of the LFB class. The 2763 above example list two frames as expected frame types: "ipv4" and 2764 "ipv6". 2766 Metadata expectations are specified by the 2767 element. In its simplest form, this element can contain a list of 2768 elements, each referring to a metadatum. When multiple 2769 instances of metadata are listed by elements, it means that 2770 "all of these" metadata must be received with each packet (except 2771 metadata that are marked as "optional" by the "dependency" attribute 2772 of the corresponding element). For a metadatum that is 2773 specified "optional", a default value MUST be provided using the 2774 "defaultValue" attribute. The above example lists three metadata as 2775 expected metadata, two of which are mandatory ("classid" and 2776 "vifid"), and one being optional ("vrfid"). 2778 The schema also allows for more complex definitions of metadata 2779 expectations. For example, using the element, a list of 2780 metadata can be specified to express that at least one of the 2781 specified metadata must be present with any packet. For example: 2783 2784 2785 prefixmask 2786 prefixlen 2787 2788 2790 The above example specifies that either the "prefixmask" or the 2791 "prefixlen" metadata must be provided with any packet. 2793 The two forms can also be combined, as it is shown in the following 2794 example: 2796 2797 classid 2798 vifid 2799 vrfid 2800 2801 prefixmask 2802 prefixlen 2803 2804 2806 Although the schema is constructed to allow even more complex 2807 definitions of metadata expectations, we do not discuss those here. 2809 4.7.3. Element to Define LFB Outputs 2811 The optional element is used to define output ports. 2812 An LFB class MAY have zero, one, or more outputs. If the LFB class 2813 has no output ports, the element MUST be omitted. The 2814 element MUST contain one or more elements, 2815 one for each port or port-group. If there are multiple outputs with 2816 the same output type, we model them as an output port group. Some 2817 special LFBs have no outputs at all (e.g., Dropper). 2819 Single output ports and output port groups are both defined by the 2820 element; they are differentiated by only an optional 2821 "group" attribute. 2823 The element MUST contain the following elements: 2825 o provides the symbolic name of the output. Example: "out". 2826 Note that the symbolic name must be unique only within the scope 2827 of the LFB class. 2829 o contains a brief description of the output port. 2830 Example: "Normal packet output". 2832 o lists the allowed frame formats. Example: {"ipv4", 2833 "ipv6"}. Note that this list should refer to symbols specified in 2834 the element in the same library document or in any 2835 included library documents. The element MAY also 2836 contain the list of emitted (generated) metadata. Example: 2837 {"classid", "color"}. This list should refer to names of metadata 2838 specified in the element in the same library 2839 document or in any included library documents. For each generated 2840 metadatum, it should be specified whether the metadatum is always 2841 generated or generated only in certain conditions. This 2842 information is important when assessing compatibility between 2843 LFBs. 2845 In addition, the optional "group" attribute of the 2846 element can specify if the port can behave as a port group, i.e., it 2847 is allowed to be instantiated. This is indicated by a "true" value 2848 (the default value is "false"). 2850 The following example specifies two output ports, the second being an 2851 output port group: 2853 2854 2855 out 2856 Normal output 2857 2858 2859 ipv4 2860 ipv4bis 2861 2862 2863 nhid 2864 nhtabid 2865 2866 2867 2868 2869 exc 2870 Exception output port group 2871 2872 2873 ipv4 2874 ipv4bis 2875 2876 2877 errorid 2878 2879 2880 2881 2883 The types of frames and metadata the port produces are defined inside 2884 the element in each . Within the 2885 element, the list of frame types the port produces is listed in the 2886 element. When more than one frame is listed, it 2887 means that "one of" these frames will be produced. 2889 The list of metadata that is produced with each packet is listed in 2890 the optional element of the . In its 2891 simplest form, this element can contain a list of elements, 2892 each referring to a metadatum type. The meaning of such a list is 2893 that "all of" these metadata are provided with each packet, except 2894 those that are listed with the optional "availability" attribute set 2895 to "conditional". Similar to the element of the 2896 , the element supports more complex 2897 forms, which we do not discuss here further. 2899 4.7.4. Element to Define LFB Operational Components 2901 Operational parameters of the LFBs that must be visible to the CEs 2902 are conceptualized in the model as the LFB components. These 2903 include, for example, flags, single parameter arguments, complex 2904 arguments, and tables. Note that the components here refer to only 2905 those operational parameters of the LFBs that must be visible to the 2906 CEs. Other variables that are internal to LFB implementation are not 2907 regarded as LFB components and hence are not covered. 2909 Some examples for LFB components are: 2911 o Configurable flags and switches selecting between operational 2912 modes of the LFB 2914 o Number of inputs or outputs in a port group 2916 o Various configurable lookup tables, including interface tables, 2917 prefix tables, classification tables, DSCP mapping tables, MAC 2918 address tables, etc. 2920 o Packet and byte counters 2922 o Various event counters 2924 o Number of current inputs or outputs for each input or output group 2926 The ForCES model supports the definition of access permission 2927 restrictions on what the CE can do with an LFB component. The 2928 following categories are supported by the model: 2930 o No-access components. This is useful for completeness, and to 2931 allow for defining objects which are used by other things, but not 2932 directly referencable by the CE. It is also useful for an FE 2933 which is reporting that certain defined, and typically accessible, 2934 Components are not supported for CE access by a reporting FE. 2936 o Read-only components. 2938 o Read-write components. 2940 o Write-only components. This could be any configurable data for 2941 which read capability is not provided to the CEs. (e.g., the 2942 security key information) 2944 o Read-reset components. The CE can read and reset this resource, 2945 but cannot set it to an arbitrary value. Example: Counters. 2947 o Firing-only components. A write attempt to this resource will 2948 trigger some specific actions in the LFB, but the actual value 2949 written is ignored. 2951 The LFB class MUST define only one possible access mode for a given 2952 component. 2954 The components of the LFB class are listed in the 2955 element. Each component is defined by an element. A 2956 element contains some or all of the following elements, 2957 some of which are mandatory: 2959 o MUST occur, and defines the name of the component. This 2960 name must be unique among the components of the LFB class. 2961 Example: "version". 2963 o is also mandatory, and provides a brief description of 2964 the purpose of the component. 2966 o is an optional element, and if present indicates that 2967 this component is optional. 2969 o The data type of the component can be defined either via a 2970 reference to a predefined data type or providing a local 2971 definition of the type. The former is provided by using the 2972 element, which must refer to the unique name of an 2973 existing data type defined in the element in the 2974 same library document or in any of the included library documents. 2975 When the data type is defined locally (unnamed type), one of the 2976 following elements can be used: , , , and 2977 . Their usage is identical to how they are used inside 2978 elements (see Section 4.5). Some form of data type 2979 definition MUST be included in the component definition. 2981 o The element is optional, and if present is used to 2982 specify a default value for a component. If a default value is 2983 specified, the FE must ensure that the component has that value 2984 when the LFB is initialized or reset. If a default value is not 2985 specified for a component, the CE MUST make no assumptions as to 2986 what the value of the component will be upon initalization. The 2987 CE must either read the value, or set the value, if it needs to 2988 know what it is. 2990 o The element MAY also appear. If included, it 2991 provides a longer description of the meaning or usage of the 2992 particular component being defined. 2994 The element also MUST have an componentID attribute, 2995 which is a numeric value used by the ForCES protocol. 2997 In addition to the above elements, the element includes 2998 an optional "access" attribute, which can take any of the following 2999 values: "read-only", "read-write", "write-only", "read-reset", and 3000 "trigger-only". The default access mode is "read-write". 3002 Whether optional components are supported, and whether components 3003 defined as read-write can actually be written can be determined for a 3004 given LFB instance by the CE by reading the property information of 3005 that component. An access control setting of "trigger-only" means 3006 that this component is included only for use in event detection. 3008 The following example defines two components for an LFB: 3010 3011 3012 foo 3013 number of things 3014 uint32 3015 3016 3017 bar 3018 number of this other thing 3019 3020 uint32 3021 3022 3023 3024 3025 10 3026 3027 3029 The first component ("foo") is a read-only 32-bit unsigned integer, 3030 defined by referring to the built-in "uint32" atomic type. The 3031 second component ("bar") is also an integer, but uses the 3032 element to provide additional range restrictions. This component has 3033 access mode of read-write allowing it to be both read and written. A 3034 default value of 10 is provided for bar. although the access for bar 3035 is read-write, some implementations MAY offer only more restrictive 3036 access, and this would be reported in the component properties. 3038 Note that not all components are likely to exist at all times in a 3039 particular implementation. While the capabilities will frequently 3040 indicate this non-existence, CEs may attempt to reference non- 3041 existent or non-permitted components anyway. The ForCES protocol 3042 mechanisms should include appropriate error indicators for this case. 3044 The mechanism defined above for non-supported components can also 3045 apply to attempts to reference non-existent array elements or to set 3046 read-only components. 3048 4.7.5. Element to Define LFB Capability Components 3050 The LFB class specification provides some flexibility for the FE 3051 implementation regarding how the LFB class is implemented. For 3052 example, the instance may have some limitations that are not inherent 3053 from the class definition, but rather the result of some 3054 implementation limitations. Some of these limitations are captured 3055 by the property information of the LFB components. The model allows 3056 for the notion of additional capability information. 3058 Such capability related information is expressed by the capability 3059 components of the LFB class. The capability components are always 3060 read-only attributes, and they are listed in a separate 3061 element in the . The 3062 element contains one or more elements, each defining one 3063 capability component. The format of the element is 3064 almost the same as the element, it differs in two 3065 aspects: it lacks the access mode attribute (because it is always 3066 read-only), and it lacks the element (because default 3067 value is not applicable to read-only attributes). 3069 Some examples of capability components follow: 3071 o The version of the LFB class that this LFB instance complies with; 3073 o Supported optional features of the LFB class; 3075 o Maximum number of configurable outputs for an output group; 3077 o Metadata pass-through limitations of the LFB; 3079 o Additional range restriction on operational components; 3081 The following example lists two capability attributes: 3083 3084 3085 version 3086 3087 LFB class version this instance is compliant with. 3088 3089 version 3090 3091 3092 limitBar 3093 3094 Maximum value of the "bar" attribute. 3095 3096 uint16 3097 3098 3100 4.7.6. Element for LFB Notification Generation 3102 The element contains the information about the occurrences 3103 for which instances of this LFB class can generate notifications to 3104 the CE. High level view on the declaration and operation of LFB 3105 events is described in Section 3.2.5. 3107 The element contains 0 or more elements, each of 3108 which declares a single event. The element has an eventID 3109 attribute giving the unique (per LFB class) ID of the event. The 3110 element will include: 3112 o element indicating which LFB field (component) is 3113 tested to generate the event; 3115 o element indicating what condition on the field will 3116 generate the event from a list of defined conditions; 3118 o element indicating what values are to be reported 3119 in the notification of the event. 3121 The example below demonstrates the different constructs. 3123 The element has a baseID attribute value, which is normally 3124 . The value of the baseID is the starting 3125 componentID for the path which identifies events. It must not be the 3126 same as the componentID of any top level components (including 3127 capabilities) of the LFB class. In derived LFBs (i.e. ones with a 3128 element) where the parent LFB class has an events 3129 declaration, the baseID must not be present in the derived LFB 3130 element. Instead, the baseID value from the parent LFB 3131 class is used. In the example shown the baseID is 7. 3133 3134 3135 Foochanged 3136 3137 An example event for a scalar 3138 3139 3140 foo 3141 3142 3143 3144 3145 3146 foo 3147 3148 3149 3151 3152 Goof1changed 3153 3154 An example event for a complex structure 3155 3156 3157 3158 goo 3159 f1 3160 3161 3162 3163 3164 3165 goo 3166 f1 3167 3168 3169 3171 3172 NewbarEntry 3173 3174 Event for a new entry created on table bar 3175 3176 3177 bar 3178 _barIndex_ 3179 3180 3181 3182 3183 bar 3184 _barIndex_ 3185 3186 3187 foo 3188 3189 3190 3192 3193 Gah11changed 3194 3195 Event for table gah, entry index 11 changing 3196 3197 3198 gah 3199 11 3200 3201 3202 3203 3204 gah 3205 11 3206 3207 3208 3210 3211 Gah10field1 3212 3213 Event for table gah, entry index 10, column field1 changing 3214 3215 3216 gah 3217 10 3218 field1 3219 3220 3221 3222 3223 gah 3224 10 3225 3227 3228 3229 3231 4.7.6.1. Element 3233 The element contains information identifying a field in 3234 the LFB that is to be monitored for events. 3236 The element contains one or more each of 3237 which MAY be followed by one or more elements. Each 3238 of these two elements represent the textual equivalent of a path 3239 select component of the LFB. 3241 The element contains the name of a component in the LFB 3242 or a component nested in an array or structure within the LFB. The 3243 name used in MUST identify a valid component within the 3244 containing LFB context. The first element in a MUST be 3245 an element. In the example shown, four LFB components 3246 foo, goo, bar and gah are used as s. 3248 In the simple case, an identifies an atomic component. 3249 This is the case illustrated in the event named Foochanged. 3250 is also used to address complex components such as 3251 arrays or structures. 3253 The first defined event, Foochanged, demonstrates how a scalar LFB 3254 component, foo, could be monitored to trigger an event. 3256 The second event, Goof1changed, demonstrates how a member of the 3257 complex structure goo could be monitored to trigger an event. 3259 The events named NewbarEntry, Gah11changed and Gah10field1 3260 represent monitoring of arrays bar and gah in differing details. 3262 If an identifies a complex component then a further 3263 MAY be used to refine the path to the target element. 3264 Defined event Goof1changed demonstrates how a second is 3265 used to point to member f1 of the structure goo. 3267 If an identifies an array then the following rules 3268 apply: 3270 o elements MUST be present as the next XML element 3271 after an which identifies an array component. 3272 MUST NOT occur other than after an array 3273 reference, as it is only meaningful in that context. 3275 o An contains either: 3277 * A numeric value to indicate that the event applies to a 3278 specific entry (by index) of the array. As an example, event 3279 Gah11changed shows how table gah's index 11 is being targeted 3280 for monitoring. 3282 * It is expected that the more common usage is to have the event 3283 being defined across all elements of the array (i.e a wildcard 3284 for all indices). In that case, the value of the 3285 MUST be a name rather than a numeric value. 3286 That same name can then be used as the value of 3287 in elements as described below. 3288 An example of a wild card table index is shown in event 3289 NewBarentry where the value is named 3290 _barIndex_ 3292 o An MAY follow an to further refine 3293 the path to the target element (Note: this is in the same spirit 3294 as the case where is used to further refine 3295 in the earlier example of a complex structure example 3296 of Goof1changed). The example event Gah10field1 illustrates how 3297 the column field1 of table gah is monitored for changes. 3299 It should be emphasized that the name in an element 3300 in defined event NewbarEntry is not a component name. It is a 3301 variable name for use in the elements (described in 3302 Section 4.7.6.3) of the given LFB definition. This name MUST be 3303 distinct from any component name that can validly occur in the 3304 clause. 3306 4.7.6.2. Element 3308 The event condition element represents a condition that triggers a 3309 notification. The list of conditions is: 3311 o the target must be an array, ending with a 3312 subscript indication. The event is generated when an entry in the 3313 array is created. This occurs even if the entry is created by CE 3314 direction. The event example NewbarEntry demonstrates the 3315 condition. 3317 o the target must be an array, ending with a 3318 subscript indication. The event is generated when an entry in the 3319 array is destroyed. This occurs even if the entry is destroyed by 3320 CE direction. 3322 o the event is generated whenever the target 3323 component changes in any way. For binary components such as up/ 3324 down, this reflects a change in state. It can also be used with 3325 numeric attributes, in which case any change in value results in a 3326 detected trigger. Event examples Foochanged, Gah11changed, and 3327 Gah10field1 illustrate the condition. 3329 o the event is generated whenever the target 3330 component becomes greater than the threshold. The threshold is an 3331 event property. 3333 o the event is generated whenever the target 3334 component becomes less than the threshold. The threshold is an 3335 event property. 3337 4.7.6.3. Element 3339 The element of an declare the information to 3340 be delivered by the FE along with the notification of the occurrence 3341 of the event. 3343 The element contains one or more 3344 elements. Each element identifies a piece of data from 3345 the LFB class to be reported. The notification carries that data as 3346 if the collection of elements had been defined in a 3347 structure. The syntax is exactly the same as used in the 3348 element, using and 3349 elements and so the same rules apply. Each element 3350 thus MUST identify a component in the LFB class. MAY 3351 contain integers. If they contain names, they MUST be names from 3352 elements of the in the event. The 3353 selection for the report will use the value for the subscript that 3354 identifies that specific element triggering the event. This can be 3355 used to reference the component causing the event, or to reference 3356 related information in parallel tables. 3358 In the example shown, in the case of the event Foochanged, the report 3359 will carry the value of foo; in the case of the defined event 3360 NewbarEntry acting on LFB component bar, which is an array, there are 3361 two items that are reported as indicated by the two 3362 declarations: 3364 o The first details what new entry was added in the 3365 table bar. Recall that _barIndex_ is declared as the event's 3366 and that by virtue of using a name 3367 instead of a numeric value, the is implied to be a 3368 wildcard and will carry whatever index of the new entry. 3370 o The second includes the value of LFB component foo 3371 at the time the new entry was created in bar. Reporting foo in 3372 this case is provided to demonstrate the flexibility of event 3373 reporting. 3375 This event reporting structure is designed to allow the LFB designer 3376 to specify information that is likely not known a priori by the CE 3377 and is likely needed by the CE to process the event. While the 3378 structure allows for pointing at large blocks of information (full 3379 arrays or complex structures) this is not recommended. Also, the 3380 variable reference/subscripting in reporting only captures a small 3381 portion of the kinds of related information. Chaining through index 3382 fields stored in a table, for example, is not supported. In general, 3383 the mechanism is an optimization for cases that have 3384 been found to be common, saving the CE from having to query for 3385 information it needs to understand the event. It does not represent 3386 all possible information needs. 3388 If any components referenced by the eventReport are optional, then 3389 the report MUST use a protocol format that supports optional elements 3390 and allows for the non-existence of such elements. Any components 3391 which do not exist are not reported. 3393 4.7.6.4. Runtime control of events 3395 The high level view of the declaration and operation of LFB events is 3396 described in Section 3.2.5. 3398 The provides additional components used in the path to 3399 reference the event. The path constitutes the baseID for events, 3400 followed by the ID for the specific event, followed by a value for 3401 each element if it exists in the . 3403 The event path will uniquely identify a specific occurrence of the 3404 event in the event notification to the CE. In the example provided 3405 above, at the end of Section 4.7.6, a notification with path of 7.7 3406 uniquely identifies the event to be that caused by the change of foo; 3407 an event with path 7.9.100 uniquely identifies the event to be that 3408 caused by a creation of table bar entry with index/subscript 100. 3410 As described in the Section 4.8.5, event elements have properties 3411 associated with them. These properties include the subscription 3412 information indicating whether the CE wishes the FE to generate event 3413 reports for the event at all, thresholds for events related to level 3414 crossing, and filtering conditions that may reduce the set of event 3415 notifications generated by the FE. Details of the filtering 3416 conditions that can be applied are given in that section. The 3417 filtering conditions allow the FE to suppress floods of events that 3418 could result from oscillation around a condition value. For FEs that 3419 do not wish to support filtering, the filter properties can either be 3420 read only or not supported. 3422 In addition to identifying the event sources, the CE also uses the 3423 event path to activate runtime control of the event via the event 3424 properties (defined in Section 4.8.5) utilizing SET-PROP as defined 3425 in ForCES Protocol [2] operation. 3427 To activate event generation on the FE, a SET-PROP message 3428 referencing the event and registration property of the event is 3429 issued to the FE by the CE with any prefix of the path of the event. 3430 So, for an event defined on the example table bar, a SET-PROP with a 3431 path of 7.9 will subscribe the CE to all occurrences of that event on 3432 any entry of the table. This is particularly useful for the 3433 and conditions on tables. Events 3434 using those conditions will generally be defined with a field/ 3435 subscript sequence that identifies an array and ends with an 3436 element. Thus, the event notification will indicate 3437 which array entry has been created or destroyed. A typical 3438 subscriber will subscribe for the array, as opposed to a specific 3439 entry in an array, so it will use a shorter path. 3441 In the example provided, subscribing to 7.8 implies receiving all 3442 declared events from table bar. Subscribing to 7.8.100 implies 3443 receiving an event when subscript/index 100 table entry is created. 3445 Threshold and filtering conditions can only be applied to individual 3446 events. For events defined on elements of an array, this 3447 specification does not allow for defining a threshold or filtering 3448 condition on an event for all elements of an array. 3450 4.7.7. Element for LFB Operational Specification 3452 The element of the provides unstructured 3453 text (in XML sense) to explain what the LFB does to a human user. 3455 4.8. Properties 3457 Components of LFBs have properties which are important to the CE. 3458 The most important property is the existence / readability / 3459 writeability of the element. Depending on the type of the component, 3460 other information may be of importance. 3462 The model provides the definition of the structure of property 3463 information. There is a base class of property information. For the 3464 array, alias, and event components there are subclasses of property 3465 information providing additional fields. This information is 3466 accessed by the CE (and updated where applicable) via the ForCES 3467 protocol. While some property information is writeable, there is no 3468 mechanism currently provided for checking the properties of a 3469 property element. Writeability can only be checked by attempting to 3470 modify the value. 3472 4.8.1. Basic Properties 3474 The basic property definition, along with the scalar dataTypeDef for 3475 accessibility is below. Note that this access permission information 3476 is generally read-only. 3478 3479 accessPermissionValues 3480 3481 The possible values of component access permission 3482 3483 3484 uchar 3485 3486 3487 None 3488 Access is prohibited 3489 3490 3491 Read-Only 3492 3493 Access to the component is read only 3494 3495 3496 3497 Write-Only 3498 3499 The component MAY be written, but not read 3500 3501 3502 3503 Read-Write 3504 3505 The component MAY be read or written 3506 3507 3508 3509 3510 3511 3512 baseElementProperties 3513 basic properties, accessibility 3514 3515 3516 accessibility 3517 3518 does the component exist, and 3519 can it be read or written 3520 3521 accessPermissionValues 3522 3523 3524 3526 4.8.2. Array Properties 3528 The properties for an array add a number of important pieces of 3529 information. These properties are also read-only. 3531 3532 arrayElementProperties 3533 Array Element Properties definition 3534 3535 baseElementProperties 3536 3537 entryCount 3538 the number of entries in the array 3539 uint32 3540 3541 3542 highestUsedSubscript 3543 the last used subscript in the array 3544 uint32 3545 3546 3547 firstUnusedSubscript 3548 3549 The subscript of the first unused array element 3550 3551 uint32 3552 3553 3554 3556 4.8.3. String Properties 3558 The properties of a string specify the actual octet length and the 3559 maximum octet length for the element. The maximum length is included 3560 because an FE implementation MAY limit a string to be shorter than 3561 the limit in the LFB Class definition. 3563 3564 stringElementProperties 3565 string Element Properties definition 3566 3567 baseElementProperties 3568 3569 stringLength 3570 the number of octets in the string 3571 uint32 3572 3573 3574 maxStringLength 3575 3576 the maximum number of octets in the string 3577 3578 uint32 3579 3580 3581 3583 4.8.4. Octetstring Properties 3585 The properties of an octetstring specify the actual length and the 3586 maximum length, since the FE implementation MAY limit an octetstring 3587 to be shorter than the LFB Class definition. 3589 3590 octetstringElementProperties 3591 octetstring Element Properties definition 3592 3593 3594 baseElementProperties 3595 3596 octetstringLength 3597 3598 the number of octets in the octetstring 3599 3600 uint32 3601 3602 3603 maxOctetstringLength 3604 3605 the maximum number of octets in the octetstring 3606 3607 uint32 3608 3609 3610 3612 4.8.5. Event Properties 3614 The properties for an event add three (usually) writeable fields. 3615 One is the subscription field. 0 means no notification is generated. 3616 Any non-zero value (typically 1 is used) means that a notification is 3617 generated. The hysteresis field is used to suppress generation of 3618 notifications for oscillations around a condition value, and is 3619 described below (Section 4.8.5.2). The threshold field is used for 3620 the and conditions. It 3621 indicates the value to compare the event target against. Using the 3622 properties allows the CE to set the level of interest. FEs which do 3623 not support setting the threshold for events will make this field 3624 read-only. 3626 3627 eventElementProperties 3628 event Element Properties definition 3629 3630 baseElementProperties 3631 3632 registration 3633 3634 has the CE registered to be notified of this event 3635 3636 uint32 3637 3638 3639 threshold 3640 comparison value for level crossing events 3641 3642 3643 uint32 3644 3645 3646 eventHysteresis 3647 region to suppress event recurrence notices 3648 3649 3650 uint32 3651 3652 3653 eventCount 3654 number of occurrences to suppress 3655 3656 3657 uint32 3658 3659 3660 eventInterval 3661 time interval in ms between notifications 3662 3663 3664 uint32 3665 3666 3667 3669 4.8.5.1. Common Event Filtering 3671 The event properties have values for controlling several filter 3672 conditions. Support of these conditions is optional, but all 3673 conditions SHOULD be supported. Events which are reliably known not 3674 to be subject to rapid occurrence or other concerns MAY not support 3675 all filter conditions. 3677 Currently, three different filter condition variables are defined. 3678 These are eventCount, eventInterval, and eventHysteresis. Setting 3679 the condition variables to 0 (their default value) means that the 3680 condition is not checked. 3682 Conceptually, when an event is triggered, all configured conditions 3683 are checked. If no filter conditions are triggered, or if any 3684 trigger conditions are met, the event notification is generated. If 3685 there are filter conditions, and no condition is met, then no event 3686 notification is generated. Event filter conditions have reset 3687 behavior when an event notification is generated. If any condition 3688 is passed, and the notification is generated, the notification reset 3689 behavior is performed on all conditions, even those which had not 3690 passed. This provides a clean definition of the interaction of the 3691 various event conditions. 3693 An example of the interaction of conditions is an event with an 3694 eventCount property set to 5 and an eventInterval property set to 500 3695 milliseconds. Suppose that a burst of occurrences of this event is 3696 detected by the FE. The first occurrence will cause a notification 3697 to be sent to the CE. Then, if four more occurrences are detected 3698 rapidly (less than 0.5 seconds) they will not result in 3699 notifications. If two more occurrences are detected, then the second 3700 of those will result in a notification. Alternatively, if more than 3701 500 milliseconds has passed since the notification and an occurrence 3702 is detected, that will result in a notification. In either case, the 3703 count and time interval suppression is reset no matter which 3704 condition actually caused the notification. 3706 4.8.5.2. Event Hysteresis Filtering 3708 Events with numeric conditions can have hysteresis filters applied to 3709 them. The hysteresis level is defined by a property of the event. 3710 This allows the FE to notify the CE of the hysteresis applied, and if 3711 it chooses, the FE can allow the CE to modify the hysteresis. This 3712 applies to for a numeric field, and to 3713 and . The content of a 3714 element is a numeric value. When supporting hysteresis, 3715 the FE MUST track the value of the element and make sure that the 3716 condition has become untrue by at least the hysteresis from the event 3717 property. To be specific, if the hysteresis is V, then 3719 o For a condition, if the last notification was for 3720 value X, then the notification MUST NOT be generated 3721 until the value reaches X +/- V. 3723 o For a condition with threshold T, once the 3724 event has been generated at least once it MUST NOT be generated 3725 again until the field first becomes less than or equal to T - V, 3726 and then exceeds T. 3728 o For a condition with threshold T, once the event 3729 has been generate at least once it MUST NOT be generated again 3730 until the field first becomes greater than or equal to T + V, and 3731 then becomes less than T. 3733 4.8.5.3. Event Count Filtering 3735 Events MAY have a count filtering condition. This property, if set 3736 to a non-zero value, indicates the number of occurrences of the event 3737 that should be considered redundant and not result in a notification. 3738 Thus, if this property is set to 1, and no other conditions apply, 3739 then every other detected occurrence of the event will result in a 3740 notification. This particular meaning is chosen so that the value 1 3741 has a distinct meaning from the value 0. 3743 A conceptual implementation (not required) for this might be an 3744 internal suppression counter. Whenever an event is triggered, the 3745 counter is checked. If the counter is 0, a notification is 3746 generated. Whether a notification is generated or not, the counter 3747 is incremented. If the counter exceeds the configured value, it is 3748 set to 0. 3750 4.8.5.4. Event Time Filtering 3752 Events MAY have a time filtering condition. This property represents 3753 the minimum time interval (in the absence of some other filtering 3754 condition being passed) between generating notifications of detected 3755 events. This condition MUST only be passed if the time since the 3756 last notification of the event is longer than the configured interval 3757 in milliseconds. 3759 Conceptually, this can be thought of as a stored timestamp which is 3760 compared with the detection time, or as a timer that is running that 3761 resets a suppression flag. In either case, if a notification is 3762 generated due to passing any condition then the time interval 3763 detection MUST be restarted. 3765 4.8.6. Alias Properties 3767 The properties for an alias add three (usually) writeable fields. 3768 These combine to identify the target component the subject alias 3769 refers to. 3771 3772 aliasElementProperties 3773 alias Element Properties defintion 3774 3775 baseElementProperties 3776 3777 targetLFBClass 3778 the class ID of the alias target 3779 uint32 3780 3781 3782 targetLFBInstance 3783 the instance ID of the alias target 3784 uint32 3785 3786 3787 targetComponentPath 3788 3789 the path to the component target 3790 each 4 octets is read as one path element, 3791 using the path construction in the ForCES protocol, 3792 [2]. 3793 3794 octetstring[128] 3795 3796 3797 3799 4.9. XML Schema for LFB Class Library Documents 3801 3802 3808 3809 3810 Schema for Defining LFB Classes and associated types (frames, 3811 data types for LFB attributes, and metadata). 3812 3813 3814 3815 3816 3817 3818 3819 3820 3821 3823 3825 3827 3829 3831 3832 3833 3834 3835 3836 3837 3838 3839 3840 3841 3842 3843 3844 3845 3846 3847 3848 3849 3850 3851 3852 3853 3854 3855 3856 3857 3858 3859 3860 3861 3862 3863 3864 3865 3866 3868 3869 3870 3871 3872 3873 3874 3875 3876 3877 3878 3879 3880 3881 3882 3883 3884 3890 3891 3892 3893 3894 3895 3896 3897 3898 3899 3900 3901 3902 3903 3904 3905 3906 3907 3908 3909 3910 3911 3913 3915 3917 3918 3919 3920 3921 3922 3924 3926 3927 3928 3929 3930 3931 3932 3933 3934 3935 3936 3937 3938 3939 3940 3941 3942 3943 3944 3945 3946 3948 3949 3950 3952 3953 3955 3956 3957 3958 3959 3960 3961 3962 3963 3966 3967 3968 3969 3970 3971 3972 3973 3974 3976 3977 3978 3979 3981 3982 3983 3984 3985 3986 3987 3988 3989 3990 3992 3993 3995 3996 3997 3998 3999 4000 4001 4002 4003 4004 4005 4006 4007 4008 4009 4010 4011 4012 4013 4015 4016 4017 4018 4019 4020 4021 4022 4023 4024 4025 4026 4027 4028 4029 4031 4033 4035 4037 4039 4041 4042 4043 4045 4046 4049 4050 4051 4052 4053 4054 4055 4056 4057 4058 4059 4060 4061 4062 4063 4064 4065 4066 4067 4068 4069 4070 4071 4072 4073 4074 4075 4077 4078 4079 4080 4081 4082 4083 4084 4085 4086 4088 4089 4090 4091 4092 4093 4094 4095 4097 4098 4099 4100 4101 4102 4103 4104 4105 4107 4108 4109 4110 4112 4113 4114 4115 4116 4117 4118 4119 4120 4121 4122 4123 4124 4125 4126 4127 4128 4129 4130 4131 4133 4134 4135 4136 4137 4138 4139 4140 4142 4143 4144 4145 4146 4147 4149 4150 4151 4152 4153 4154 4155 4156 4157 4158 4161 4162 4163 4164 4165 4166 4167 4169 4171 4172 4173 4174 4175 4176 4177 4179 4180 4182 4183 4184 4185 4186 4187 4188 4189 4190 4191 4192 4193 4194 4195 4196 4197 4198 4199 4200 4201 4202 4203 4204 4205 4207 4208 4209 4210 4211 4212 4213 4214 4215 4216 4217 4218 4219 4220 4221 4222 4223 4224 4225 4226 4227 4229 4230 4232 4233 4234 4235 4236 4238 4239 4240 4241 4242 4243 4244 4245 4246 4247 4248 4249 4250 4251 4252 4253 4254 4255 4256 4257 4258 4259 4260 4261 4262 4264 4265 4266 4267 4268 4269 4270 4271 4272 4273 4274 4275 4276 4277 4279 4280 4281 4283 4284 4285 4286 4288 4289 4290 4291 4293 4295 4297 4299 4301 4302 4303 4304 4306 4307 4308 4310 4312 4314 4315 4316 4318 4319 4320 4321 4322 4323 4324 4325 4326 4328 5. FE Components and Capabilities 4330 A ForCES forwarding element handles traffic on behalf of a ForCES 4331 control element. While the standards will describe the protocol and 4332 mechanisms for this control, different implementations and different 4333 instances will have different capabilities. The CE MUST be able to 4334 determine what each instance it is responsible for is actually 4335 capable of doing. As stated previously, this is an approximation. 4336 The CE is expected to be prepared to cope with errors in requests and 4337 variations in detail not captured by the capabilities information 4338 about an FE. 4340 In addition to its capabilities, an FE will have information that can 4341 be used in understanding and controlling the forwarding operations. 4342 Some of this information will be read only, while others parts may 4343 also be writeable. 4345 In order to make the FE information easily accessible, the 4346 information is represented in an LFB. This LFB has a class, 4347 FEObject. The LFBClassID for this class is 1. Only one instance of 4348 this class will ever be present in an FE, and the instance ID of that 4349 instance in the protocol is 1. Thus, by referencing the components 4350 of class:1, instance:1 a CE can get the general information about the 4351 FE. The FEObject LFB Class is described in this section. 4353 There will also be an FEProtocol LFB Class. LFBClassID 2 is reserved 4354 for that class. There will be only one instance of that class as 4355 well. Details of that class are defined in the ForCES Protocol [2] 4356 document. 4358 5.1. XML for FEObject Class definition 4360 4361 4364 4365 4366 LFBAdjacencyLimitType 4367 Describing the Adjacent LFB 4368 4369 4370 NeighborLFB 4371 ID for that LFB Class 4372 uint32 4373 4374 4375 ViaPorts 4376 4377 the ports on which we can connect 4378 4379 4380 string 4381 4382 4383 4384 4385 4386 PortGroupLimitType 4387 4388 Limits on the number of ports in a given group 4389 4390 4391 4392 PortGroupName 4393 Group Name 4394 string 4395 4396 4397 MinPortCount 4398 Minimum Port Count 4399 4400 uint32 4402 4403 4404 MaxPortCount 4405 Max Port Count 4406 4407 uint32 4408 4409 4410 4411 4412 SupportedLFBType 4413 table entry for supported LFB 4414 4415 4416 LFBName 4417 4418 The name of a supported LFB Class 4419 4420 string 4421 4422 4423 LFBClassID 4424 the id of a supported LFB Class 4425 uint32 4426 4427 4428 LFBVersion 4429 4430 The version of the LFB Class used 4431 by this FE. 4432 4433 string 4434 4435 4436 LFBOccurrenceLimit 4437 4438 the upper limit of instances of LFBs of this class 4439 4440 4441 uint32 4442 4443 4445 4446 PortGroupLimits 4447 Table of Port Group Limits 4448 4449 4450 PortGroupLimitType 4451 4452 4453 4454 4455 CanOccurAfters 4456 4457 List of LFB Classes that this LFB class can follow 4458 4459 4460 4461 LFBAdjacencyLimitType 4462 4463 4464 4466 4467 CanOccurBefores 4468 4469 List of LFB Classes that can follow this LFB class 4470 4471 4472 4473 LFBAdjacencyLimitType 4474 4475 4476 4477 UseableParentLFBClasses 4478 4479 List of LFB Classes from which this class has 4480 inherited, and which the FE is willing to allow 4481 for references to instances of this class. 4482 4483 4484 4485 uint32 4486 4487 4488 4489 4490 4491 FEStateValues 4492 The possible values of status 4493 4494 uchar 4495 4496 4497 AdminDisable 4498 4499 FE is administratively disabled 4500 4501 4502 4503 OperDisable 4504 FE is operatively disabled 4505 4506 4507 OperEnable 4508 FE is operating 4509 4510 4511 4512 4513 4514 FEConfiguredNeighborType 4515 Details of the FE's Neighbor 4516 4517 4518 NeighborID 4519 Neighbors FEID 4520 uint32 4521 4522 4523 InterfaceToNeighbor 4524 4525 FE's interface that connects to this neighbor 4526 4527 4528 string 4529 4530 4531 NeighborInterface 4532 4533 The name of the interface on the neighbor to 4534 which this FE is adjacent. This is required 4535 In case two FEs are adjacent on more than 4536 one interface. 4537 4538 4539 string 4540 4541 4542 4543 4544 LFBSelectorType 4545 4546 Unique identification of an LFB class-instance 4547 4548 4549 4550 LFBClassID 4551 LFB Class Identifier 4552 uint32 4553 4554 4555 LFBInstanceID 4556 LFB Instance ID 4557 uint32 4558 4559 4560 4561 4562 LFBLinkType 4563 4564 Link between two LFB instances of topology 4565 4566 4567 4568 FromLFBID 4569 LFB src 4570 LFBSelectorType 4571 4572 4573 FromPortGroup 4574 src port group 4575 string 4576 4577 4578 FromPortIndex 4579 src port index 4580 uint32 4581 4582 4583 ToLFBID 4584 dst LFBID 4585 LFBSelectorType 4586 4587 4588 ToPortGroup 4589 dst port group 4590 string 4591 4592 4593 ToPortIndex 4594 dst port index 4595 uint32 4596 4597 4598 4599 4600 4601 4602 FEObject 4603 Core LFB: FE Object 4604 1.0 4605 4606 4607 LFBTopology 4608 the table of known Topologies 4609 4610 LFBLinkType 4611 4612 4613 4614 LFBSelectors 4615 4616 table of known active LFB classes and 4617 instances 4618 4619 4620 LFBSelectorType 4621 4622 4623 4624 FEName 4625 name of this FE 4626 string[40] 4627 4628 4629 FEID 4630 ID of this FE 4631 uint32 4632 4633 4634 FEVendor 4635 vendor of this FE 4636 string[40] 4637 4638 4639 FEModel 4640 model of this FE 4641 string[40] 4643 4644 4645 FEState 4646 State of this FE 4647 FEStateValues 4648 4649 4650 FENeighbors 4651 table of known neighbors 4652 4653 4654 FEConfiguredNeighborType 4655 4656 4657 4658 4659 4660 ModifiableLFBTopology 4661 4662 Whether Modifiable LFB is supported 4663 4664 4665 boolean 4666 4667 4668 SupportedLFBs 4669 List of all supported LFBs 4670 4671 4672 SupportedLFBType 4673 4674 4675 4676 4677 4678 4680 5.2. FE Capabilities 4682 The FE Capability information is contained in the capabilities 4683 element of the class definition. As described elsewhere, capability 4684 information is always considered to be read-only. 4686 The currently defined capabilities are ModifiableLFBTopology and 4687 SupportedLFBs. Information as to which components of the FEObject 4688 LFB are supported is accessed by the properties information for those 4689 components. 4691 5.2.1. ModifiableLFBTopology 4693 This component has a boolean value that indicates whether the LFB 4694 topology of the FE may be changed by the CE. If the component is 4695 absent, the default value is assumed to be true, and the CE presumes 4696 the LFB topology may be changed. If the value is present and set to 4697 false, the LFB topology of the FE is fixed. If the topology is 4698 fixed, the SupportedLFBs element may be omitted, and the list of 4699 supported LFBs is inferred by the CE from the LFB topology 4700 information. If the list of supported LFBs is provided when 4701 ModifiableLFBTopology is false, the CanOccurBefore and CanOccurAfter 4702 information should be omitted. 4704 5.2.2. SupportedLFBs and SupportedLFBType 4706 One capability that the FE should include is the list of supported 4707 LFB classes. The SupportedLFBs component, is an array that contains 4708 the information about each supported LFB Class. The array structure 4709 type is defined as the SupportedLFBType dataTypeDef. 4711 Each entry in the SupportedLFBs array describes an LFB class that the 4712 FE supports. In addition to indicating that the FE supports the 4713 class, FEs with modifiable LFB topology SHOULD include information 4714 about how LFBs of the specified class may be connected to other LFBs. 4715 This information SHOULD describe which LFB classes the specified LFB 4716 class may succeed or precede in the LFB topology. The FE SHOULD 4717 include information as to which port groups may be connected to the 4718 given adjacent LFB class. If port group information is omitted, it 4719 is assumed that all port groups may be used. This capability 4720 information on the acceptable ordering and connection of LFBs MAY be 4721 omitted if the implementor concludes that the actual constraints are 4722 such that the information would be misleading for the CE. 4724 5.2.2.1. LFBName 4726 This component has as its value the name of the LFB Class being 4727 described. 4729 5.2.2.2. LFBClassID 4731 The numeric ID of the LFB Class being described. While conceptually 4732 redundant with the LFB Name, both are included for clarity and to 4733 allow consistency checking. 4735 5.2.2.3. LFBVersion 4737 The version string specifying the LFB Class version supported by this 4738 FE. As described above in versioning, an FE can support only a 4739 single version of a given LFB Class. 4741 5.2.2.4. LFBOccurrenceLimit 4743 This component, if present, indicates the largest number of instances 4744 of this LFB class the FE can support. For FEs that do not have the 4745 capability to create or destroy LFB instances, this can either be 4746 omitted or be the same as the number of LFB instances of this class 4747 contained in the LFB list attribute. 4749 5.2.2.5. PortGroupLimits and PortGroupLimitType 4751 The PortGroupLimits component is an array of information about the 4752 port groups supported by the LFB class. The structure of the port 4753 group limit information is defined by the PortGroupLimitType 4754 dataTypeDef. 4756 Each PortGroupLimits array entry contains information describing a 4757 single port group of the LFB class. Each array entry contains the 4758 name of the port group in the PortGroupName component, the fewest 4759 number of ports that can exist in the group in the MinPortCount 4760 component, and the largest number of ports that can exist in the 4761 group in the MaxPortCount component. 4763 5.2.2.6. CanOccurAfters and LFBAdjacencyLimitType 4765 The CanOccurAfters component is an array that contains the list of 4766 LFBs the described class can occur after. The array entries are 4767 defined in the LFBAdjacencyLimitType dataTypeDef. 4769 The array entries describe a permissible positioning of the described 4770 LFB class, referred to here as the SupportedLFB. Specifically, each 4771 array entry names an LFB that can topologically precede that LFB 4772 class. That is, the SupportedLFB can have an input port connected to 4773 an output port of an LFB that appears in the CanOccurAfters array. 4774 The LFB class that the SupportedLFB can follow is identified by the 4775 NeighborLFB component (of the LFBAdjacencyLimitType dataTypeDef) of 4776 the CanOccurAfters array entry. If this neighbor can only be 4777 connected to a specific set of input port groups, then the viaPort 4778 component is included. This component is an array, with one entry 4779 for each input port group of the SupportedLFB that can be connected 4780 to an output port of the NeighborLFB. 4782 [e.g., Within a SupportedLFBs entry, each array entry of the 4783 CanOccurAfters array must have a unique NeighborLFB, and within each 4784 such array entry each viaPort must represent a distinct and valid 4785 input port group of the SupportedLFB. The LFB Class definition 4786 schema does not include these uniqueness constraints.] 4788 5.2.2.7. CanOccurBefores and LFBAdjacencyLimitType 4790 The CanOccurBefores array holds the information about which LFB 4791 classes can follow the described class. Structurally this element 4792 parallels CanOccurAfters, and uses the same type definition for the 4793 array entries. 4795 The array entries list those LFB classes that the SupportedLFB may 4796 precede in the topology. In this component, the entries in the 4797 viaPort component of the array value represent the output port groups 4798 of the SupportedLFB that may be connected to the NeighborLFB. As 4799 with CanOccurAfters, viaPort may have multiple entries if multiple 4800 output ports may legitimately connect to the given NeighborLFB class. 4802 [And a similar set of uniqueness constraints apply to the 4803 CanOccurBefore clauses, even though an LFB may occur both in 4804 CanOccurAfter and CanOccurBefore.] 4806 5.2.2.8. UseableParentLFBClasses 4808 The UseableParentLFBClasses array, if present, is used to hold a list 4809 of parent LFB class IDs. All the entries in the list must be IDs of 4810 classes from which the SupportedLFB Class being described has 4811 inherited (either directly, or through an intermediate parent.) (If 4812 an FE includes improper values in this list, improper manipulations 4813 by the CE are likely, and operational failures are likely.) In 4814 addition, the FE, by including a given class in the last, is 4815 indicating to the CE that a given parent class may be used to 4816 manipulate an instance of this supported LFB class. 4818 By allowing such substitution, the FE allows for the case where an 4819 instantiated LFB may be of a class not known to the CE, but could 4820 still be manipulated. While it is hoped that such situations are 4821 rare, it is desirable for this to be supported. This can occur if an 4822 FE locally defines certain LFB instances, or if an earlier CE had 4823 configured some LFB instances. It can also occur if the FE would 4824 prefer to instantiate a more recent, more specific and suitable, LFB 4825 class rather than a common parent. 4827 In order to permit this, the FE MUST be more restrained in assigning 4828 LFB Instance IDs. Normally, instance IDs are qualified by the LFB 4829 class. However, if two LFB classes share a parent, and if that 4830 parent is listed in the UseableParentLFBClasses for both specific LFB 4831 classes, then all the instances of both (or any, if multiple classes 4832 are listing the common parent) MUST use distinct instances. This 4833 permits the FE to determine which LFB Instance is intended by CE 4834 manipulation operations even when a parent class is used. 4836 5.2.2.9. LFBClassCapabilities 4838 While it would be desirable to include class capability level 4839 information, this is not included in the model. While such 4840 information belongs in the FE Object in the supported class table, 4841 the contents of that information would be class specific. The 4842 currently expected encoding structures for transferring information 4843 between the CE and FE are such that allowing completely unspecified 4844 information would be likely to induce parse errors. We could specify 4845 that the information is encoded in an octetstring, but then we would 4846 have to define the internal format of that octet string. 4848 As there also are not currently any defined LFB Class level 4849 Capabilities that the FE needs to report, this information is not 4850 present now, but may be added in a future version of the FE Object. 4851 (This is an example of a case where versioning, rather than 4852 inheritance, would be needed, since the FE Object must have class ID 4853 1 and instance ID 1 so that the protocol behavior can start by 4854 finding this object.) 4856 5.3. FE Components 4858 The element is included if the class definition contains 4859 the definition of the components of the FE Object that are not 4860 considered "capabilities". Some of these components are writeable, 4861 and some are read-only, which is determinable by examining the 4862 property information of the components. 4864 5.3.1. FEState 4866 This component carries the overall state of the FE. The possible 4867 values are the strings AdminDisable, OperDisable and OperEnable. The 4868 starting state is OperDisable, and the transition to OperEnable is 4869 controlled by the FE. The CE controls the transition from OperEnable 4870 to/from AdminDisable. For details refer to the ForCES Protocol 4871 document [2]. 4873 5.3.2. LFBSelectors and LFBSelectorType 4875 The LFBSelectors component is an array of information about the LFBs 4876 currently accessible via ForCES in the FE. The structure of the LFB 4877 information is defined by the LFBSelectorType dataTypeDef. 4879 Each entry in the array describes a single LFB instance in the FE. 4880 The array entry contains the numeric class ID of the class of the LFB 4881 instance and the numeric instance ID for this instance. 4883 5.3.3. LFBTopology and LFBLinkType 4885 The optional LFBTopology component contains information about each 4886 inter-LFB link inside the FE, where each link is described in an 4887 LFBLinkType dataTypeDef. The LFBLinkType component contains 4888 sufficient information to identify precisely the end points of a 4889 link. The FromLFBID and ToLFBID components specify the LFB instances 4890 at each end of the link, and MUST reference LFBs in the LFB instance 4891 table. The FromPortGroup and ToPortGroup MUST identify output and 4892 input port groups defined in the LFB classes of the LFB instances 4893 identified by FromLFBID and ToLFBID. The FromPortIndex and 4894 ToPortIndex components select the entries from the port groups that 4895 this link connects. All links are uniquely identified by the 4896 FromLFBID, FromPortGroup, and FromPortIndex fields. Multiple links 4897 may have the same ToLFBID, ToPortGroup, and ToPortIndex as this model 4898 supports fan-in of inter- LFB links but not fan-out. 4900 5.3.4. FENeighbors and FEConfiguredNeighborType 4902 The FENeighbors component is an array of information about manually 4903 configured adjacencies between this FE and other FEs. The content of 4904 the array is defined by the FEConfiguredNeighborType dataTypeDef. 4906 This array is intended to capture information that may be configured 4907 on the FE and is needed by the CE, where one array entry corresponds 4908 to each configured neighbor. Note that this array is not intended to 4909 represent the results of any discovery protocols, as those will have 4910 their own LFBs. This component is optional. 4912 While there may be many ways to configure neighbors, the FE-ID is the 4913 best way for the CE to correlate entities. And the interface 4914 identifier (name string) is the best correlator. The CE will be able 4915 to determine the IP address and media level information about the 4916 neighbor from the neighbor directly. Omitting that information from 4917 this table avoids the risk of incorrect double configuration. 4919 Information about the intended forms of exchange with a given 4920 neighbor is not captured here, only the adjacency information is 4921 included. 4923 5.3.4.1. NeighborID 4925 This is the ID in some space meaningful to the CE for the neighbor. 4927 5.3.4.2. InterfaceToNeighbor 4929 This identifies the interface through which the neighbor is reached. 4931 5.3.4.3. NeighborInterface 4933 This identifies the interface on the neighbor through which the 4934 neighbor is reached. The interface identification is needed when 4935 either only one side of the adjacency has configuration information, 4936 or the two FEs are adjacent on more than one interface. 4938 6. Satisfying the Requirements on FE Model 4940 This section describes how the proposed FE model meets the 4941 requirements outlined in Section 5 of RFC3654 [4]. The requirements 4942 can be separated into general requirements (Section 5, 5.1 - 5.4) and 4943 the specification of the minimal set of logical functions that the FE 4944 model must support (Section 5.5). 4946 The general requirement on the FE model is that it be able to express 4947 the logical packet processing capability of the FE, through both a 4948 capability and a state model. In addition, the FE model is expected 4949 to allow flexible implementations and be extensible to allow defining 4950 new logical functions. 4952 A major component of the proposed FE model is the Logical Function 4953 Block (LFB) model. Each distinct logical function in an FE is 4954 modeled as an LFB. Operational parameters of the LFB that must be 4955 visible to the CE are conceptualized as LFB components. These 4956 components express the capability of the FE and support flexible 4957 implementations by allowing an FE to specify which optional features 4958 are supported. The components also indicate whether they are 4959 configurable by the CE for an LFB class. Configurable components 4960 provide the CE some flexibility in specifying the behavior of an LFB. 4961 When multiple LFBs belonging to the same LFB class are instantiated 4962 on an FE, each of those LFBs could be configured with different 4963 component settings. By querying the settings of the components for 4964 an instantiated LFB, the CE can determine the state of that LFB. 4966 Instantiated LFBs are interconnected in a directed graph that 4967 describes the ordering of the functions within an FE. This directed 4968 graph is described by the topology model. The combination of the 4969 components of the instantiated LFBs and the topology describe the 4970 packet processing functions available on the FE (current state). 4972 Another key component of the FE model is the FE components. The FE 4973 components are used mainly to describe the capabilities of the FE, 4974 but they also convey information about the FE state. 4976 The FE model includes only the definition of the FE Object LFB 4977 itself. Meeting the full set of working group requirements requires 4978 other LFBs. The class definitions for those LFBs will be provided in 4979 other documents. 4981 7. Using the FE model in the ForCES Protocol 4983 The actual model of the forwarding plane in a given NE is something 4984 the CE must learn and control by communicating with the FEs (or by 4985 other means). Most of this communication will happen in the post- 4986 association phase using the ForCES protocol. The following types of 4987 information must be exchanged between CEs and FEs via the ForCES 4988 Protocol [2]: 4990 1. FE topology query; 4992 2. FE capability declaration; 4994 3. LFB topology (per FE) and configuration capabilities query; 4996 4. LFB capability declaration; 4998 5. State query of LFB components; 5000 6. Manipulation of LFB components; 5002 7. LFB topology reconfiguration. 5004 Items 1) through 5) are query exchanges, where the main flow of 5005 information is from the FEs to the CEs. Items 1) through 4) are 5006 typically queried by the CE(s) in the beginning of the post- 5007 association (PA) phase, though they may be repeatedly queried at any 5008 time in the PA phase. Item 5) (state query) will be used at the 5009 beginning of the PA phase, and often frequently during the PA phase 5010 (especially for the query of statistical counters). 5012 Items 6) and 7) are "command" types of exchanges, where the main flow 5013 of information is from the CEs to the FEs. Messages in Item 6) (the 5014 LFB re-configuration commands) are expected to be used frequently. 5015 Item 7) (LFB topology re-configuration) is needed only if dynamic LFB 5016 topologies are supported by the FEs and it is expected to be used 5017 infrequently. 5019 The inter-FE topology (item 1 above) can be determined by the CE in 5020 many ways. Neither this document nor the ForCES Protocol [2] 5021 document mandates a specific mechanism. The LFB Class definition 5022 does include the capability for an FE to be configured with, and to 5023 provide to the CE in response to a query, the identity of its 5024 neighbors. There may also be defined specific LFB classes and 5025 protocols for neighbor discovery. Routing protocols may be used by 5026 the CE for adjacency determination. The CE may be configured with 5027 the relevant information. 5029 The relationship between the FE model and the seven post-association 5030 messages are visualized in Figure 12: 5032 +--------+ 5033 ..........-->| CE | 5034 /----\ . +--------+ 5035 \____/ FE Model . ^ | 5036 | |................ (1),2 | | 6, 7 5037 | | (off-line) . 3, 4, 5 | | 5038 \____/ . | v 5039 . +--------+ 5040 e.g. RFCs ..........-->| FE | 5041 +--------+ 5043 Figure 12: Relationship between the FE model and the ForCES protocol 5044 messages, where (1) is part of the ForCES base protocol, and the 5045 rest are defined by the FE model. 5047 The actual encoding of these messages is defined by the ForCES 5048 Protocol [2] document and is beyond the scope of the FE model. Their 5049 discussion is nevertheless important here for the following reasons: 5051 o These PA model components have considerable impact on the FE 5052 model. For example, some of the above information can be 5053 represented as components of the LFBs, in which case such 5054 components must be defined in the LFB classes. 5056 o The understanding of the type of information that must be 5057 exchanged between the FEs and CEs can help to select the 5058 appropriate protocol format and the actual encoding method (such 5059 as XML, TLVs). 5061 o Understanding the frequency of these types of messages should 5062 influence the selection of the protocol format (efficiency 5063 considerations). 5065 The remaining sub-sections of this section address each of the seven 5066 message types. 5068 7.1. FE Topology Query 5070 An FE may contain zero, one or more external ingress ports. 5071 Similarly, an FE may contain zero, one or more external egress ports. 5072 In other words, not every FE has to contain any external ingress or 5073 egress interfaces. For example, Figure 13 shows two cascading FEs. 5074 FE #1 contains one external ingress interface but no external egress 5075 interface, while FE #2 contains one external egress interface but no 5076 ingress interface. It is possible to connect these two FEs together 5077 via their internal interfaces to achieve the complete ingress-to- 5078 egress packet processing function. This provides the flexibility to 5079 spread the functions across multiple FEs and interconnect them 5080 together later for certain applications. 5082 While the inter-FE communication protocol is out of scope for ForCES, 5083 it is up to the CE to query and understand how multiple FEs are 5084 inter-connected to perform a complete ingress-egress packet 5085 processing function, such as the one described in Figure 13. The 5086 inter-FE topology information may be provided by FEs, may be hard- 5087 coded into CE, or may be provided by some other entity (e.g., a bus 5088 manager) independent of the FEs. So while the ForCES Protocol [2] 5089 supports FE topology query from FEs, it is optional for the CE to use 5090 it, assuming the CE has other means to gather such topology 5091 information. 5093 +-----------------------------------------------------+ 5094 | +---------+ +------------+ +---------+ | 5095 input| | | | | | output | 5096 ---+->| Ingress |-->|Header |-->|IPv4 |---------+--->+ 5097 | | port | |Decompressor| |Forwarder| FE | | 5098 | +---------+ +------------+ +---------+ #1 | | 5099 +-----------------------------------------------------+ V 5100 | 5101 +-----------------------<-----------------------------+ 5102 | 5103 | +----------------------------------------+ 5104 V | +------------+ +----------+ | 5105 | input | | | | output | 5106 +->--+->|Header |-->| Egress |---------+--> 5107 | |Compressor | | port | FE | 5108 | +------------+ +----------+ #2 | 5109 +----------------------------------------+ 5111 Figure 13: An example of two FEs connected together 5113 Once the inter-FE topology is discovered by the CE after this query, 5114 it is assumed that the inter-FE topology remains static. However, it 5115 is possible that an FE may go down during the NE operation, or a 5116 board may be inserted and a new FE activated, so the inter-FE 5117 topology will be affected. It is up to the ForCES protocol to 5118 provide a mechanism for the CE to detect such events and deal with 5119 the change in FE topology. FE topology is outside the scope of the 5120 FE model. 5122 7.2. FE Capability Declarations 5124 FEs will have many types of limitations. Some of the limitations 5125 must be expressed to the CEs as part of the capability model. The 5126 CEs must be able to query these capabilities on a per-FE basis. 5127 Examples: 5129 o Metadata passing capabilities of the FE. Understanding these 5130 capabilities will help the CE to evaluate the feasibility of LFB 5131 topologies, and hence to determine the availability of certain 5132 services. 5134 o Global resource query limitations (applicable to all LFBs of the 5135 FE). 5137 o LFB supported by the FE. 5139 o LFB class instantiation limit. 5141 o LFB topological limitations (linkage constraint, ordering etc.) 5143 7.3. LFB Topology and Topology Configurability Query 5145 The ForCES protocol must provide the means for the CEs to discover 5146 the current set of LFB instances in an FE and the interconnections 5147 between the LFBs within the FE. In addition, sufficient information 5148 should be available to determine whether the FE supports any CE- 5149 initiated (dynamic) changes to the LFB topology, and if so, determine 5150 the allowed topologies. Topology configurability can also be 5151 considered as part of the FE capability query as described in Section 5152 9.3. 5154 7.4. LFB Capability Declarations 5156 LFB class specifications define a generic set of capabilities. When 5157 an LFB instance is implemented (instantiated) on a vendor's FE, some 5158 additional limitations may be introduced. Note that we discuss only 5159 those limitations that are within the flexibility of the LFB class 5160 specification. That is, the LFB instance will remain compliant with 5161 the LFB class specification despite these limitations. For example, 5162 certain features of an LFB class may be optional, in which case it 5163 must be possible for the CE to determine if an optional feature is 5164 supported by a given LFB instance or not. Also, the LFB class 5165 definitions will probably contain very few quantitative limits (e.g., 5166 size of tables), since these limits are typically imposed by the 5167 implementation. Therefore, quantitative limitations should always be 5168 expressed by capability arguments. 5170 LFB instances in the model of a particular FE implementation will 5171 possess limitations on the capabilities defined in the corresponding 5172 LFB class. The LFB class specifications must define a set of 5173 capability arguments, and the CE must be able to query the actual 5174 capabilities of the LFB instance via querying the value of such 5175 arguments. The capability query will typically happen when the LFB 5176 is first detected by the CE. Capabilities need not be re-queried in 5177 case of static limitations. In some cases, however, some 5178 capabilities may change in time (e.g., as a result of adding/removing 5179 other LFBs, or configuring certain components of some other LFB when 5180 the LFBs share physical resources), in which case additional 5181 mechanisms must be implemented to inform the CE about the changes. 5183 The following two broad types of limitations will exist: 5185 o Qualitative restrictions. For example, a standardized multi- 5186 field classifier LFB class may define a large number of 5187 classification fields, but a given FE may support only a subset of 5188 those fields. 5190 o Quantitative restrictions, such as the maximum size of tables, 5191 etc. 5193 The capability parameters that can be queried on a given LFB class 5194 will be part of the LFB class specification. The capability 5195 parameters should be regarded as special components of the LFB. The 5196 actual values of these components may be, therefore, obtained using 5197 the same component query mechanisms as used for other LFB components. 5199 Capability components are read-only arguments. In cases where some 5200 implementations may allow CE modification of the value, the 5201 information must be represented as an operational component, not a 5202 capability component. 5204 Assuming that capabilities will not change frequently, the efficiency 5205 of the protocol/schema/encoding is of secondary concern. 5207 Much of this restrictive information is captured by the component 5208 property information, and so can be access uniformly for all 5209 information within the model. 5211 7.5. State Query of LFB Components 5213 This feature must be provided by all FEs. The ForCES protocol and 5214 the data schema/encoding conveyed by the protocol must together 5215 satisfy the following requirements to facilitate state query of the 5216 LFB components: 5218 o Must permit FE selection. This is primarily to refer to a single 5219 FE, but referring to a group of (or all) FEs may optionally be 5220 supported. 5222 o Must permit LFB instance selection. This is primarily to refer to 5223 a single LFB instance of an FE, but optionally addressing of a 5224 group of LFBs (or all) may be supported. 5226 o Must support addressing of individual components of an LFB. 5228 o Must provide efficient encoding and decoding of the addressing 5229 info and the configured data. 5231 o Must provide efficient data transmission of the component state 5232 over the wire (to minimize communication load on the CE-FE link). 5234 7.6. LFB Component Manipulation 5236 The FE Model provides for the definition of LFB Classes. Each class 5237 has a globally unique identifier. Information within the class is 5238 represented as components and assigned identifiers within the scope 5239 of that class. This model also specifies that instances of LFB 5240 Classes have identifiers. The combination of class identifiers, 5241 instance identifiers, and component identifiers are used by the 5242 protocol to reference the LFB information in the protocol operations. 5244 7.7. LFB Topology Re-configuration 5246 Operations that will be needed to reconfigure LFB topology: 5248 o Create a new instance of a given LFB class on a given FE. 5250 o Connect a given output of LFB x to the given input of LFB y. 5252 o Disconnect: remove a link between a given output of an LFB and a 5253 given input of another LFB. 5255 o Delete a given LFB (automatically removing all interconnects to/ 5256 from the LFB). 5258 8. Example LFB Definition 5260 This section contains an example LFB definition. While some 5261 properties of LFBs are shown by the FE Object LFB, this endeavors to 5262 show how a data plane LFB might be build. This example is a 5263 fictional case of an interface supporting a coarse WDM optical 5264 interface that carries Frame Relay traffic. The statistical 5265 information (including error statistics) is omitted. 5267 Later portions of this example include references to protocol 5268 operations. The operations described are operations the protocol 5269 needs to support. The exact format and fields are purely 5270 informational here, as the ForCES Protocol [2] document defines the 5271 precise syntax and semantics of its operations. 5273 5274 5277 5278 5279 FRFrame 5280 5281 A frame relay frame, with DLCI without 5282 stuffing) 5283 5284 5285 5286 IPFrame 5287 An IP Packet 5288 5289 5290 5291 5292 frequencyInformationType 5293 5294 Information about a single CWDM frequency 5295 5296 5297 5298 LaserFrequency 5299 encoded frequency(channel) 5300 uint32 5301 5302 5303 FrequencyState 5304 state of this frequency 5305 PortStatusValues 5307 5308 5309 LaserPower 5310 current observed power 5311 uint32 5312 5313 5314 FrameRelayCircuits 5315 5316 Information about circuits on this Frequency 5317 5318 5319 frameCircuitsType 5320 5321 5322 5323 5324 5325 frameCircuitsType 5326 5327 Information about a single Frame Relay circuit 5328 5329 5330 5331 DLCI 5332 DLCI of the circuit 5333 uint32 5334 5335 5336 CircuitStatus 5337 state of the circuit 5338 PortStatusValues 5339 5340 5341 isLMI 5342 is this the LMI circuit 5343 boolean 5344 5345 5346 associatedPort 5347 5348 which input / output port is associated 5349 with this circuit 5350 5351 uint32 5352 5353 5354 5355 5356 PortStatusValues 5357 5358 The possible values of status. Used for both 5359 administrative and operational status 5360 5361 5362 uchar 5363 5364 5365 Disabled 5366 the component is disabled 5367 5368 5369 Enabled 5370 FE is operatively enabled 5371 5372 5373 5374 5375 5376 5377 5378 DLCI 5379 The DLCI the frame arrived on 5380 12 5381 uint32 5382 5383 5384 LaserChannel 5385 The index of the laser channel 5386 34 5387 uint32 5388 5389 5390 5391 5392 5393 FrameLaserLFB 5394 Fictional LFB for Demonstrations 5395 1.0 5396 5397 5398 LMIfromFE 5399 5400 Ports for LMI traffic, for transmission 5401 5402 5403 5404 FRFrame 5405 5406 5407 DLCI 5408 LaserChannel 5409 5410 5411 5412 5413 DatafromFE 5414 5415 Ports for data to be sent on circuits 5416 5417 5418 5419 IPFrame 5420 5421 5422 DLCI 5423 LaserChannel 5424 5425 5426 5427 5428 5429 5430 LMItoFE 5431 5432 Ports for LMI traffic for processing 5433 5434 5435 5436 FRFrame 5437 5438 5439 DLCI 5440 LaserChannel 5441 5442 5443 5444 5445 DatatoFE 5446 5447 Ports for Data traffic for processing 5448 5449 5450 5451 IPFrame 5452 5453 5454 DLCI 5455 LaserChannel 5456 5457 5458 5459 5460 5461 5462 AdminPortState 5463 is this port allowed to function 5464 PortStatusValues 5465 5466 5467 FrequencyInformation 5468 5469 table of information per CWDM frequency 5470 5471 5472 frequencyInformationType 5473 5474 5475 5476 5477 5478 OperationalState 5479 5480 whether the port over all is operational 5481 5482 PortStatusValues 5483 5484 5485 MaximumFrequencies 5486 5487 how many laser frequencies are there 5488 5489 uint16 5490 5491 5492 MaxTotalCircuits 5493 5494 Total supportable Frame Relay Circuits, across 5495 all laser frequencies 5496 5497 5498 uint32 5500 5501 5502 5503 5504 FrequencyState 5505 5506 The state of a frequency has changed 5507 5508 5509 FrequencyInformation 5510 _FrequencyIndex_ 5511 FrequencyState 5512 5513 5514 5515 5516 5517 FrequencyInformation 5518 _FrequencyIndex_ 5519 FrequencyState 5520 5521 5522 5523 5524 CreatedFrequency 5525 A new frequency has appeared 5526 5527 FrequencyInformation> 5528 _FrequencyIndex_ 5529 5530 5531 5532 5533 FrequencyInformation 5534 _FrequencyIndex_ 5535 LaserFrequency 5536 5537 5538 5539 5540 DeletedFrequency 5541 5542 A frequency Table entry has been deleted 5543 5544 5545 FrequencyInformation 5546 _FrequencyIndex_ 5547 5548 5549 5550 5551 PowerProblem 5552 5553 there are problems with the laser power level 5554 5555 5556 FrequencyInformation 5557 _FrequencyIndex_ 5558 LaserPower 5559 5560 5561 5562 5563 FrequencyInformation 5564 _FrequencyIndex_ 5565 LaserPower 5566 5567 5568 FrequencyInformation 5569 _FrequencyIndex_ 5570 LaserFrequency 5571 5572 5573 5574 5575 FrameCircuitChanged 5576 5577 the state of an Fr circuit on a frequency 5578 has changed 5579 5580 5581 FrequencyInformation 5582 _FrequencyIndex_ 5583 FrameRelayCircuits 5584 FrameCircuitIndex 5585 CircuitStatus 5586 5587 5588 5589 5590 FrequencyInformation 5591 _FrequencyIndex_ 5592 FrameRelayCircuits 5593 FrameCircuitIndex 5594 CircuitStatus 5595 5596 5597 FrequencyInformation 5598 _FrequencyIndex_ 5599 FrameRelayCircuits 5600 FrameCircuitIndex 5601 DLCI 5602 5603 5604 5605 5606 5607 5608 5610 8.1. Data Handling 5612 This LFB is designed to handle data packets coming in from or going 5613 out to the external world. It is not a full port, and it lacks many 5614 useful statistics, but it serves to show many of the relevant 5615 behaviors. The following paragraphs describe a potential operational 5616 device and how it might use this LFB definition. 5618 Packets arriving without error from the physical interface come in on 5619 a Frame Relay DLCI on a laser channel. These two values are used by 5620 the LFB to look up the handling for the packet. If the handling 5621 indicates that the packet is LMI, then the output index is used to 5622 select an LFB port from the LMItoFE port group. The packet is sent 5623 as a full Frame Relay frame (without any bit or byte stuffing) on the 5624 selected port. The laser channel and DLCI are sent as meta-data, 5625 even though the DLCI is also still in the packet. 5627 Good packets that arrive and are not LMI and have a frame relay type 5628 indicator of IP are sent as IP packets on the port in the DatatoFE 5629 port group, using the same index field from the table based on the 5630 laser channel and DLCI. The channel and DLCI are attached as meta- 5631 data for other use (classifiers, for example.) 5633 The current definition does not specify what to do if the Frame Relay 5634 type information is not IP. 5636 Packets arriving on input ports arrive with the Laser Channel and 5637 Frame Relay DLCI as meta-data. As such, a single input port could 5638 have been used. With the structure that is defined (which parallels 5639 the output structure), the selection of channel and DLCI could be 5640 restricted by the arriving input port group (LMI vs. data) and port 5641 index. As an alternative LFB design, the structures could require a 5642 1-1 relationship between DLCI and LFB port, in which case no meta- 5643 data would be needed. This would however be quite complex and noisy. 5645 The intermediate level of structure here allows parallelism between 5646 input and output, without requiring excessive ports. 5648 8.1.1. Setting up a DLCI 5650 When a CE chooses to establish a DLCI on a specific laser channel, it 5651 sends a SET request directed to this LFB. The request might look 5652 like 5654 T = SET 5655 T = PATH-DATA 5656 Path: flags = none, length = 4, path = 2, channel, 4, entryIdx 5657 DataRaw: DLCI, Enabled(1), false, out-idx 5659 Which would establish the DLCI as enabled, with traffic going to a 5660 specific entry of the output port group DatatoFE. (The CE would 5661 ensure that output port is connected to the right place before 5662 issuing this request.) 5664 The response would confirm the creation of the specified entry. This 5665 table is structured to use separate internal indices and DLCIs. An 5666 alternative design could have used the DLCI as index, trading off 5667 complexities. 5669 One could also imagine that the FE has an LMI LFB. Such an LFB would 5670 be connected to the LMItoFE and LMIfromFE port groups. It would 5671 process LMI information. It might be the LFBs job to set up the 5672 frame relay circuits. The LMI LFB would have an alias entry that 5673 points to the Frame Relay circuits table it manages, so that it can 5674 manipulate those entities. 5676 8.1.2. Error Handling 5678 The LFB will receive invalid packets over the wire. Many of these 5679 will simply result in incrementing counters. The LFB designer might 5680 also specify some error rate measures. This puts more work on the 5681 FE, but allows for more meaningful alarms. 5683 There may be some error conditions that should cause parts of the 5684 packet to be sent to the CE. The error itself is not something that 5685 can cause an event in the LFB. There are two ways this can be 5686 handled. 5688 One way is to define a specific component to count the error, and a 5689 component in the LFB to hold the required portion of the packet. The 5690 component could be defined to hold the portion of the packet from the 5691 most recent error. One could then define an event that occurs 5692 whenever the error count changes, and declare that reporting the 5693 event includes the LFB field with the packet portion. For rare but 5694 extremely critical errors, this is an effective solution. It ensures 5695 reliable delivery of the notification. And it allows the CE to 5696 control if it wants the notification. 5698 Another approach is for the LFB to have a port that connects to a 5699 redirect sink. The LFB would attach the laser channel, the DLCI, and 5700 the error indication as meta-data, and ship the packet to the CE. 5702 Other aspects of error handling are discussed under events below. 5704 8.2. LFB Components 5706 This LFB is defined to have two top level components. One reflects 5707 the administrative state of the LFB. This allows the CE to disable 5708 the LFB completely. 5710 The other component is the table of information about the laser 5711 channels. It is a variable sized array. Each array entry contains 5712 an identifier for what laser frequency this entry is associated with, 5713 whether that frequency is operational, the power of the laser at that 5714 frequency, and a table of information about frame relay circuits on 5715 this frequency. There is no administrative status since a CE can 5716 disable an entry simply by removing it. (Frequency and laser power 5717 of a non-operational channel are not particularly useful. Knowledge 5718 about what frequencies can be supported would be a table in the 5719 capabilities section.) 5721 The Frame Relay circuit information contains the DLCI, the 5722 operational circuit status, whether this circuit is to be treated as 5723 carrying LMI information, and which port in the output port group of 5724 the LFB traffic is to be sent to. As mentioned above, the circuit 5725 index could, in some designs, be combined with the DLCI. 5727 8.3. Capabilities 5729 The capability information for this LFB includes whether the 5730 underlying interface is operational, how many frequencies are 5731 supported, and how many total circuits, across all channels, are 5732 permitted. The maximum number for a given laser channel can be 5733 determined from the properties of the FrameRelayCircuits table. A 5734 GET-PROP on path 2.channel.4 will give the CE the properties of that 5735 FrameRelayCircuits array which include the number of entries used, 5736 the first available entry, and the maximum number of entries 5737 permitted. 5739 8.4. Events 5741 This LFB is defined to be able to generate several events that the CE 5742 may be interested in. There are events to report changes in 5743 operational state of frequencies, and the creation and deletion of 5744 frequency entries. There is an event for changes in status of 5745 individual frame relay circuits. So an event notification of 5746 61.5.3.11 would indicate that there had been a circuit status change 5747 on subscript 11 of the circuit table in subscript 3 of the frequency 5748 table. The event report would include the new status of the circuit 5749 and the DLCI of the circuit. Arguably, the DLCI is redundant, since 5750 the CE presumably knows the DLCI based on the circuit index. It is 5751 included here to show including two pieces of information in an event 5752 report. 5754 As described above, the event declaration defines the event target, 5755 the event condition, and the event report content. The event 5756 properties indicate whether the CE is subscribed to the event, the 5757 specific threshold for the event, and any filter conditions for the 5758 event. 5760 Another event shown is a laser power problem. This event is 5761 generated whenever the laser falls below the specified threshold. 5762 Thus, a CE can register for the event of laser power loss on all 5763 circuits. It would do this by: 5765 T = SET-PROP 5766 Path-TLV: flags=0, length = 2, path = 61.4 5767 Path-TLV: flags = property-field, length = 1, path = 2 5768 Content = 1 (register) 5769 Path-TLV: flags = property-field, length = 1, path = 3 5770 Content = 15 (threshold) 5772 This would set the registration for the event on all entries in the 5773 table. It would also set the threshold for the event, causing 5774 reporting if the power falls below 15. (Presumably, the CE knows 5775 what the scale is for power, and has chosen 15 as a meaningful 5776 problem level.) 5778 If a laser oscillates in power near the 15 mark, one could get a lot 5779 of notifications. (If it flips back and forth between 14 and 15, 5780 each flip down will generate an event.) Suppose that the CE decides 5781 to suppress this oscillation somewhat on laser channel 5. It can do 5782 this by setting the hysteresis property on that event. The request 5783 would look like: 5785 T = SET-PROP 5786 Path-TLV: flags=0, length = 3, path = 61.4.5 5787 Path-TLV: flags = property-field, length = 1, path = 4 5788 Content = 2 (hysteresis) 5790 Setting the hysteresis to 2 suppress a lot of spurious notifications. 5791 When the level first falls below 10, a notification is generated. If 5792 the power level increases to 10 or 11, and then falls back below 10, 5793 an event will not be generated. The power has to recover to at least 5794 12 and fall back below 10 to generate another event. One common 5795 cause of this form of oscillation is when the actual value is right 5796 near the border. If it is really 9.5, tiny changes might flip it 5797 back and forth between 9 and 10. A hysteresis level of 1 will 5798 suppress this sort of condition. Many other events have oscillations 5799 that are somewhat wider, so larger hysteresis settings can be used 5800 with those. 5802 9. IANA Considerations 5804 The ForCES model creates the need for a unique XML namespace for 5805 ForCES library definition usage, and unique class names and numeric 5806 class identifiers. 5808 9.1. URN Namespace Registration 5810 IANA is requested to register a new XML namespace, as per the 5811 guidelines in RFC3688 [3]. 5813 URI: The URI for this namespace is 5814 urn:ietf:params:xml:ns:forces:lfbmodel:1.0 5816 Registrant Contact: IESG 5818 XML: none, this is an XML namespace 5820 9.2. LFB Class Names and LFB Class Identifiers 5822 In order to have well defined ForCES LFB Classes, and well defined 5823 identifiers for those classes, a registry of LFB Class names, 5824 corresponding class identifiers, and the document which defines the 5825 LFB Class is needed. The registry policy is simply first come first 5826 served(FCFS) with regard to LFB Class names. With regard to LFB 5827 Class identifiers, identifiers less than 65536 are reserved for 5828 assignment by IETF Standards Track RFCs. Identifiers above 65536, in 5829 the 32 bit class ID space, are available for assignment on a first 5830 come, first served basis. All Registry entries must be documented in 5831 a stable, publicly available form. 5833 Since this registry provides for FCFS allocation of a portion of the 5834 class identifier space, it is necessary to define rules for naming 5835 classes that are using that space. As these can be defined by 5836 anyone, the needed rule is to keep the FCFS class names from 5837 colliding with IETF defined class names. Therefore, all FCFS class 5838 names MUST start with the string "Ext-". 5840 Table 1 tabulates the above information. 5842 IANA is requested to create a register of ForCES LFB Class Names and 5843 the corresponding ForCES LFB Class Identifiers, with the location of 5844 the definition of the ForCES LFB Class, in accordance with the rules 5845 in the following table. 5847 +----------------+------------+---------------+---------------------+ 5848 | LFB Class Name | LFB Class | Place Defined | Description | 5849 | | Identifier | | | 5850 +----------------+------------+---------------+---------------------+ 5851 | Reserved | 0 | RFCxxxx | Reserved | 5852 | | | | -------- | 5853 | FE Object | 1 | RFCxxxx | Defines ForCES | 5854 | | | | Forwarding Element | 5855 | | | | information | 5856 | FE Protocol | 2 | [2] | Defines parameters | 5857 | Object | | | for the ForCES | 5858 | | | | protocol operation | 5859 | | | | -------- | 5860 | IETF defined | 3-65535 | Standards | Reserved for IETF | 5861 | LFBs | | Track RFCs | defined RFCs | 5862 | | | | -------- | 5863 | Forces LFB | >65535 | Any Publicly | First Come, First | 5864 | Class names | | Available | Served for any use | 5865 | beginning EXT- | | Document | | 5866 +----------------+------------+---------------+---------------------+ 5868 Table 1 5870 [Note to RFC Editor, RFCxxxx above is to be changed to the RFC number 5871 assigned to this document for publication.] 5873 10. Authors Emeritus 5875 The following are the authors who were instrumental in the creation 5876 of earlier releases of this document. 5878 Ellen Delganes, Intel Corp. 5879 Lily Yang, Intel Corp. 5880 Ram Gopal, Nokia Research Center 5881 Alan DeKok, Infoblox, Inc. 5882 Zsolt Haraszti, Clovis Solutions 5884 11. Acknowledgments 5886 Many of the colleagues in our companies and participants in the 5887 ForCES mailing list have provided invaluable input into this work. 5888 Particular thanks to Evangelos Haleplidis for help getting the XML 5889 right. 5891 12. Security Considerations 5893 The FE model describes the representation and organization of data 5894 sets and components in the FEs. The ForCES framework document [2] 5895 provides a comprehensive security analysis for the overall ForCES 5896 architecture. For example, the ForCES protocol entities must be 5897 authenticated per the ForCES requirements before they can access the 5898 information elements described in this document via ForCES. Access 5899 to the information contained in the FE model is accomplished via the 5900 ForCES protocol, which will be defined in separate documents, and 5901 thus the security issues will be addressed there. 5903 13. References 5905 13.1. Normative References 5907 [1] Bradner, S., "Key words for use in RFCs to Indicate Requirement 5908 Levels", BCP 14, RFC 2119, March 1997. 5910 [2] Doria, A., Haas, R., Hadi Salim, J., Khosravi, H., and W. Wang, 5911 "ForCES Protocol Specification", work in progress, draft-ietf - 5912 forces-protocol-11.txt, December 2007. 5914 [3] Mealling, M., "The IETF XML Registry", BCP 81, RFC 3688, 5915 January 2004. 5917 13.2. Informative References 5919 [4] Khosravi, H. and T. Anderson, "Requirements for Separation of 5920 IP Control and Forwarding", RFC 3654, November 2003. 5922 [5] Yang, L., Dantu, R., Anderson, T., and R. Gopal, "Forwarding 5923 and Control Element Separation (ForCES) Framework", RFC 3746, 5924 April 2004. 5926 [6] Chan, K., Sahita, R., Hahn, S., and K. McCloghrie, 5927 "Differentiated Services Quality of Service Policy Information 5928 Base", RFC 3317, March 2003. 5930 [7] Sahita, R., Hahn, S., Chan, K., and K. McCloghrie, "Framework 5931 Policy Information Base", RFC 3318, March 2003. 5933 [8] Pras, A. and J. Schoenwaelder, "On the Difference between 5934 Information Models and Data Models", RFC 3444, January 2003. 5936 [9] Hollenbeck, S., Rose, M., and L. Masinter, "Guidelines for the 5937 Use of Extensible Markup Language (XML) within IETF Protocols", 5938 BCP 70, RFC 3470, January 2003. 5940 [10] Thompson, H., Beech, D., Maloney, M., and N. Mendelsohn, "XML 5941 Schema Part 1: Structures", W3C REC-xmlschema-1, 5942 http://www.w3.org/TR/ xmlschema-1/, May 2001. 5944 [11] Biron, P. and A. Malhotra, "XML Schema Part 2: Datatypes", 5945 W3C REC-xmlschema-2, http://www.w3.org/TR /xmlschema-2/, 5946 May 2001. 5948 [12] Davis, M. and M. Suignard, "UNICODE Security Considerations", 5949 http://www.unicode.org/ reports/tr36/tr36-3.html, July 2005. 5951 Authors' Addresses 5953 Joel Halpern 5954 Self 5955 P.O. Box 6049 5956 Leesburg,, VA 20178 5958 Phone: +1 703 371 3043 5959 Email: jmh@joelhalpern.com 5961 Jamal Hadi Salim 5962 Znyx Networks 5963 Ottawa, Ontario 5964 Canada 5966 Email: hadi@znyx.com 5968 Full Copyright Statement 5970 Copyright (C) The IETF Trust (2008). 5972 This document is subject to the rights, licenses and restrictions 5973 contained in BCP 78, and except as set forth therein, the authors 5974 retain all their rights. 5976 This document and the information contained herein are provided on an 5977 "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS 5978 OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY, THE IETF TRUST AND 5979 THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS 5980 OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF 5981 THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED 5982 WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. 5984 Intellectual Property 5986 The IETF takes no position regarding the validity or scope of any 5987 Intellectual Property Rights or other rights that might be claimed to 5988 pertain to the implementation or use of the technology described in 5989 this document or the extent to which any license under such rights 5990 might or might not be available; nor does it represent that it has 5991 made any independent effort to identify any such rights. Information 5992 on the procedures with respect to rights in RFC documents can be 5993 found in BCP 78 and BCP 79. 5995 Copies of IPR disclosures made to the IETF Secretariat and any 5996 assurances of licenses to be made available, or the result of an 5997 attempt made to obtain a general license or permission for the use of 5998 such proprietary rights by implementers or users of this 5999 specification can be obtained from the IETF on-line IPR repository at 6000 http://www.ietf.org/ipr. 6002 The IETF invites any interested party to bring to its attention any 6003 copyrights, patents or patent applications, or other proprietary 6004 rights that may cover technology that may be required to implement 6005 this standard. Please address the information to the IETF at 6006 ietf-ipr@ietf.org.