idnits 2.17.00 (12 Aug 2021) /tmp/idnits28107/draft-ietf-nsis-nslp-natfw-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 19. -- Found old boilerplate from RFC 3978, Section 5.5, updated by RFC 4748 on line 3682. -- Found old boilerplate from RFC 3979, Section 5, paragraph 1 on line 3693. -- Found old boilerplate from RFC 3979, Section 5, paragraph 2 on line 3700. -- Found old boilerplate from RFC 3979, Section 5, paragraph 3 on line 3706. 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 : ---------------------------------------------------------------------------- == There are 1 instance of lines with private range IPv4 addresses in the document. If these are generic example addresses, they should be changed to use any of the ranges defined in RFC 6890 (or successor): 192.0.2.x, 198.51.100.x or 203.0.113.x. -- The document has examples using IPv4 documentation addresses according to RFC6890, but does not use any IPv6 documentation addresses. Maybe there should be IPv6 examples, too? Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust Copyright Line does not match the current year == Using lowercase 'not' together with uppercase 'MUST', 'SHALL', 'SHOULD', or 'RECOMMENDED' is not an accepted usage according to RFC 2119. Please use uppercase 'NOT' together with RFC 2119 keywords (if that is what you mean). Found 'MUST not' in this paragraph: o NSLP forwarder being either edge-NAT or edge-firewall: When the NF accepts a EXT_PROXY message, it generates a successful RESPONSE message as if it were the NR and additionally, it generates a CREATE message as defined in Section 3.7.1 and includes a NATFW_NONCE object having the same value as of the received NATFW_NONCE object. The NF MUST not generate a CREATE-PROXY message. The NF MUST refresh the CREATE message signaling session only if a EXT-PROXY refresh message has been received first. This also includes tearing down signaling sessions, i.e., the NF must teardown the CREATE signaling session only if a EXT-PROXY message with lifetime set to 0 has been received first. -- 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 (July 9, 2007) is 5429 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) == Outdated reference: draft-ietf-nsis-ntlp has been published as RFC 5971 ** Downref: Normative reference to an Experimental draft: draft-ietf-nsis-ntlp (ref. '2') == Outdated reference: draft-ietf-nsis-qos-nslp has been published as RFC 5974 -- Obsolete informational reference (is this intentional?): RFC 2434 (ref. '13') (Obsoleted by RFC 5226) -- Obsolete informational reference (is this intentional?): RFC 3852 (ref. '14') (Obsoleted by RFC 5652) -- Obsolete informational reference (is this intentional?): RFC 3489 (ref. '15') (Obsoleted by RFC 5389) == Outdated reference: draft-ietf-dime-diameter-qos has been published as RFC 5866 == Outdated reference: A later version (-04) exists of draft-manner-nsis-nslp-auth-02 Summary: 2 errors (**), 0 flaws (~~), 7 warnings (==), 11 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 NSIS Working Group M. Stiemerling 3 Internet-Draft NEC 4 Intended status: Standards Track H. Tschofenig 5 Expires: January 10, 2008 NSN 6 C. Aoun 7 E. Davies 8 Folly Consulting 9 July 9, 2007 11 NAT/Firewall NSIS Signaling Layer Protocol (NSLP) 12 draft-ietf-nsis-nslp-natfw-15.txt 14 Status of this Memo 16 By submitting this Internet-Draft, each author represents that any 17 applicable patent or other IPR claims of which he or she is aware 18 have been or will be disclosed, and any of which he or she becomes 19 aware will be disclosed, in accordance with Section 6 of BCP 79. 21 Internet-Drafts are working documents of the Internet Engineering 22 Task Force (IETF), its areas, and its working groups. Note that 23 other groups may also distribute working documents as Internet- 24 Drafts. 26 Internet-Drafts are draft documents valid for a maximum of six months 27 and may be updated, replaced, or obsoleted by other documents at any 28 time. It is inappropriate to use Internet-Drafts as reference 29 material or to cite them other than as "work in progress." 31 The list of current Internet-Drafts can be accessed at 32 http://www.ietf.org/ietf/1id-abstracts.txt. 34 The list of Internet-Draft Shadow Directories can be accessed at 35 http://www.ietf.org/shadow.html. 37 This Internet-Draft will expire on January 10, 2008. 39 Copyright Notice 41 Copyright (C) The IETF Trust (2007). 43 Abstract 45 This memo defines the NSIS Signaling Layer Protocol (NSLP) for 46 Network Address Translators (NATs) and firewalls. This NSLP allows 47 hosts to signal on the data path for NATs and firewalls to be 48 configured according to the needs of the application data flows. It 49 enables hosts behind NATs to obtain a public reachable address and 50 hosts behind firewalls to receive data traffic. The overall 51 architecture is given by the framework and requirements defined by 52 the Next Steps in Signaling (NSIS) working group. The network 53 scenarios, the protocol itself, and examples for path-coupled 54 signaling are given in this memo. 56 Table of Contents 58 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 5 59 1.1. Terminology and Abbreviations . . . . . . . . . . . . . . 7 60 1.2. Middleboxes . . . . . . . . . . . . . . . . . . . . . . . 10 61 1.3. General Scenario for NATFW Traversal . . . . . . . . . . . 11 63 2. Network Deployment Scenarios using the NATFW NSLP . . . . . . 13 64 2.1. Firewall Traversal . . . . . . . . . . . . . . . . . . . . 13 65 2.2. NAT with two private Networks . . . . . . . . . . . . . . 14 66 2.3. NAT with Private Network on Sender Side . . . . . . . . . 15 67 2.4. NAT with Private Network on Receiver Side Scenario . . . . 15 68 2.5. Both End Hosts behind twice-NATs . . . . . . . . . . . . . 16 69 2.6. Both End Hosts Behind Same NAT . . . . . . . . . . . . . . 17 70 2.7. Multihomed Network with NAT . . . . . . . . . . . . . . . 18 71 2.8. Multihomed Network with Firewall . . . . . . . . . . . . . 19 73 3. Protocol Description . . . . . . . . . . . . . . . . . . . . . 20 74 3.1. Policy Rules . . . . . . . . . . . . . . . . . . . . . . . 20 75 3.2. Basic Protocol Overview . . . . . . . . . . . . . . . . . 21 76 3.2.1. Message Types . . . . . . . . . . . . . . . . . . . . 25 77 3.2.2. Classification of RESPONSE Messages . . . . . . . . . 25 78 3.2.3. NATFW NSLP Signaling Sessions . . . . . . . . . . . . 26 79 3.3. Basic Message Processing . . . . . . . . . . . . . . . . . 27 80 3.4. Calculation of Signaling Session Lifetime . . . . . . . . 27 81 3.5. Message Sequencing . . . . . . . . . . . . . . . . . . . . 30 82 3.6. Authentication, Authorization, and Policy Decisions . . . 31 83 3.7. Protocol Operations . . . . . . . . . . . . . . . . . . . 32 84 3.7.1. Creating Signaling Sessions . . . . . . . . . . . . . 32 85 3.7.2. Reserving External Addresses . . . . . . . . . . . . . 35 86 3.7.3. NATFW NSLP Signaling Session Refresh . . . . . . . . . 42 87 3.7.4. Deleting Signaling Sessions . . . . . . . . . . . . . 43 88 3.7.5. Reporting Asynchronous Events . . . . . . . . . . . . 44 89 3.7.6. Proxy Mode of Operation . . . . . . . . . . . . . . . 46 91 3.8. De-Multiplexing at NATs . . . . . . . . . . . . . . . . . 49 92 3.9. Reacting to Route Changes . . . . . . . . . . . . . . . . 51 93 3.10. Updating Policy Rules . . . . . . . . . . . . . . . . . . 51 95 4. NATFW NSLP Message Components . . . . . . . . . . . . . . . . 53 96 4.1. NSLP Header . . . . . . . . . . . . . . . . . . . . . . . 53 97 4.2. NSLP Objects . . . . . . . . . . . . . . . . . . . . . . . 54 98 4.2.1. Signaling Session Lifetime Object . . . . . . . . . . 55 99 4.2.2. External Address Object . . . . . . . . . . . . . . . 55 100 4.2.3. Extended Flow Information Object . . . . . . . . . . . 56 101 4.2.4. Information Code Object . . . . . . . . . . . . . . . 57 102 4.2.5. Nonce Object . . . . . . . . . . . . . . . . . . . . . 60 103 4.2.6. Message Sequence Number Object . . . . . . . . . . . . 60 104 4.2.7. Data Terminal Information Object . . . . . . . . . . . 61 105 4.2.8. ICMP Types Object . . . . . . . . . . . . . . . . . . 62 106 4.3. Message Formats . . . . . . . . . . . . . . . . . . . . . 63 107 4.3.1. CREATE . . . . . . . . . . . . . . . . . . . . . . . . 64 108 4.3.2. EXTERNAL (EXT) . . . . . . . . . . . . . . . . . . . . 64 109 4.3.3. RESPONSE . . . . . . . . . . . . . . . . . . . . . . . 65 110 4.3.4. NOTIFY . . . . . . . . . . . . . . . . . . . . . . . . 65 112 5. Security Considerations . . . . . . . . . . . . . . . . . . . 67 113 5.1. Authorization Framework . . . . . . . . . . . . . . . . . 67 114 5.1.1. Peer-to-Peer Relationship . . . . . . . . . . . . . . 67 115 5.1.2. Intra-Domain Relationship . . . . . . . . . . . . . . 68 116 5.1.3. End-to-Middle Relationship . . . . . . . . . . . . . . 69 117 5.2. Security Framework for the NAT/Firewall NSLP . . . . . . . 70 118 5.2.1. Security Protection between neighboring NATFW NSLP 119 Nodes . . . . . . . . . . . . . . . . . . . . . . . . 70 120 5.2.2. Security Protection between non-neighboring NATFW 121 NSLP Nodes . . . . . . . . . . . . . . . . . . . . . . 71 123 6. IAB Considerations on UNSAF . . . . . . . . . . . . . . . . . 73 125 7. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 74 127 8. Open Issues . . . . . . . . . . . . . . . . . . . . . . . . . 76 129 9. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 77 131 10. References . . . . . . . . . . . . . . . . . . . . . . . . . . 78 132 10.1. Normative References . . . . . . . . . . . . . . . . . . . 78 133 10.2. Informative References . . . . . . . . . . . . . . . . . . 78 135 Appendix A. Selecting Signaling Destination Addresses for EXT . . 80 137 Appendix B. Applicability Statement on Data Receivers behind 138 Firewalls . . . . . . . . . . . . . . . . . . . . . . 81 140 Appendix C. Firewall and NAT Resources . . . . . . . . . . . . . 83 141 C.1. Wildcarding of Policy Rules . . . . . . . . . . . . . . . 83 142 C.2. Mapping to Firewall Rules . . . . . . . . . . . . . . . . 83 143 C.3. Mapping to NAT Bindings . . . . . . . . . . . . . . . . . 84 144 C.4. NSLP Handling of Twice-NAT . . . . . . . . . . . . . . . . 84 146 Appendix D. Assigned Numbers for Testing . . . . . . . . . . . . 86 148 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 87 149 Intellectual Property and Copyright Statements . . . . . . . . . . 88 151 1. Introduction 153 Firewalls and Network Address Translators (NAT) have both been used 154 throughout the Internet for many years, and they will remain present 155 for the foreseeable future. Firewalls are used to protect networks 156 against certain types of attacks from internal networks and the 157 Internet, whereas NATs provide a virtual extension of the IP address 158 space. Both types of devices may be obstacles to some applications, 159 since they only allow traffic created by a limited set of 160 applications to traverse them, typically those that use protocols 161 with relatively predetermined and static properties (e.g., most HTTP 162 traffic, and other client/server applications). Other applications, 163 such as IP telephony and most other peer-to-peer applications, which 164 have more dynamic properties, create traffic that is unable to 165 traverse NATs and firewalls unassisted. In practice, the traffic of 166 many applications cannot traverse autonomous firewalls or NATs, even 167 when they have additional functionality which attempts to restore the 168 transparency of the network. 170 Several solutions to enable applications to traverse such entities 171 have been proposed and are currently in use. Typically, application 172 level gateways (ALG) have been integrated with the firewall or NAT to 173 configure the firewall or NAT dynamically. Another approach is 174 middlebox communication (MIDCOM). In this approach, ALGs external to 175 the firewall or NAT configure the corresponding entity via the MIDCOM 176 protocol [7]. Several other work-around solutions are available, 177 such as STUN [15]. However, all of these approaches introduce other 178 problems that are generally hard to solve, such as dependencies on 179 the type of NAT implementation (full-cone, symmetric, etc), or 180 dependencies on certain network topologies. 182 NAT and firewall (NATFW) signaling shares a property with Quality of 183 Service (QoS) signaling. The signaling of both must reach any device 184 on the data path that is involved in, respectively, NATFW or QoS 185 treatment of data packets. This means, that for both, NATFW and QoS, 186 it is convenient if signaling travels path-coupled, meaning that the 187 signaling messages follow exactly the same path that the data packets 188 take. RSVP [11] is an example of a current QoS signaling protocol 189 that is path-coupled. [21] proposes the use of RSVP as firewall 190 signaling protocol but does not include NATs. 192 This memo defines a path-coupled signaling protocol for NAT and 193 firewall configuration within the framework of NSIS, called the NATFW 194 NSIS Signaling Layer Protocol (NSLP). The general requirements for 195 NSIS are defined in [5] and the general framework of NSIS is outlined 196 in [4]. It introduces the split between an NSIS transport layer and 197 an NSIS signaling layer. The transport of NSLP messages is handled 198 by an NSIS Network Transport Layer Protocol (NTLP, with General 199 Internet Signaling Transport (GIST) [2] being the implementation of 200 the abstract NTLP). The signaling logic for QoS and NATFW signaling 201 is implemented in the different NSLPs. The QoS NSLP is defined in 202 [6]. 204 The NATFW NSLP is designed to request the dynamic configuration of 205 NATs and/or firewalls along the data path. Dynamic configuration 206 includes enabling data flows to traverse these devices without being 207 obstructed, as well as blocking of particular data flows at inbound 208 firewalls. Enabling data flows requires the loading of firewall 209 rules with an action that allows the data flow packets to be 210 forwarded and creating NAT bindings. Blocking of data flows requires 211 the loading of firewalls rules with an action that will deny 212 forwarding of the data flow packets. A simplified example for 213 enabling data flows: A source host sends a NATFW NSLP signaling 214 message towards its data destination. This message follows the data 215 path. Every NATFW NSLP-enabled NAT/firewall along the data path 216 intercepts these messages, processes them, and configures itself 217 accordingly. Thereafter, the actual data flow can traverse all these 218 configured firewalls/NATs. 220 It is necessary to distinguish between two different basic scenarios 221 when operating the NATFW NSLP, independent of the type of the 222 middleboxes to be configured. 224 1. Both, data sender and data receiver, are NSIS NATFW NSLP aware. 225 This includes the cases where the data sender is logically 226 decomposed from the NSIS initiator or the data receiver logically 227 decomposed from the NSIS receiver, but both sides support NSIS. 228 This scenario assumes deployment of NSIS all over the Internet, 229 or at least at all NATs and firewalls. This scenario is used as 230 base assumption, if not otherwise noted. 232 2. Only one end host or region of the network is NSIS NATFW NSLP 233 aware, either data receiver or data sender. This scenario is 234 referred to as proxy mode. 236 The NATFW NSLP has two basic signaling messages which are sufficient 237 to cope with the various possible scenarios likely to be encountered 238 before and after widespread deployment of NSIS: 240 CREATE message: The basic message for configuring a path outbound 241 from a data sender to a data receiver. 243 EXTERNAL (EXT) message: Used to locate inbound NATs/firewalls and 244 prime them to expect outbound signaling and at NATs to pre- 245 allocate a public address. This is used for data receivers behind 246 these devices to enable their reachability. 248 CREATE and EXT messages are sent by the NSIS initiator (NI) towards 249 the NSIS responder (NR). Both type of messages are acknowledged by a 250 subsequent RESPONSE message. This RESPONSE message is generated by 251 the NR if the requested configuration can be established, otherwise 252 the NR or any of the NSIS forwarders (NFs) can also generate such a 253 message if an error occurs. NFs and the NR can also generate 254 asynchronous messages to notify the NI, the so called NOTIFY 255 messages. 257 If the data receiver resides in a private addressing realm or 258 firewall, and needs to preconfigure the edge-NAT/edge-firewall to 259 provide a (publicly) reachable address for use by the data sender, a 260 combination of EXTERNAL and CREATE messages is used. 262 During the introduction of NSIS, it is likely that one or other of 263 the data sender and receiver will not be NSIS aware. In these cases, 264 the NATFW NSLP can utilize NSIS aware middleboxes on the path between 265 the data sender and data receiver to provide proxy NATFW NSLP 266 services (i.e., the proxy mode operation). Typically, these boxes 267 will be at the boundaries of the realms in which the end hosts are 268 located. 270 The CREATE and EXT messages create NATFW NSLP and NTLP state in NSIS 271 entities. NTLP state allows signaling messages to travel in the 272 forward (outbound) and the reverse (inbound) direction along the path 273 between a NAT/firewall NSLP sender and a corresponding receiver. 274 This state is managed using a soft-state mechanism, i.e., it expires 275 unless it is refreshed from time to time. The NAT bindings and 276 firewall rules being installed during the state setup are bound to 277 the particular signaling session. However, the exact local 278 implementation of the NAT bindings and firewall rules are NAT/ 279 firewall specific. 281 This memo is structured as follows. Section 2 describes the network 282 environment for NATFW NSLP signaling. Section 3 defines the NATFW 283 signaling protocol and Section 4 defines the message components and 284 the overall messages used in the protocol. The remaining parts of 285 the main body of the document, covers security considerations 286 Section 5, IAB considerations on UNilateral Self-Address Fixing 287 (UNSAF) [12] in Section 6 and IANA considerations in Section 7. 288 Please note that readers familiar with firewalls and NATs and their 289 possible location within networks can safely skip Section 2. 291 1.1. Terminology and Abbreviations 293 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 294 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 295 document are to be interpreted as described in [1]. 297 This document uses a number of terms defined in [5] and [4]. The 298 following additional terms are used: 300 o Policy rule: A policy rule is "a basic building block of a policy- 301 based system. It is the binding of a set of actions to a set of 302 conditions - where the conditions are evaluated to determine 303 whether the actions are performed" [16]. In the context of NSIS 304 NATFW NSLP, the conditions are the specification of a set of 305 packets to which the rule is applied. The set of actions always 306 contains just a single element per rule, and is limited to either 307 action "deny" or action "allow". 309 o Reserved policy rule: A policy rule stored at NATs or firewalls 310 for activation by a later, different signaling exchange. This 311 type of policy rule is kept in the NATFW NSLP and is not loaded 312 into the firewall or NAT engine, i.e., it does not affect the data 313 flow handling. 315 o Installed policy rule: A policy rule in operation at NATs or 316 firewalls. This type of rule is kept in the NATFW NSLP and is 317 loaded into the firewall or NAT engine, i.e., it is affecting the 318 data flow. 320 o Remembered policy rule: A policy rule stored at NATs and firewalls 321 for immediate use, as soon as the signaling exchange is 322 successfully completed. 324 o Firewall: A packet filtering device that matches packets against a 325 set of policy rules and applies the actions. In the context of 326 NSIS NATFW NSLP we refer to this device as a firewall. 328 o Network Address Translator: Network Address Translation is a 329 method by which IP addresses are mapped from one IP address realm 330 to another, in an attempt to provide transparent routing between 331 hosts (see [9]). Network Address Translators are devices that 332 perform this work by modifying packets passing through them. 334 o Middlebox: "A middlebox is defined as any intermediate device 335 performing functions other than the normal, standard functions of 336 an IP router on the datagram path between a source host and a 337 destination host" [10]. In the context of this document, the term 338 middlebox refers to firewalls and NATs only. Other types of 339 middlebox are outside of the scope of this document. 341 o Data Receiver (DR): The node in the network that is receiving the 342 data packets of a flow. 344 o Data Sender (DS): The node in the network that is sending the data 345 packets of a flow. 347 o NATFW NSLP peer or peer: An NSIS NATFW NSLP node with which an 348 NSIS adjacency has been created as defined in [2]. 350 o NATFW NSLP signaling session or signaling session: A signaling 351 session defines an association between the NI, NFs, and the NR 352 related to a data flow. All the NATFW NSLP peers on the path, 353 including the NI and the NR, use the same identifier to refer to 354 the state stored for the association. The same NI and NR may have 355 more than one signaling session active at any time. The state for 356 NATFW NSLP consists of NSLP state and associated policy rules at a 357 middlebox. 359 o Edge-NAT: An edge-NAT is a NAT device with a globally routable IP 360 address which is reachable from the public Internet. 362 o Edge-firewall: An edge-firewall is a firewall device that is 363 located on the border line of an administrative domain. 365 o Public Network: "A Global or Public Network is an address realm 366 with unique network addresses assigned by Internet Assigned 367 Numbers Authority (IANA) or an equivalent address registry. This 368 network is also referred as external network during NAT 369 discussions" [9]. 371 o Private/Local Network: "A private network is an address realm 372 independent of external network addresses. Private network may 373 also be referred alternately as Local Network. Transparent 374 routing between hosts in private realm and external realm is 375 facilitated by a NAT router" [9]. 377 o Public/Global IP address: An IP address located in the public 378 network according to Section 2.7 of [9]. 380 o Private/Local IP address: An IP address located in the private 381 network according to Section 2.8 of [9]. 383 o Signaling Destination Address (SDA): An IP address generally taken 384 from the public/global IP address range, although, the SDA may in 385 certain circumstances be part of the private/local IP address 386 range. This address is used in EXT signaling message exchanges, 387 if the data receiver's IP address is unknown. 389 1.2. Middleboxes 391 The term middlebox covers a range of devices which intercept the flow 392 of packets between end hosts and perform actions other than standard 393 forwarding expected in an IP router. As such, middleboxes fall into 394 a number of categories with a wide range of functionality, not all of 395 which is pertinent to the NATFW NSLP. Middlebox categories in the 396 scope of this memo are firewalls that filter data packets against a 397 set of filter rules, and NATs that translate packet addresses from 398 one address realm to another address realm. Other categories of 399 middleboxes, such as QoS traffic shapers, are out of scope of this 400 memo. 402 The term NAT used in this document is a placeholder for a range of 403 different NAT flavors. We consider the following types of NATs: 405 o Traditional NAT (basic NAT and NAPT) 407 o Bi-directional NAT 409 o Twice-NAT 411 o Multihomed NAT 413 For definitions and a detailed discussion about the characteristics 414 of each NAT type please see [9]. 416 All types of middleboxes under consideration here, use policy rules 417 to make a decision on data packet treatment. Policy rules consist of 418 a flow identifier which selects the packets to which the policy 419 applies and an associated action; data packets matching the flow 420 identifier are subjected to the policy rule action. A typical flow 421 identifier is the 5-tuple selector which matches the following fields 422 of a packet to configured values: 424 o Source and destination IP addresses 426 o Transport protocol number 428 o Transport source and destination port numbers 430 Actions for firewalls are usually one or more of: 432 o Allow: forward data packet 434 o Deny: block data packet and discard it 435 o Other actions such as logging, diverting, duplicating, etc 437 Actions for NATs include (amongst many others): 439 o Change source IP address and transport port number to a globally 440 routeable IP address and associated port number. 442 o Change destination IP address and transport port number to a 443 private IP address and associated port number. 445 It should be noted that a middlebox may contain two logical 446 representations of the policy rule. The policy rule has a 447 representation within the NATFW NSLP, comprising the message routing 448 information (MRI) of the NTLP and NSLP information (such as the rule 449 action). The other representation is the implementation of the NATFW 450 NSLP policy rule within the NAT and firewall engine of the particular 451 device. Refer to Appendix C for further details. 453 1.3. General Scenario for NATFW Traversal 455 The purpose of NSIS NATFW signaling is to enable communication 456 between endpoints across networks, even in the presence of NAT and 457 firewall middleboxes that have not been specially engineered to 458 facilitate communication with the application protocols used. This 459 removes the need to create and maintain application layer gateways 460 for specific protocols that have been commonly used to provide 461 transparency in previous generations of NAT and firewall middleboxes. 462 It is assumed that these middleboxes will be statically configured in 463 such a way that NSIS NATFW signaling messages themselves are allowed 464 to reach the locally installed NATFW NSLP daemon. NSIS NATFW NSLP 465 signaling is used to dynamically install additional policy rules in 466 all NATFW middleboxes along the data path that will allow 467 transmission of the application data flow(s). Firewalls are 468 configured to forward data packets matching the policy rule provided 469 by the NSLP signaling. NATs are configured to translate data packets 470 matching the policy rule provided by the NSLP signaling. An 471 additional capability, that is an exception to the primary goal of 472 NSIS NATFW signaling, is that the NATFW nodes can request blocking of 473 particular data flows instead of enabling these flows at inbound 474 firewalls. 476 The basic high-level picture of NSIS usage is that end hosts are 477 located behind middleboxes, meaning that there is a middlebox on the 478 data path from the end host in a private network and the external 479 network (NATFW in Figure 1). Applications located at these end hosts 480 try to establish communication with corresponding applications on 481 other such end hosts. They trigger the NSIS entity at the local host 482 to control provisioning for middlebox traversal along the prospective 483 data path (e.g., via an API call). The NSIS entity in turn uses NSIS 484 NATFW NSLP signaling to establish policy rules along the data path, 485 allowing the data to travel from the sender to the receiver 486 unobstructed. 488 Application Application Server (0, 1, or more) Application 490 +----+ +----+ +----+ 491 | +------------------------+ +------------------------+ | 492 +-+--+ +----+ +-+--+ 493 | | 494 | NSIS Entities NSIS Entities | 495 +-+--+ +----+ +-----+ +-+--+ 496 | +--------+ +----------------------------+ +-----+ | 497 +-+--+ +-+--+ +--+--+ +-+--+ 498 | | ------ | | 499 | | //// \\\\\ | | 500 +-+--+ +-+--+ |/ | +-+--+ +-+--+ 501 | | | | | Internet | | | | | 502 | +--------+ +-----+ +----+ +-----+ | 503 +----+ +----+ |\ | +----+ +----+ 504 \\\\ ///// 505 sender NATFW (1+) ------ NATFW (1+) receiver 507 Note that 1+ refers to one or more NATFW nodes. 509 Figure 1: Generic View of NSIS with NATs and/or Firewalls 511 For end-to-end NATFW signaling, it is necessary that each firewall 512 and each NAT along the path between the data sender and the data 513 receiver implements the NSIS NATFW NSLP. There might be several NATs 514 and FWs in various possible combinations on a path between two hosts. 515 Section 2 presents a number of likely scenarios with different 516 combinations of NATs and firewalls. 518 2. Network Deployment Scenarios using the NATFW NSLP 520 This section introduces several scenarios for middlebox placement 521 within IP networks. Middleboxes are typically found at various 522 different locations, including at enterprise network borders, within 523 enterprise networks, as mobile phone network gateways, etc. Usually, 524 middleboxes are placed more towards the edge of networks than in 525 network cores. Firewalls and NATs may be found at these locations 526 either alone, or they may be combined; other categories of 527 middleboxes may also be found at such locations, possibly combined 528 with the NATs and/or firewalls. Using combined middleboxes typically 529 reduces the number of network elements needed. 531 NSIS initiators (NI) send NSIS NATFW NSLP signaling messages via the 532 regular data path to the NSIS responder (NR). On the data path, 533 NATFW NSLP signaling messages reach different NSIS nodes that 534 implement the NATFW NSLP. Each NATFW NSLP node processes the 535 signaling messages according to Section 3 and, if necessary, installs 536 policy rules for subsequent data packets. 538 Each of the following sub-sections introduces a different scenario 539 for a different set of middleboxes and their ordering within the 540 topology. It is assumed that each middlebox implements the NSIS 541 NATFW NSLP signaling protocol. 543 2.1. Firewall Traversal 545 This section describes a scenario with firewalls only; NATs are not 546 involved. Each end host is behind a firewall. The firewalls are 547 connected via the public Internet. Figure 2 shows the topology. The 548 part labeled "public" is the Internet connecting both firewalls. 550 +----+ //----\\ +----+ 551 NI -----| FW |---| |------| FW |--- NR 552 +----+ \\----// +----+ 554 private public private 556 FW: Firewall 557 NI: NSIS Initiator 558 NR: NSIS Responder 560 Figure 2: Firewall Traversal Scenario 562 Each firewall on the data path must provide traversal service for 563 NATFW NSLP in order to permit the NSIS message to reach the other end 564 host. All firewalls process NSIS signaling and establish appropriate 565 policy rules, so that the required data packet flow can traverse 566 them. 568 There are several very different ways to place firewalls in a network 569 topology. To distinguish firewalls located at network borders, such 570 as administrative domains, from others located internally, the term 571 edge-firewall is used. A similar distinction can be made for NATs, 572 with an edge-NAT fulfilling the equivalent role. 574 2.2. NAT with two private Networks 576 Figure 3 shows a scenario with NATs at both ends of the network. 577 Therefore, each application instance, the NSIS initiator and the NSIS 578 responder, are behind NATs. The outermost NAT, known as the edge-NAT 579 (MB2 and MB3), at each side is connected to the public Internet. The 580 NATs are generically labeled as MBX (for middlebox No. X), since 581 those devices certainly implement NAT functionality, but can 582 implement firewall functionality as well. 584 Only two middleboxes MB are shown in Figure 3 at each side, but in 585 general, any number of MBs on each side must be considered. 587 +----+ +----+ //----\\ +----+ +----+ 588 NI --| MB1|-----| MB2|---| |---| MB3|-----| MB4|--- NR 589 +----+ +----+ \\----// +----+ +----+ 591 private public private 593 MB: Middlebox 594 NI: NSIS Initiator 595 NR: NSIS Responder 597 Figure 3: NAT with two Private Networks Scenario 599 Signaling traffic from NI to NR has to traverse all the middleboxes 600 on the path (MB1 to MB4, in this order), and all the middleboxes must 601 be configured properly to allow NSIS signaling to traverse them. The 602 NATFW signaling must configure all middleboxes and consider any 603 address translation that will result from this configuration in 604 further signaling. The sender (NI) has to know the IP address of the 605 receiver (NR) in advance, otherwise it will not be possible to send 606 any NSIS signaling messages towards the responder. Note that this IP 607 address is not the private IP address of the responder but the NAT's 608 public IP address (here MB3's IP address). Instead a NAT binding 609 (including a public IP address) has to be previously installed on the 610 NAT MB3. This NAT binding subsequently allows packets reaching the 611 NAT to be forwarded to the receiver within the private address realm. 612 The receiver might have a number of ways to learn its public IP 613 address and port number (including the NATFW NSLP) and might need to 614 signal this information to the sender using the application level 615 signaling protocol. 617 2.3. NAT with Private Network on Sender Side 619 This scenario shows an application instance at the sending node that 620 is behind one or more NATs (shown as generic MB, see discussion in 621 Section 2.2). The receiver is located in the public Internet. 623 +----+ +----+ //----\\ 624 NI --| MB |-----| MB |---| |--- NR 625 +----+ +----+ \\----// 627 private public 629 MB: Middlebox 630 NI: NSIS Initiator 631 NR: NSIS Responder 633 Figure 4: NAT with Private Network on Sender Side Scenario 635 The traffic from NI to NR has to traverse middleboxes only on the 636 sender's side. The receiver has a public IP address. The NI sends 637 its signaling message directly to the address of the NSIS responder. 638 Middleboxes along the path intercept the signaling messages and 639 configure the policy rules accordingly. 641 The data sender does not necessarily know whether the receiver is 642 behind a NAT or not, hence, it is the receiving side that has to 643 detect whether itself is behind a NAT or not. 645 2.4. NAT with Private Network on Receiver Side Scenario 647 The application instance receiving data is behind one or more NATs 648 shown as MB (see discussion in Section 2.2). 650 //----\\ +----+ +----+ 651 NI ---| |---| MB |-----| MB |--- NR 652 \\----// +----+ +----+ 654 public private 656 MB: Middlebox 657 NI: NSIS Initiator 658 NR: NSIS Responder 660 Figure 5: NAT with Private Network on Receiver Scenario 662 Initially, the NSIS responder must determine its publicly reachable 663 IP address at the external middlebox and notify the NSIS initiator 664 about this address. One possibility is that an application level 665 protocol is used, meaning that the public IP address is signaled via 666 this protocol to the NI. Afterwards the NI can start its signaling 667 towards the NR and therefore establish the path via the middleboxes 668 in the receiver side private network. 670 This scenario describes the use case for the EXTERNAL message of the 671 NATFW NSLP. 673 2.5. Both End Hosts behind twice-NATs 675 This is a special case, where the main problem arises from the need 676 to detect that both end hosts are logically within the same address 677 space, but are also in two partitions of the address realm on either 678 side of a twice-NAT (see [9] for a discussion of twice-NAT 679 functionality). 681 Sender and receiver are both within a single private address realm 682 but the two partitions potentially have overlapping IP address 683 ranges. Figure 6 shows the arrangement of NATs. 685 public 687 +----+ +----+ //----\\ 688 NI --| MB |--+--| MB |---| | 689 +----+ | +----+ \\----// 690 | 691 | +----+ 692 +--| MB |------------ NR 693 +----+ 695 private 697 MB: Middlebox 698 NI: NSIS Initiator 699 NR: NSIS Responder 701 Figure 6: NAT to Public, Sender and Receiver on either side of a 702 twice-NAT Scenario 704 The middleboxes shown in Figure 6 are twice-NATs, i.e., they map IP 705 addresses and port numbers on both sides, meaning the mapping of 706 source and destination address at the private and public interfaces. 708 This scenario requires the assistance of application level entities, 709 such as a DNS server. The application level entities must handle 710 requests that are based on symbolic names, and configure the 711 middleboxes so that data packets are correctly forwarded from NI to 712 NR. The configuration of those middleboxes may require other 713 middlebox communication protocols, such as MIDCOM [7]. NSIS 714 signaling is not required in the twice-NAT only case, since 715 middleboxes of the twice-NAT type are normally configured by other 716 means. Nevertheless, NSIS signaling might be useful when there are 717 also firewalls on the path. In this case NSIS will not configure any 718 policy rule at twice-NATs, but will configure policy rules at the 719 firewalls on the path. The NSIS signaling protocol must be at least 720 robust enough to survive this scenario. This requires that twice- 721 NATs must implement the NATFW NSLP also and participate in NATFW 722 signaling sessions but they do not change the configuration of the 723 NAT, i.e., they only read the address mapping information out of the 724 NAT and translate the Message Routing Information (MRI, [2]) within 725 the NSLP and NTLP accordingly. For more information see Appendix C.4 727 2.6. Both End Hosts Behind Same NAT 729 When NSIS initiator and NSIS responder are behind the same NAT (thus 730 being in the same address realm, see Figure 7), they are most likely 731 not aware of this fact. As in Section 2.4 the NSIS responder must 732 determine its public IP address in advance and transfer it to the 733 NSIS initiator. Afterwards, the NSIS initiator can start sending the 734 signaling messages to the responder's public IP address. During this 735 process, a public IP address will be allocated for the NSIS initiator 736 at the same middlebox as for the responder. Now, the NSIS signaling 737 and the subsequent data packets will traverse the NAT twice: from 738 initiator to public IP address of responder (first time) and from 739 public IP address of responder to responder (second time). 741 NI public 742 \ +----+ //----\\ 743 +-| MB |----| | 744 / +----+ \\----// 745 NR 746 private 748 MB: Middlebox 749 NI: NSIS Initiator 750 NR: NSIS Responder 752 Figure 7: NAT to Public, Both Hosts Behind Same NAT 754 2.7. Multihomed Network with NAT 756 The previous sub-sections sketched network topologies where several 757 NATs and/or firewalls are ordered sequentially on the path. This 758 section describes a multihomed scenario with two NATs placed on 759 alternative paths to the public network. 761 +----+ //---\\ 762 NI -------| MB |---| | 763 \ +----+ \\-+-// 764 \ | 765 \ +----- NR 766 \ | 767 \ +----+ //-+-\\ 768 --| MB |---| | 769 +----+ \\---// 771 private public 773 MB: Middlebox 774 NI: NSIS Initiator 775 NR: NSIS Responder 776 Figure 8: Multihomed Network with Two NATs 778 Depending on the destination, either one or the other middlebox is 779 used for the data flow. Which middlebox is used, depends on local 780 policy or routing decisions. NATFW NSLP must be able to handle this 781 situation properly, see Section 3.7.2 for an extended discussion of 782 this topic with respect to NATs. 784 2.8. Multihomed Network with Firewall 786 This section describes a multihomed scenario with two firewalls 787 placed on alternative paths to the public network (Figure 9). The 788 routing in the private and public network decides which firewall is 789 being taken for data flows. Depending on the data flow's direction, 790 either outbound or inbound, a different firewall could be traversed. 791 This is a challenge for the EXT message of the NATFW NSLP where the 792 NSIS responder is located behind these firewalls within the private 793 network. The EXT message is used to block a particular data flow on 794 an inbound firewall. NSIS must route the EXT message inbound from NR 795 to NI probably without knowing which path the data traffic will take 796 from NI to NR (see also Appendix B). 798 +----+ 799 NR -------| MB |\ 800 \ +----+ \ //---\\ 801 \ -| |-- NI 802 \ \\---// 803 \ +----+ | 804 --| MB |-------+ 805 +----+ 806 private 808 private public 810 MB: Middlebox 811 NI: NSIS Initiator 812 NR: NSIS Responder 814 Figure 9: Multihomed Network with two Firewalls 816 3. Protocol Description 818 This section defines messages, objects, and protocol semantics for 819 the NATFW NSLP. 821 3.1. Policy Rules 823 Policy rules, bound to a NATFW NSLP signaling session, are the 824 building blocks of middlebox devices considered in the NATFW NSLP. 825 For firewalls the policy rule usually consists of a 5-tuple, source/ 826 destination addresses, transport protocol, and source/destination 827 port numbers, plus an action, such as allow or deny. For NATs the 828 policy rule consists of the action 'translate this address' and 829 further mapping information, that might be, in the simplest case, 830 internal IP address and external IP address. 832 The NATFW NSLP carries, in conjunction with the NTLP's Message 833 Routing Information (MRI), the policy rules to be installed at NATFW 834 peers. This policy rule is an abstraction with respect to the real 835 policy rule to be installed at the respective firewall or NAT. It 836 conveys the initiator's request and must be mapped to the possible 837 configuration on the particular used NAT and/or firewall in use. For 838 pure firewalls one or more filter rules must be created and for pure 839 NATs one or more NAT bindings must be created. In mixed firewall and 840 NAT boxes, the policy rule must be mapped to filter rules and 841 bindings observing the ordering of the firewall and NAT engine. 842 Depending on the ordering, NAT before firewall or vice versa, the 843 firewall rules must carry public or private IP addresses. However, 844 the exact mapping depends on the implementation of the firewall or 845 NAT which is different for each vendor. 847 The policy rule at the NATFW NSLP level comprises the message routing 848 information (MRI) part, carried in the NTLP, and the information 849 available in the NATFW NSLP. The information provided by the NSLP is 850 stored in the 'extend flow information' (NATFW_EFI) and 'data 851 terminal information' (NATFW_DTINFO) objects, and the message type. 852 Additional information, such as the external IP address and port 853 number, stored in the NAT or firewall, will be used as well. The MRI 854 carries the filter part of the NAT/firewall-level policy rule that is 855 to be installed. 857 The NATFW NSLP specifies two actions for the policy rules: deny and 858 allow. A policy rule with action set to deny will result in all 859 packets matching this rule to be dropped. A policy rule with action 860 set to allow will result in all packets matching this rule to be 861 forwarded. 863 3.2. Basic Protocol Overview 865 The NSIS NATFW NSLP is carried over the General Internet Signaling 866 Transport (GIST, the implementation of the NTLP) defined in [2]. 867 NATFW NSLP messages are initiated by the NSIS initiator (NI), handled 868 by NSIS forwarders (NF) and received by the NSIS responder (NR). It 869 is required that at least NI and NR implement this NSLP, intermediate 870 NFs only implement this NSLP when they provide relevant middlebox 871 functions. NSIS forwarders that do not have any NATFW NSLP functions 872 just forward these packets as they have no interest in them. 874 A Data Sender (DS), intending to send data to a Data Receiver (DR) 875 has to start NATFW NSLP signaling. This causes the NI associated 876 with the data sender (DS) to launch NSLP signaling towards the 877 address of data receiver (DR) (see Figure 10). Although it is 878 expected that the DS and the NATFW NSLP NI will usually reside on the 879 same host, this specification does not rule out scenarios where the 880 DS and NI reside on different hosts, the so-called proxy mode (see 881 Section 3.7.6.) 883 +-------+ +-------+ +-------+ +-------+ 884 | DS/NI |<~~~| MB1/ |<~~~| MB2/ |<~~~| DR/NR | 885 | |--->| NF1 |--->| NF2 |--->| | 886 +-------+ +-------+ +-------+ +-------+ 888 ========================================> 889 Data Traffic Direction (outbound) 891 ---> : NATFW NSLP request signaling 892 ~~~> : NATFW NSLP response signaling 893 DS/NI : Data sender and NSIS initiator 894 DR/NR : Data receiver and NSIS responder 895 MB1 : Middlebox 1 and NSIS forwarder 1 896 MB2 : Middlebox 2 and NSIS forwarder 2 898 Figure 10: General NSIS signaling 900 The following list shows the normal sequence of NSLP events without 901 detailing the interaction with the NTLP and the interactions on the 902 the NTLP level. 904 o NSIS initiators generate NATFW NSLP CREATE/EXT messages and send 905 these towards the NSIS responder. This CREATE/EXT message is the 906 initial message which creates a new NATFW NSLP signaling session. 907 The NI and the NR will most likely already share an application 908 session before they start the NATFW NSLP signaling session. Note 909 the difference between both sessions. 911 o NSLP CREATE/EXT messages are processed each time a NF with NATFW 912 NSLP support is traversed. Each NF that is intercepting a CREATE/ 913 EXT message and is accepting it for further treatment is joining 914 the particular NATFW NSLP signaling session. These nodes process 915 the message, check local policies for authorization and 916 authentication, possibly create policy rules, and forward the 917 signaling message to the next NSIS node. The request message is 918 forwarded until it reaches the NSIS responder. 920 o NSIS responders will check received messages and process them if 921 applicable. NSIS responders generate RESPONSE messages and send 922 them hop-by-hop back to the NI via the same chain of NFs 923 (traversal of the same NF chain is guaranteed through the 924 established reverse message routing state in the NTLP). The NR is 925 also joining the NATFW NSLP signaling session if the CREATE/EXT 926 message is accepted. 928 o The RESPONSE message is processed at each NF that has been 929 included in the prior NATFW NSLP signaling session setup. 931 o If the NI has received a successful RESPONSE message and if the 932 signaling NATFW NSLP session started with a CREATE message, the 933 data sender can start sending its data flow to the data receiver. 934 If the Ni has received a successful RESPONSE message and if the 935 signaling NATFW NSLP session started with a EXT message, the data 936 receiver is ready to receive further CREATE messages. 938 Because NATFW NSLP signaling follows the data path from DS to DR, 939 this immediately enables communication between both hosts for 940 scenarios with only firewalls on the data path or NATs on the sender 941 side. For scenarios with NATs on the receiver side certain problems 942 arise, as described in Section 2.4. 944 When the NR and the NI are located in different address realms and 945 the NR is located behind a NAT, the NI cannot signal to the NR 946 address directly. The DR/NR are not reachable from the NIs using the 947 private address of the NR and thus NATFW signaling messages cannot be 948 sent to the NR/DR's address. Therefore, the NR must first obtain a 949 NAT binding that provides an address that is reachable for the NI. 950 Once the NR has acquired a public IP address, it forwards this 951 information to the DS via a separate protocol. This application 952 layer signaling, which is out of scope of the NATFW NSLP, may involve 953 third parties that assist in exchanging these messages. 955 The same holds partially true for NRs located behind firewalls that 956 block all traffic by default. In this case, NR must tell its inbound 957 firewalls of inbound NATFW NSLP signaling and corresponding data 958 traffic. Once the NR has informed the inbound firewalls, it can 959 start its application level signaling to initiate communication with 960 the NI. This application layer signaling, which is out of scope of 961 the NATFW NSLP, may involve third parties that assist in exchanging 962 these messages. This mechanism can be used by machines hosting 963 services behind firewalls as well. In this case, the NR informs the 964 inbound firewalls as described, but does not need to communicate this 965 to the NIs. 967 NATFW NSLP signaling supports this scenario by using the EXT message 969 1. The DR acquires a public address by signaling on the reverse path 970 (DR towards DS) and thus making itself available to other hosts. 971 This process of acquiring public addresses is called reservation. 972 During this process the DR reserves publicly reachable addresses 973 and ports suitable for further usage in application level 974 signaling and the publicly reachable address for further NATFW 975 NSLP signaling. However, the data traffic will not be allowed to 976 use this address/port initially (see next point). In the process 977 of reservation the DR becomes the NI for the messages necessary 978 to obtain the publicly reachable IP address, i.e., the NI for 979 this specific NATFW NSLP signaling session. 981 2. Now on the side of DS, the NI creates a new NATFW NSLP signaling 982 session and signals directly to the public IP address of DR. 983 This public IP address is used as NR's address, as the NI would 984 do if there is no NAT in between, and creates policy rules at 985 middleboxes. Note, that the reservation will only allow 986 forwarding of signaling messages, but not data flow packets. 987 Policy rules allowing forwarding of data flow packets set up by 988 the prior EXT message signaling will be activated when the 989 signaling from NI towards NR is confirmed with a positive 990 RESPONSE message. The EXTERNAL (EXT) message is described 991 inSection 3.7.2. 993 +-------+ +-------+ +-------+ +-------+ 994 | DS/NI |<~~~| MB1/ |<~~~| NR | | DR | 995 | |--->| NF1 |--->| | | | 996 +-------+ +-------+ +-------+ +-------+ 998 ========================================> 999 Data Traffic Direction (outbound) 1001 ---> : NATFW NSLP request signaling 1002 ~~~> : NATFW NSLP response signaling 1003 DS/NI : Data sender and NSIS initiator 1004 DR/NR : Data receiver and NSIS responder 1005 MB1 : Middlebox 1 and NSIS forwarder 1 1006 MB2 : Middlebox 2 and NSIS forwarder 2 1008 Figure 11: A NSIS proxy mode signaling 1010 The above usage assumes that both ends of a communication support 1011 NSIS, but fails when NSIS is only deployed at one end of the path. 1012 In this case only one of the receiving or sending side is NSIS aware 1013 and not both at the same time. NATFW NSLP supports this scenario 1014 (i.e., the DR does not support NSIS) by using a proxy mode, as 1015 described in Section 3.7.6; the proxy mode operation also supports 1016 scenarios with a data sender that does not support NSIS, i.e. the 1017 data receiver must act to enable data flows towards itself. 1019 The basic functionality of the NATFW NSLP provides for opening 1020 firewall pin holes and creating NAT bindings to enable data flows to 1021 traverse these devices. Firewalls are normally expected to work on a 1022 'deny-all' policy, meaning that traffic not explicitly matching any 1023 firewall filter rule will be blocked. Similarly, the normal behavior 1024 of NATs is to block all traffic that does not match any already 1025 configured/installed binding or NATFW NSLP session. However, some 1026 scenarios require support of firewalls having 'allow-all' policies, 1027 allowing data traffic to traverse the firewall unless it is blocked 1028 explicitly. Data receivers can utilize NATFW NSLP's EXT message with 1029 action set to 'deny' to install policy rules at inbound firewalls to 1030 block unwanted traffic. 1032 The protocol works on a soft-state basis, meaning that whatever state 1033 is installed or reserved on a middlebox will expire, and thus be de- 1034 installed or forgotten after a certain period of time. To prevent 1035 premature removal of state that is needed for ongoing communication, 1036 the NATFW NI involved will have to specifically request a NATFW NSLP 1037 signaling session extension. An explicit NATFW NSLP state deletion 1038 capability is also provided by the protocol. 1040 If the actions requested by a NATFW NSLP message cannot be carried 1041 out, NFs and the NR must return a failure, such that appropriate 1042 actions can be taken. They can do this either during a the request 1043 message handling (synchronously) by sending an error RESPONSE 1044 message, or at any time (asynchronously) by sending a notification 1045 message. 1047 The next sections define the NATFW NSLP message types and formats, 1048 protocol operations, and policy rule operations. 1050 3.2.1. Message Types 1052 The protocol uses four messages types: 1054 o CREATE: a request message used for creating, changing, refreshing, 1055 and deleting NATFW NSLP signaling sessions, i.e., open the data 1056 path from DS to DR. 1058 o EXTERNAL (EXT): a request message used for reserving, changing, 1059 refreshing, and deleting EXT NATFW NSLP signaling sessions. EXT 1060 messages are forwarded to the edge-NAT or edge-firewall and allow 1061 inbound CREATE messages to be forwarded to the NR. Additionally, 1062 EXT messages reserve an external address and, if applicable, port 1063 number at an edge-NAT. 1065 o NOTIFY: an asynchronous message used by NATFW peers to alert 1066 inbound NATFW peers about specific events (especially failures). 1068 o RESPONSE: used as a response to CREATE and EXT request messages. 1070 3.2.2. Classification of RESPONSE Messages 1072 RESPONSE messages will be generated synchronously to CREATE and EXT 1073 messages by NSIS Forwarders and Responders to report success or 1074 failure of operations or some information relating to the NATFW NSLP 1075 signaling session or a node. RESPONSE messages MUST NOT be generated 1076 for any other message, such as NOTIFY and RESPONSE. 1078 All RESPONSE messages MUST carry a NATFW_INFO object which contains a 1079 severity class code and a response code (see Section 4.2.4). This 1080 section defines terms for groups of RESPONSE messages depending on 1081 the severity class. 1083 o Successful RESPONSE: Messages carrying NATFW_INFO with severity 1084 class 'Success' (0x2). 1086 o Informational RESPONSE: Messages carrying NATFW_INFO with severity 1087 class 'Informational' (0x1) (only used with NOTIFY messages). 1089 o Error RESPONSE: Messages carrying NATFW_INFO with severity class 1090 other than 'Success' or 'Informational'. 1092 3.2.3. NATFW NSLP Signaling Sessions 1094 A NATFW NSLP signaling session defines an association between the NI, 1095 NFs, and the NR related to a data flow. This association is created 1096 when the initial CREATE or EXT message is successfully received at 1097 the NFs or the NR. There is signaling NATFW NSLP session state 1098 stored at the NTLP layer and at the NATFW NSLP level. The NATFW NSLP 1099 signaling session state for the NATFW NSLP comprises NSLP state and 1100 the associated policy rules at a middlebox. 1102 The NATFW NSLP signaling session is identified by the session ID 1103 (plus other information at the NTLP level). The session ID is 1104 generated by the NI before the initial CREATE or EXT message is sent. 1105 The value of the session ID MUST generated in a random way, i.e., the 1106 output MUST NOT be easily guessable by third parties. The session ID 1107 is not stored in any NATFW NSLP message but passed on to the NTLP. 1109 A NATFW NSLP signaling session can conceptually be in different 1110 states, implementations may use other or even more states. The 1111 signaling session can have these states at a node: 1113 o Pending: The NATFW NSLP signaling session has been created and the 1114 node is waiting for a RESPONSE message to the CREATE or EXT 1115 message. A NATFW NSLP signaling session in state 'Pending' MUST 1116 be marked as 'Dead' if no corresponding RESPONSE message has been 1117 received within the time of the locally granted NATFW NSLP 1118 signaling session lifetime of the forwarded CREATE or EXT message 1119 (as described in Section 3.4). 1121 o Established: The NATFW NSLP signaling session is established, i.e, 1122 the signaling has been successfully performed and the lifetime of 1123 NATFW NSLP signaling session is counted from now on. A NATFW NSLP 1124 signaling session in state 'Established' MUST be marked as 'Dead' 1125 if no refresh message has been received within the time of the 1126 locally granted NATFW NSLP signaling session lifetime of the 1127 RESPONSE message (as described in Section 3.4). 1129 o Dead: Either the NATFW NSLP signaling session is timed out or the 1130 node has received an error RESPONSE message for the NATFW NSLP 1131 signaling session and the NATFW NSLP signaling session can be 1132 deleted. 1134 o Transit: The node has received an asynchronous message, i.e., a 1135 NOTIFY, and can delete the NATFW NSLP signaling session if needed. 1136 When a node has received a NOTIFY message (for instance, 1137 indicating a route change) it marks it as 'Transit' and deletes 1138 this NATFW NSLP signaling session if it is unused for some time 1139 specific to the local node. This idle time does not need to be 1140 fixed, since it can depend on the node local maintenance cycle, 1141 i.e., the NATFW NSLP signaling session could be deleted if the 1142 node runs it garbage collection cycle. 1144 3.3. Basic Message Processing 1146 All NATFW messages are subject to some basic message processing when 1147 received at a node, independent of message type. Initially, the 1148 syntax of the NSLP message is checked and a RESPONSE message with an 1149 appropriate error of class 'Protocol error' (0x1) code is generated 1150 if any problem is detected. If a message is delivered to the NATFW 1151 NSLP, this implies that the NTLP layer has been able to correlate it 1152 with the SID and MRI entries in its database. There is therefore 1153 enough information to identify the source of the message and routing 1154 information to route the message back to the NI through an 1155 established chain of NTLP messaging associations. The message is not 1156 further forwarded if any error in the syntax is detected. The 1157 specific response codes stemming from the processing of objects are 1158 described in the respective object definition section (see 1159 Section 4). After passing this check, the NATFW NSLP node performs 1160 authentication/authorization related checks described in Section 3.6. 1161 Further processing is executed only if these tests have been 1162 successfully passed, otherwise the processing stops and an error 1163 RESPONSE is returned. 1165 Further message processing stops whenever an error RESPONSE message 1166 is generated, and the EXT or CREATE message is discarded. 1168 3.4. Calculation of Signaling Session Lifetime 1170 NATFW NSLP signaling sessions, and the corresponding policy rules 1171 which may have been installed, are maintained via a soft-state 1172 mechanism. Each signaling session is assigned a signaling session 1173 lifetime and the signaling session is kept alive as long as the 1174 lifetime is valid. After the expiration of the signaling session 1175 lifetime, signaling sessions and policy rules MUST be removed 1176 automatically and resources bound to them MUST be freed as well. 1177 Signaling session lifetime is handled at every NATFW NSLP node. The 1178 NSLP forwarders and NSLP responder MUST NOT trigger signaling session 1179 lifetime extension refresh messages (see Section 3.7.3): this is the 1180 task of the NSIS initiator. 1182 The NSIS initiator MUST choose a NATFW NSLP signaling session 1183 lifetime value (expressed in seconds) before sending any message, 1184 including the initial message which creates the NATFW NSLP signaling 1185 session, to other NSLP nodes. The NATFW NSLP signaling session 1186 lifetime value is calculated based on: 1188 o the number of lost refresh messages that NFs should cope with; 1190 o the end-to-end delay between the NI and NR; 1192 o network vulnerability due to NATFW NSLP signaling session 1193 hijacking ([8]), NATFW NSLP signaling session hijacking is made 1194 easier when the NI does not explicitly remove the NATFW NSLP 1195 signaling session); 1197 o the user application's data exchange duration, in terms of time 1198 and networking needs. This duration is modeled as M x R, with R 1199 the message refresh period (in seconds) and M as a multiplier for 1200 R; 1202 o the load on the signalling plane. Short lifetimes imply more 1203 frequent signaling messages. 1205 o the acceptable time for a NATFW NSLP signaling session to be 1206 present after it is no longer actually needed. For example, if 1207 the existence of the NATFW NSLP signaling session implies a 1208 monetary cost and teardown cannot be guaranteed, shorter lifetimes 1209 would be preferable. 1211 o the lease time of the NI's IP address. The chosen NATFW NSLP 1212 signaling session lifetime must be larger than the lease time, 1213 otherwise the IP address can be re-assigned to a different node. 1214 This node may receive unwanted traffic, although it never has 1215 requested a NAT/firewall configuration, which might be an issue in 1216 mobile environments. 1218 The RSVP specification [11] provides an appropriate algorithm for 1219 calculating the NATFW NSLP signaling session lifetime as well as 1220 means to avoid refresh message synchronization between NATFW NSLP 1221 signaling sessions. [11] recommends: 1223 1. The refresh message timer to be randomly set to a value in the 1224 range [0.5R, 1.5R]. 1226 2. To avoid premature loss of state, lt (with lt being the NATFW 1227 NSLP signaling session lifetime) must satisfy lt >= (K + 1228 0.5)*1.5*R, where K is a small integer. Then in the worst case, 1229 K-1 successive messages may be lost without state being deleted. 1230 Currently K = 3 is suggested as the default. However, it may be 1231 necessary to set a larger K value for hops with high loss rate. 1232 Other algorithms could be used to define the relation between the 1233 NATFW NSLP signaling session lifetime and the refresh message 1234 period; the algorithm provided is only given as an example. 1236 This requested NATFW NSLP signaling session lifetime value lt is 1237 stored in the NATFW_LT object of the NSLP message. 1239 NSLP forwarders can execute the following behavior with respect to 1240 the lifetime handling: 1242 Requested signaling session lifetime acceptable: 1244 No changes to the NATFW NSLP signaling session lifetime values are 1245 needed. The CREATE or EXT message is forwarded. 1247 Signaling session lifetime can be lowered: 1249 The NSLP responder MAY also lower the requested NATFW NSLP 1250 signaling session lifetime to an acceptable value (based on its 1251 local policies). If an NF changes the NATFW NSLP signaling 1252 session lifetime value, it MUST store the new value in the 1253 NATFW_LT object. The CREATE or EXT message is forwarded. 1255 Requested signaling session lifetime is too big: 1257 The NSLP responder MAY reject the requested NATFW NSLP signaling 1258 session lifetime value as being too big and MUST generate an error 1259 RESPONSE message of class 'Signaling session failures' (0x6) with 1260 response code 'Requested lifetime is too big' (0x02) upon 1261 rejection. Lowering the lifetime is preferred instead of 1262 generating an error message. 1264 Requested signaling session lifetime is too small: 1266 The NSLP responder MAY reject the requested NATFW NSLP signaling 1267 session lifetime value as being to small and MUST generate an 1268 error RESPONSE message of class 'Signaling session failures' (0x6) 1269 with response code 'Requested lifetime is too small' (0x10) upon 1270 rejection. 1272 NFs MUST NOT increase the NATFW NSLP signaling session lifetime 1273 value. Messages can be rejected on the basis of the NATFW NSLP 1274 signaling session lifetime being too long when a NATFW NSLP signaling 1275 session is first created and also on refreshes. 1277 The NSLP responder generates a successful RESPONSE for the received 1278 CREATE or EXT message, sets the NATFW NSLP signaling session lifetime 1279 value in the NATFW_LT object to the above granted lifetime and sends 1280 the message back towards NSLP initiator. 1282 Each NSLP forwarder processes the RESPONSE message, reads and stores 1283 the granted NATFW NSLP signaling session lifetime value. The 1284 forwarders MUST accept the granted NATFW NSLP signaling session 1285 lifetime, as long as this value is less than or equal to the 1286 acceptable value. The acceptable value refers to the value accepted 1287 by the NSLP forwarder when processing the CREATE or EXT message. For 1288 received values greater than the acceptable value, NSLP forwarders 1289 MUST generate a RESPONSE message of class 'Signaling session 1290 failures' (0x6) with response code 'Requested lifetime is too big' 1291 (0x02). For received values lower than the values acceptable by the 1292 node local policy, NSLP forwarders MUST generate a RESPONSE message 1293 of class 'Signaling session failures' (0x6) with response code 1294 'Requested lifetime is too small' (0x10). Figure 12 shows the 1295 procedure with an example, where an initiator requests 60 seconds 1296 lifetime in the CREATE message and the lifetime is shortened along 1297 the path by the forwarder to 20 seconds and by the responder to 15 1298 seconds. When the NSLP forwarder receives the RESPONSE message with 1299 a NATFW NSLP signaling session lifetime value of 15 seconds it checks 1300 whether this value is lower or equal to the acceptable value. 1302 +-------+ CREATE(lt=60s) +-------------+ CREATE(lt=20s) +--------+ 1303 | |---------------->| NSLP |---------------->| | 1304 | NI | | forwarder | | NR | 1305 | |<----------------| check 15<20 |<----------------| | 1306 +-------+ RESPONSE(lt=15s)+-------------+ RESPONSE(lt=15s)+--------+ 1308 lt = lifetime 1310 Figure 12: Signaling Session Lifetime Setting Example 1312 3.5. Message Sequencing 1314 NATFW NSLP messages need to carry an identifier so that all nodes 1315 along the path can distinguish messages sent at different points in 1316 time. Messages can be lost along the path or duplicated. So all 1317 NATFW NSLP nodes should be able to identify either old messages that 1318 have been received before (duplicated), or the case that messages 1319 have been lost before (loss). For message replay protection it is 1320 necessary to keep information about messages that have already been 1321 received and requires every NATFW NSLP message to carry a message 1322 sequence number (MSN), see also Section 4.2.6. 1324 The MSN MUST be set by the NI and MUST NOT be set or modified by any 1325 other node. The initial value for the MSN MUST be generated randomly 1326 and MUST be unique only within the NATFW NSLP signaling session for 1327 which it is used. The NI MUST increment the MSN by one for every 1328 message sent. Once the MSN has reached the maximum value, the next 1329 value it takes is zero. All NATFW NSLP nodes MUST use the algorithm 1330 defined in [3] to detect MSN wrap-arounds. 1332 NSIS forwarders and the responder store the MSN from the initial 1333 CREATE or EXT packet which creates the NATFW NSLP signaling session 1334 as the start value for the NATFW NSLP signaling session. NFs and NRs 1335 MUST include the received MSN value in the corresponding RESPONSE 1336 message that they generate. 1338 When receiving a CREATE or EXT message, a NATFW NSLP node uses the 1339 MSN given in the message to determine whether the state being 1340 requested is different to the state already installed. The message 1341 MUST be discarded if the received MSN value is equal to or lower than 1342 the stored MSN value. Such a received MSN value can indicate a 1343 duplicated and delayed message or replayed message. If the received 1344 MSN value is greater than the already stored MSN value, the NATFW 1345 NSLP MUST update its stored state accordingly, if permitted by all 1346 security checks (see Section 3.6), and stores the updated MSN value 1347 accordingly. 1349 3.6. Authentication, Authorization, and Policy Decisions 1351 NATFW NSLP nodes receiving signaling messages MUST first check 1352 whether this message is authenticated and authorized to perform the 1353 requested action. NATFW NSLP nodes requiring more information than 1354 provided MUST generate an error RESPONSE of class 'Permanent failure' 1355 (0x5) with response code 'Authentication failed' (0x01) or with 1356 response code 'Authorization failed' (0x02). 1358 The NATFW NSLP is expected to run in various environments, such as 1359 IP-based telephone systems, enterprise networks, home networks, etc. 1360 The requirements on authentication and authorization are quite 1361 different between these use cases. While a home gateway, or an 1362 Internet cafe, using NSIS may well be happy with a "NATFW signaling 1363 coming from inside the network" policy for authorization of 1364 signaling, enterprise networks are likely to require more strongly 1365 authenticated/authorized signaling. This enterprise scenario may 1366 require the use of an infrastructure and administratively assigned 1367 identities to operate the NATFW NSLP. 1369 Once the NI is authenticated and authorized, another step is 1370 performed. The requested policy rule for the NATFW NSLP signaling 1371 session is checked against a set of policy rules, i.e., whether the 1372 requesting NI is allowed to request the policy rule to be loaded in 1373 the device. If this fails the NF or NR must send an error RESPONSE 1374 of class 'Permanent failure' (0x5) and with response code 1375 'Authorization failed' (0x02). 1377 3.7. Protocol Operations 1379 This section defines the protocol operations including, how to create 1380 NATFW NSLP signaling sessions, maintain them, and how to reserve 1381 addresses. 1383 3.7.1. Creating Signaling Sessions 1385 Allowing two hosts to exchange data even in the presence of 1386 middleboxes is realized in the NATFW NSLP by use of the CREATE 1387 message. The NI (either the data sender or a proxy) generates a 1388 CREATE message as defined in Section 4.3.1 and hands it to the NTLP. 1389 The NTLP forwards the whole message on the basis of the message 1390 routing information (MRI) towards the NR. Each NSIS forwarder along 1391 the path that implements NATFW NSLP, processes the NSLP message. 1392 Forwarding is thus managed NSLP hop-by-hop but may pass transparently 1393 through NSIS forwarders which do not contain NATFW NSLP functionality 1394 and non-NSIS aware routers between NSLP hop way points. When the 1395 message reaches the NR, the NR can accept the request or reject it. 1396 The NR generates a response to CREATE and this response is 1397 transported hop-by-hop towards the NI. NATFW NSLP forwarders may 1398 reject requests at any time. Figure 13 sketches the message flow 1399 between NI (DS in this example), a NF (e.g., NAT), and NR (DR in this 1400 example). 1402 NI Private Network NF Public Internet NR 1403 | | | 1404 | CREATE | | 1405 |----------------------------->| | 1406 | | | 1407 | | | 1408 | | CREATE | 1409 | |--------------------------->| 1410 | | | 1411 | | RESPONSE | 1412 | RESPONSE |<---------------------------| 1413 |<-----------------------------| | 1414 | | | 1415 | | | 1417 Figure 13: CREATE message flow with success RESPONSE 1419 There are several processing rules for a NATFW peer when generating 1420 and receiving CREATE messages, since this message type is used for 1421 creating new NATFW NSLP signaling session, updating existing, 1422 extending the lifetime and deleting NATFW NSLP signaling session. 1423 The three latter functions operate in the same way for all kinds of 1424 CREATE message, and are therefore described in separate sections: 1426 o Extending the lifetime of NATFW NSLP signaling sessions is 1427 described in Section 3.7.3. 1429 o Deleting NATFW NSLP signaling sessions is described in 1430 Section 3.7.4. 1432 o Updating policy rules is described in Section 3.10. 1434 For an initial CREATE message creating a new NATFW NSLP signaling 1435 session, the processing of CREATE messages is different for every 1436 NATFW node type: 1438 o NSLP initiator: An NI only generates CREATE messages and hands 1439 them over to the NTLP. The NI should never receive CREATE 1440 messages and MUST discard it. 1442 o NATFW NSLP forwarder: NFs that are unable to forward the CREATE 1443 message to the next hop MUST generate an error RESPONSE of class 1444 'Permanent failure' (0x6) with response code 'Did not reach the 1445 NR' (0x07). This case may occur if the NTLP layer cannot find an 1446 NATFW NSLP peer, either another NF or the NR, and returns an error 1447 via the GIST API. The NSLP message processing at the NFs depends 1448 on the middlebox type: 1450 * NAT: When the initial CREATE message is received at the public 1451 side of the NAT, it looks for a reservation made in advance, by 1452 using a EXT message (see Section 3.7.2). The matching process 1453 considers the received MRI information and the stored MRI 1454 information, as described in Section 3.8. If no matching 1455 reservation can be found, i.e. no reservation has been made in 1456 advance, the NSLP MUST return an error RESPONSE of class 1457 'Signaling session failure' (0x6) with response code 'No 1458 reservation found matching the MRI of the CREATE request' 1459 (0x03) MUST be generated. If there is a matching reservation, 1460 the NSLP stores the data sender's address (and if applicable 1461 port number) as part of the source address of the policy rule 1462 ('the remembered policy rule') to be loaded and forwards the 1463 message with the destination address set to the internal 1464 (private in most cases) address of NR. When the initial CREATE 1465 message is received at the private side, the NAT binding is 1466 allocated, but not activated (see also Appendix C.3). The MRI 1467 information is updated to reflect the address, and if 1468 applicable port, translation. The NSLP message is forwarded 1469 towards the NR with source address set to the NAT's external 1470 address from the newly remembered binding. 1472 * Firewall: When the initial CREATE message is received, the NSLP 1473 just remembers the requested policy rule, but does not install 1474 any policy rule. Afterwards, the message is forwarded towards 1475 the NR. 1477 * Combined NAT and firewall: Processing at combined firewall and 1478 NAT middleboxes is the same as in the NAT case. No policy 1479 rules are installed. Implementations MUST take into account 1480 the order of packet processing in the firewall and NAT 1481 functions within the device. This will be referred to as 1482 'order of functions' and is generally different depending on 1483 whether the packet arrives at the external or internal side of 1484 the middlebox. 1486 o NSLP receiver: NRs receiving initial CREATE messages MUST reply 1487 with a success RESPONSE of class 'Success' (0x2) with response 1488 code set to 'All successfully processed' (0x01), if they accept 1489 the CREATE message. Otherwise they MUST generate a RESPONSE 1490 message with a suitable response code. RESPONSE messages are sent 1491 back NSLP hop-by-hop towards the NI, irrespective of the response 1492 codes, either success or error. 1494 Remembered policy rules at middleboxes MUST be only installed upon 1495 receiving a corresponding successful RESPONSE message with the same 1496 SID and MSN as the CREATE message that caused them to be remembered. 1497 This is a countermeasure to several problems, for example, wastage of 1498 resources due to loading policy rules at intermediate NFs when the 1499 CREATE message does not reach the final NR for some reason. 1501 Processing of a RESPONSE message is different for every NSIS node 1502 type: 1504 o NSLP initiator: After receiving a successful RESPONSE, the data 1505 path is configured and the DS can start sending its data to the 1506 DR. After receiving an error RESPONSE message, the NI MAY try to 1507 generate the CREATE message again or give up and report the 1508 failure to the application, depending on the error condition. 1510 o NSLP forwarder: NFs install the remembered policy rules, if a 1511 successful RESPONSE message with matching SID and MSN is received. 1512 If an ERROR RESPONSE message with matching SID and MSN is 1513 received, the NATFW NSLP session is marked as dead, no policy rule 1514 is installed and the remembered rule is discarded. 1516 o NSIS responder: The NR should never receive RESPONSE messages and 1517 MUST silently drop any such messages received. 1519 3.7.2. Reserving External Addresses 1521 NSIS signaling is intended to travel end-to-end, even in the presence 1522 of NATs and firewalls on-path. This works well in cases where the 1523 data sender is itself behind a NAT or a firewall as described in 1524 Section 3.7.1. For scenarios where the data receiver is located 1525 behind a NAT or a firewall and it needs to receive data flows from 1526 outside its own network (usually referred to as inbound flows, see 1527 Figure 5) the problem is more troublesome. 1529 NSIS signaling, as well as subsequent data flows, are directed to a 1530 particular destination IP address that must be known in advance and 1531 reachable. Data receivers must tell the local NSIS infrastructure 1532 (i.e., the inbound firewalls/NATs) about incoming NATFW NSLP 1533 signaling and data flows before they can receive these flows. It is 1534 necessary to differentiate between data receivers behind NATs and 1535 behind firewalls for understanding the further NATFW procedures. 1536 Data receivers that are only behind firewalls already have a public 1537 IP address and they need only to be reachable for NATFW signaling. 1538 Unlike data receivers behind just firewalls, data receivers behind 1539 NATs do not have public IP addresses; consequently they are not 1540 reachable for NATFW signaling by entities outside their addressing 1541 realm. 1543 The preceding discussion addresses the situation where a DR node that 1544 wants to be reachable is unreachable because the NAT lacks a suitable 1545 rule with the 'allow' action which would forward inbound data. 1546 However, in certain scenarios, a node situated behind inbound 1547 firewalls that do not block inbound data traffic (firewalls with 1548 "default to allow") unless requested might wish to prevent traffic 1549 being sent to it from specified addresses. In this case, NSIS NATFW 1550 signaling can be used to achieve this by installing a policy rule 1551 with its action set to 'deny' using the same mechanisms as for 1552 'allow' rules. 1554 The required result is obtained by sending a EXTERNAL (EXT) message 1555 in the inbound direction of the intended data flow. When using this 1556 functionality the NSIS initiator for the 'Reserve External Address' 1557 signaling is typically the node that will become the DR for the 1558 eventual data flow. To distinguish this initiator from the usual 1559 case where the NI is associated with the DS, the NI is denoted by NI+ 1560 and the NSIS responder is similarly denoted by NR+. 1562 Public Internet Private Address 1563 Space 1565 Edge 1566 NI(DS) NAT/FW NAT NR(DR) 1567 NR+ NI+ 1569 | | | | 1570 | | | | 1571 | | | | 1572 | | EXT[(DTInfo)] | EXT[(DTInfo)] | 1573 | |<----------------------|<----------------------| 1574 | | | | 1575 | |RESPONSE[Success/Error]|RESPONSE[Success/Error]| 1576 | |---------------------->|---------------------->| 1577 | | | | 1578 | | | | 1580 ============================================================> 1581 Data Traffic Direction 1583 Figure 14: Reservation message flow for DR behind NAT or firewall 1585 Figure 14 shows the EXT message flow for enabling inbound NATFW NSLP 1586 signaling messages. In this case the roles of the different NSIS 1587 entities are: 1589 o The data receiver (DR) for the anticipated data traffic is the 1590 NSIS initiator (NI+) for the EXTERNAL (EXT) message, but becomes 1591 the NSIS responder (NR) for following CREATE messages. 1593 o The actual data sender (DS) will be the NSIS initiator (NI) for 1594 later CREATE messages and may be the NSIS target of the signaling 1595 (NR+). 1597 o It may be necessary to use a signaling destination address (SDA) 1598 as the actual target of the EXT message (NR+) if the DR is located 1599 behind a NAT and the address of the DS is unknown. The SDA is an 1600 arbitrary address in the outermost address realm on the other side 1601 of the NAT from the DR. Typically this will be a suitable public 1602 IP address when the 'outside' realm is the public Internet. This 1603 choice of address causes the EXT message to be routed through the 1604 NATs towards the outermost realm and would force interception of 1605 the message by the outermost NAT in the network at the boundary 1606 between the private address and the public address realm (the 1607 edge-NAT). It may also be intercepted by other NATs and firewalls 1608 on the path to the edge-NAT. 1610 Basically, there are two different signaling scenarios. Either 1612 1. the DR behind the NAT/firewall knows the IP address of the DS in 1613 advance, 1615 2. or the address of DS is not known in advance. 1617 Case 1 requires the NATFW NSLP to request the path-coupled message 1618 routing method (PC-MRM) from the NTLP. The EXT message MUST be sent 1619 with PC-MRM (see Section 5.8.1 in [2]) with the direction set to 1620 'upstream' (inbound). The handling of case 2 depends on the 1621 situation of DR: If DR is solely located behind a firewall, the EXT 1622 message MUST be sent with the PC-MRM, direction 'upstream' (inbound), 1623 and data flow source IP address set to wildcard. If DR is located 1624 behind a NAT, the EXT message MUST be sent with the loose-end message 1625 routing method (LE-MRM, see Section 5.8.2 in [2]), the destination- 1626 address set to the signaling destination address (SDA, see also 1627 Appendix A). For scenarios with DR being behind a firewall, special 1628 conditions apply (applicability statement, Appendix B). The data 1629 receiver is challenged to determine whether it is solely located 1630 behind firewalls or NATs, for choosing the right message routing 1631 method. This decision can depend on a local configuration parameter, 1632 possibly given through DHCP, or it could be discovered through other 1633 non-NSLP related testing of the network configuration. 1635 For case 2 with NAT, the NI+ (which could be on the data receiver DR 1636 or on any other host within the private network) sends the EXT 1637 message targeted to the signaling destination address. The message 1638 routing for the EXT message is in the reverse direction to the normal 1639 message routing used for path-coupled signaling where the signaling 1640 is sent outbound (as opposed to inbound in this case). When 1641 establishing NAT bindings (and an NATFW NSLP signaling session) the 1642 signaling direction does not matter since the data path is modified 1643 through route pinning due to the external IP address at the NAT. 1644 Subsequent NSIS messages (and also data traffic) will travel through 1645 the same NAT boxes. However, this is only valid for the NAT boxes, 1646 but not for any intermediate firewall. That is the reason for having 1647 a separate CREATE message enabling the reservations made with EXT at 1648 the NATs and either enabling prior reservations or creating new 1649 pinholes at the firewalls which are encountered on the outbound path 1650 depending on whether the inbound and outbound routes coincide. 1652 The EXT signaling message creates an NSIS NATFW signaling session at 1653 any intermediate NSIS NATFW peer(s) encountered, independent of the 1654 message routing method used. Furthermore, it has to be ensured that 1655 the edge-NAT or edge-firewall device is discovered as part of this 1656 process. The end host cannot be assumed to know this device - 1657 instead the NAT or firewall box itself is assumed to know that it is 1658 located at the outer perimeter of the network. Forwarding of the EXT 1659 message beyond this entity is not necessary, and MUST be prohibited 1660 as it may provide information on the capabilities of internal hosts. 1661 It should be noted, that it is the outermost NAT or firewall that is 1662 the edge-device that must be found during this discovery process. 1663 For instance, when there are a NAT and afterwards a firewall on the 1664 outbound path at the network border, the firewall is the edge- 1665 firewall. All messages must be forwarded to the topology-wise 1666 outermost edge-device, to ensure that this devices knows about the 1667 NATFW NSLP signaling sessions for incoming CREATE messages. However, 1668 the NAT is still the edge-NAT because it has a public globally 1669 routable IP address on its public side: this is not affected by any 1670 firewall between the edge-NAT and the public network. 1672 Possible edge arrangements are: 1674 Public Net ----------------- Private net -------------- 1676 | Public Net|--|Edge-FW|--|FW|...|FW|--|DR| 1678 | Public Net|--|Edge-FW|--|Edge-NAT|...|NAT or FW|--|DR| 1680 | Public Net|--|Edge-NAT|--|NAT or FW|...|NAT or FW|--|DR| 1682 The edge-NAT or edge-firewall device closest to the public realm 1683 responds to the EXT message with a successful RESPONSE message. An 1684 edge-NAT includes an NATFW_EXT_IP object (see Section 4.2.2), 1685 carrying the public reachable IP address, and if applicable port 1686 number. 1688 There are several processing rules for a NATFW peer when generating 1689 and receiving EXT messages, since this message type is used for 1690 creating new reserve NATFW NSLP signaling sessions, updating 1691 existing, extending the lifetime and deleting NATFW NSLP signaling 1692 session. The three latter functions operate in the same way for all 1693 kinds of CREATE and EXT messages, and are therefore described in 1694 separate sections: 1696 o Extending the lifetime of NATFW NSLP signaling sessions is 1697 described in Section 3.7.3. 1699 o Deleting NATFW NSLP signaling sessions is described in 1700 Section 3.7.4. 1702 o Updating policy rules is described in Section 3.10. 1704 The NI+ MUST include a NATFW_DTINFO object in the EXT message when 1705 using the LE-MRM. The LE-MRM does not include enough information for 1706 some types of NATs (basically, those NATs which also translate port 1707 numbers) to perform the address translation. This information is 1708 provided in the NATFW_DTINFO (see Section 4.2.7). This information 1709 MUST include at least the 'dst port number' and 'protocol' fields, in 1710 the NATFW_DTINFO object as these may be required by en-route NATs, 1711 depending on the type of the NAT. All other fields MAY be set by the 1712 NI+ to restrict the set of possible NIs. An edge-NAT will use the 1713 information provided in the NATFW_DTINFO object to allow only NATFW 1714 CREATE message with the MRI matching ('src IPv4/v6 address', 'src 1715 port number', 'protocol') to be forwarded. A NAT requiring 1716 information carried in the NATFW_DTINFO can generate a number of 1717 error RESPONSE messages of class 'Signaling session failures' (0x6): 1719 o 'Requested policy rule denied due to policy conflict' (0x04) 1721 o 'NATFW_DTINFO object is required' (0x07) 1723 o 'Requested value in sub_ports field in NATFW_EFI not permitted' 1724 (0x08) 1726 o 'Requested IP protocol not supported' (0x09) 1728 o 'Plain IP policy rules not permitted -- need transport layer 1729 information' (0x0A) 1731 o 'source IP address range is too large' (0x0C) 1733 o 'destination IP address range is too large' (0x0D) 1735 o 'source L4-port range is too large' (0x0E) 1737 o 'destination L4-port range is too large' (0x0F) 1739 Processing of EXT messages is specific to the NSIS node type: 1741 o NSLP initiator: NI+ only generate EXT messages. When the data 1742 sender's address information is known in advance the NI+ can 1743 include a NATFW_DTINFO object in the EXT message, if not anyway 1744 required to do so (see above). When the data sender's IP address 1745 is not known, the NI+ MUST NOT include a NATFW_DTINFO object. The 1746 NI should never receive EXT messages and MUST silently discard it. 1748 o NSLP forwarder: The NSLP message processing at NFs depends on the 1749 middlebox type: 1751 * NAT: NATs check whether the message is received at the external 1752 (public in most cases) address or at the internal (private) 1753 address. If received at the external an NF MUST generate an 1754 error RESPONSE of class 'Protocol error' (0x3) with response 1755 code 'Received EXT request message on external side' (0x0B). 1756 If received at the internal (private address) and the NATFW_EFI 1757 object contains the action 'deny', an error RESPONSE of class 1758 'Protocol error' (0x3) with response code 'Requested rule 1759 action not applicable' (0x06) MUST be generated. If received 1760 at the internal address, an IP address, and if applicable one 1761 or more ports, are reserved. If it is an edge-NAT and there is 1762 no edge-firewall beyond, the NSLP message is not forwarded any 1763 further and a successful RESPONSE message is generated 1764 containing an NATFW_EXT_IP object holding the translated 1765 address, and if applicable port, information from the binding 1766 reserved as a result of the EXT message. The RESPONSE message 1767 is sent back towards the NI+. If it is not an edge-NAT, the 1768 NSLP message is forwarded further using the translated IP 1769 address as signaling source address in the LE-MRM and 1770 translated port in the NATFW_DTINFO object in the field 'DR 1771 port number', i.e., the NATFW_DTINFO object is updated to 1772 reflect the translated port number. The edge-NAT or any other 1773 NAT MUST reject EXT messages not carrying a NATFW_DTINFO object 1774 or if the address information within this object is invalid or 1775 is not compliant with local policies (e.g., the information 1776 provided relates to a range of addresses ('wildcarded') but the 1777 edge-NAT requires exact information about DS' IP address and 1778 port) with the above mentioned response codes. 1780 * Firewall: Non edge-firewalls remember the requested policy 1781 rule, keep NATFW NSLP signaling session state, and forward the 1782 message. Edge-firewalls stop forwarding the EXT message. The 1783 policy rule is immediately loaded if the action in the 1784 NATFW_EFI object is set to 'deny' and the node is an edge- 1785 firewall. The policy rule is remembered, but not activated, if 1786 the action in the NATFW_EFI object is set to 'allow'. In both 1787 cases, a successful RESPONSE message is generated. If the 1788 action is 'allow', and the NATFW_DTINFO object is included, and 1789 the MRM is set to LE-MRM in the request, additionally an 1790 NATFW_EXT_IP object is included in the RESPONSE message, 1791 holding the translated address, and if applicable port, 1792 information. This information is obtained from the 1793 NATFW_DTINFO object's 'DR port number' and the source-address 1794 of the LE-MRM. 1796 * Combined NAT and firewall: Processing at combined firewall and 1797 NAT middleboxes is the same as in the NAT case. 1799 o NSLP receiver: This type of message should never be received by 1800 any NR+ and it MUST generate an error RESPONSE message of class 1801 'Permanent failure' (0x5) with response code 'No edge-device here' 1802 (0x06). 1804 Processing of a RESPONSE message is different for every NSIS node 1805 type: 1807 o NSLP initiator: Upon receiving a successful RESPONSE message, the 1808 NI+ can rely on the requested configuration for future inbound 1809 NATFW NSLP signaling sessions. If the response contains an 1810 NATFW_EXT_IP object, the NI can use IP address and port pairs 1811 carried for further application signaling. After receiving a 1812 error RESPONSE message, the NI+ MAY try to generate the EXT 1813 message again or give up and report the failure to the 1814 application, depending on the error condition. 1816 o NSLP forwarder: NFs simply forward this message as long as they 1817 keep state for the requested reservation, if the RESPONSE message 1818 contains NATFW_INFO object with class set to 'Success' (0x2). If 1819 the RESPONSE message contains NATFW_INFO object with class set not 1820 to 'Success' (0x2), the NATFW NSLP signaling session is marked as 1821 dead. 1823 o NSIS responder: This type of message should never be received by 1824 any NR+. The NF should never receive response messages and MUST 1825 silently discard it. 1827 Reservations with action 'allow' made with EXT MUST be enabled by a 1828 subsequent CREATE message. A reservation made with EXT (independent 1829 of selected action) is kept alive as long as the NI+ refreshes the 1830 particular NATFW NSLP signaling session and it can be reused for 1831 multiple, different CREATE messages. An NI+ may decide to teardown a 1832 reservation immediately after receiving a CREATE message. This 1833 implies that a new NATFW NSLP signaling session must be created for 1834 each new CREATE message. The CREATE message does not re-use the 1835 NATFW NSLP signaling session created by REA. 1837 Without using CREATE Section 3.7.1 or EXT in proxy mode Section 3.7.6 1838 no data traffic will be forwarded to DR beyond the edge-NAT or edge- 1839 firewall. The only function of EXT is to ensure that subsequent 1840 CREATE messages traveling towards the NR will be forwarded across the 1841 public-private boundary towards the DR. Correlation of incoming 1842 CREATE messages to EXT reservation states is described in 1843 Section 3.8. 1845 3.7.3. NATFW NSLP Signaling Session Refresh 1847 NATFW NSLP signaling sessions are maintained on a soft-state basis. 1848 After a specified timeout, sessions and corresponding policy rules 1849 are removed automatically by the middlebox, if they are not 1850 refreshed. Soft-state is created by CREATE and EXT and the 1851 maintenance of this state must be done by these messages. State 1852 created by CREATE must be maintained by CREATE, state created by EXT 1853 must be maintained by EXT. Refresh messages, are messages carrying 1854 the same session ID as the initial message and a NATFW_LT lifetime 1855 object with a lifetime greater than zero. Messages with the same SID 1856 but carrying a different MRI are treated as updates of the policy 1857 rules and are processed as defined in Section 3.10. Every refresh 1858 CREATE or EXT message MUST be acknowledged by an appropriate response 1859 message generated by the NR. Upon reception by each NSIS forwarder, 1860 the state for the given session ID is extended by the NATFW NSLP 1861 signaling session refresh period, a period of time calculated based 1862 on a proposed refresh message period. The lifetime extension of a 1863 NATFW NSLP signaling session is calculated as current local time plus 1864 proposed lifetime value (NATFW NSLP signaling session refresh 1865 period). Section 3.4 defines the process of calculating lifetimes in 1866 detail. 1868 NI Public Internet NAT Private address NR 1870 | | space | 1871 | CREATE[lifetime > 0] | | 1873 |----------------------------->| | 1874 | | | 1875 | | | 1876 | | CREATE[lifetime > 0] | 1877 | |--------------------------->| 1878 | | | 1879 | | RESPONSE[Success/Error] | 1880 | RESPONSE[Success/Error] |<---------------------------| 1881 |<-----------------------------| | 1882 | | | 1883 | | | 1885 Figure 16: Successful Refresh Message Flow, CREATE as example 1887 Processing of NATFW NSLP signaling session refresh CREATE and EXT 1888 messages is different for every NSIS node type: 1890 o NSLP initiator: The NI/NI+ can generate NATFW NSLP signaling 1891 session refresh CREATE/EXT messages before the NATFW NSLP 1892 signaling session times out. The rate at which the refresh 1893 CREATE/EXT messages are sent and their relation to the NATFW NSLP 1894 signaling session state lifetime is discussed further in 1895 Section 3.4. 1897 o NSLP forwarder: Processing of this message is independent of the 1898 middlebox type and is as described in Section 3.4. 1900 o NSLP responder: NRs accepting a NATFW NSLP signaling session 1901 refresh CREATE/EXT message generate a successful RESPONSE message, 1902 including the granted lifetime value of Section 3.4 in a NATFW_LT 1903 object. 1905 3.7.4. Deleting Signaling Sessions 1907 NATFW NSLP signaling sessions can be deleted at any time. NSLP 1908 initiators can trigger this deletion by using a CREATE or EXT 1909 messages with a lifetime value set to 0, as shown in Figure 17. 1910 Whether a CREATE or EXT message type is used, depends on how the 1911 NATFW NSLP signaling session was created. 1913 NI Public Internet NAT Private address NR 1915 | | space | 1916 | CREATE[lifetime=0] | | 1917 |----------------------------->| | 1918 | | | 1919 | | CREATE[lifetime=0] | 1920 | |--------------------------->| 1921 | | | 1923 Figure 17: Delete message flow, CREATE as example 1925 NSLP nodes receiving this message delete the NATFW NSLP signaling 1926 session immediately. Policy rules associated with this particular 1927 NATFW NSLP signaling session MUST be also deleted immediately. This 1928 message is forwarded until it reaches the final NR. The CREATE/EXT 1929 message with a lifetime value of 0, does not generate any response, 1930 neither positive nor negative, since there is no NSIS state left at 1931 the nodes along the path. 1933 NSIS initiators can use CREATE/EXT message with lifetime set to zero 1934 in an aggregated way, such that a single CREATE or EXT message is 1935 terminating multiple NATFW NSLP signaling sessions. NIs can follow 1936 this procedure if the like to aggregate NATFW NSLP signaling session 1937 deletion requests: The NI uses the CREATE or EXT message with the 1938 session ID set to zero and the MRI's source-address set to its used 1939 IP address. All other fields of the respective NATFW NSLP signaling 1940 sessions to be terminated are set as well, otherwise these fields are 1941 completely wildcarded. The NSLP message is transferred to the NTLP 1942 requesting 'explicit routing' as described in Sections 5.2.1 and 1943 7.1.4. in [2]. 1945 The outbound NF receiving such an aggregated CREATE or EXT message 1946 MUST reject it with an error RESPONSE of class 'Permanent failure' 1947 (0x5) with response code 'Authentication failed' (0x01) if the 1948 authentication fails and with an error RESPONSE of class 'Permanent 1949 failure' (0x5) with response code 'Authorization failed' (0x02) if 1950 the authorization fails. Per NATFW NSLP signaling session proof of 1951 ownership, as it is defined in this memo, is not possible anymore 1952 when using this aggregated way. However, the outbound NF can use the 1953 relationship between the information of the received CREATE or EXT 1954 message and the GIST messaging association where the request has been 1955 received. The outbound NF MUST only accept this aggregated CREATE or 1956 EXT message through already established GIST messaging associations 1957 with the NI. The outbound NF MUST NOT propagate this aggregated 1958 CREATE or EXT message but it MAY generate and forward per NATFW NSLP 1959 signaling session CREATE or EXT messages. 1961 3.7.5. Reporting Asynchronous Events 1963 NATFW NSLP forwarders and NATFW NSLP responders must have the ability 1964 to report asynchronous events to other NATFW NSLP nodes, especially 1965 to allow reporting back to the NATFW NSLP initiator. Such 1966 asynchronous events may be premature NATFW NSLP signaling session 1967 termination, changes in local policies, route change or any other 1968 reason that indicates change of the NATFW NSLP signaling session 1969 state. 1971 NFs and NRs may generate NOTIFY messages upon asynchronous events, 1972 with a NATFW_INFO object indicating the reason for event. These 1973 reasons can be carried in the NATFW_INFO object (class MUST be set to 1974 'Informational' (0x1)) within the NOTIFY message. This list shows 1975 the response codes and the associated actions to take at NFs and the 1976 NI: 1978 o 'Route change: possible route change on the outbound path' (0x01): 1979 Follow instructions in Section 3.9. This MUST be sent inbound. 1981 o 'Re-authentication required' (0x02): The NI should re-send the 1982 authentication. This MUST be sent inbound. 1984 o 'NATFW node is going down soon' (0x03): The NI and other NFs 1985 should be prepared for a service interruption at any time. This 1986 message MAY be sent inbound and outbound. 1988 o 'NATFW signaling session lifetime expired' (0x04): The NATFW 1989 signaling session has been expired and the signaling session is 1990 invalid now. NFs MUST mark the signaling session as 'Dead'. This 1991 message MAY be sent inbound and outbound. 1993 NOTIFY messages are always sent hop-by-hop inbound towards NI until 1994 they reach NI or outbound towards the NR as indicated in the list 1995 above. 1997 The initial processing when receiving a NOTIFY message is the same 1998 for all NATFW nodes: NATFW nodes MUST only accept NOTIFY messages 1999 through already established NTLP messaging associations. The further 2000 processing is different for each NATFW NSLP node type and depends on 2001 the events notified: 2003 o NSLP initiator: NIs analyze the notified event and behave 2004 appropriately based on the event type. NIs MUST NOT generate 2005 NOTIFY messages. 2007 o NSLP forwarder: NFs analyze the notified event and behave based on 2008 the above description per response code. NFs SHOULD generate 2009 NOTIFY messages upon asynchronous events and forward them inbound 2010 towards the NI or outbound towards the NR, depending on the 2011 received direction, i.e., inbound messages MUST be forwarded 2012 further inbound and outbound messages MUST be forwarded further 2013 inbound. NFs MUST silently discard NOTIFY messages that have been 2014 received outbound but are only allowed to be sent inbound, e.g. 2015 'Re-authentication required' (0x02). 2017 o NSLP responder: NRs SHOULD generate NOTIFY messages upon 2018 asynchronous events including a response code based on the 2019 reported event. The NR MUST silently discard NOTIFY messages that 2020 have been received outbound but are only allowed to be sent 2021 inbound, e.g. 'Re-authentication required' (0x02), 2023 NATFW NSLP forwarders, keeping multiple NATFW NSLP signaling sessions 2024 at the same time, can experience problems when shutting down service 2025 suddenly. This sudden shutdown can be result of node local failure, 2026 for instance, due to a hardware failure. This NF generates NOTIFY 2027 messages for each of the NATFW NSLP signaling sessions and tries to 2028 send them inbound. Due to the number of NOTIFY messages to be sent, 2029 the shutdown of the node may be unnecessarily prolonged, since not 2030 all messages can be sent at the same time. This case can be 2031 described as a NOTIFY storm, if a multitude of NATFW NSLP signaling 2032 sessions is involved. 2034 To avoid the need of generating per NATFW NSLP signaling session 2035 NOTIFY messages in such a scenario described or similar cases, NFs 2036 SHOULD follow this procedure: The NF uses the NOTIFY message with the 2037 session ID in the NTLP set to zero, with the MRI completely 2038 wildcarded, using the 'explicit routing' as described in Sections 2039 5.2.1 and 7.1.4. in [2]. The inbound NF receiving this type of 2040 NOTIFY immediately regards all NATFW NSLP signaling sessions from 2041 that peer matching the MRI as void. This message will typically 2042 result in multiple NOTIFY messages at the inbound NF, i.e., the NF 2043 can generate per terminated NATFW NSLP signaling session a NOTIFY 2044 message. However, a NF MAY aggregate again the NOTIFY messages as 2045 described here. 2047 3.7.6. Proxy Mode of Operation 2049 Some migration scenarios need specialized support to cope with cases 2050 where NSIS is only deployed in same areas of the Internet. End-to- 2051 end signaling is going to fail without NSIS support at or near both 2052 data sender and data receiver terminals. A proxy mode of operation 2053 is needed. This proxy mode of operation must terminate the NATFW 2054 NSLP signaling as topologically close to the terminal for which it is 2055 proxying and proxy all messages. This NATFW NSLP node doing the 2056 proxying of the signaling messages becomes either the NI or the NR 2057 for the particular NATFW NSLP signaling session, depending on whether 2058 it is the DS or DR that does not support NSIS. Typically, the edge- 2059 NAT or the edge-firewall would be used to proxy NATFW NSLP messages. 2061 This proxy mode operation does not require any new CREATE or EXT 2062 message type, but relies on extended CREATE and EXT message types. 2063 They are called respectively CREATE-PROXY and EXT-PROXY and are 2064 distinguished by setting the P flag in the NSLP header to P=1. This 2065 flag instructs edge-NATs and edge-firewalls receiving them to operate 2066 in proxy mode for the NATFW NSLP signaling session in question. The 2067 semantics of the CREATE and EXT message types are not changed and the 2068 behavior of the various node types is as defined in Section 3.7.1 and 2069 Section 3.7.2, except for the proxying node. The following 2070 paragraphs describe the proxy mode operation for data receivers 2071 behind middleboxes and data senders behind middleboxes. 2073 3.7.6.1. Proxying for a Data Sender 2075 The NATFW NSLP gives the NR the ability to install state on the 2076 inbound path towards the data sender for outbound data packets, even 2077 when only the receiving side is running NSIS (as shown in Figure 18). 2078 The goal of the method described is to trigger the edge-NAT/ 2079 edge-firewall to generate a CREATE message on behalf of the data 2080 receiver. In this case, an NR can signal towards the network border 2081 as it is performed in the standard EXT message handling scenario as 2082 in Section 3.7.2. The message is forwarded until the edge-NAT/ 2083 edge-firewall is reached. A public IP address and port number is 2084 reserved at an edge-NAT/edge-firewall. As shown in Figure 18, unlike 2085 the standard EXT message handling case, the edge-NAT/edge-firewall is 2086 triggered to send a CREATE message on a new reverse path which 2087 traverse several firewalls or NATs. The new reverse path for CREATE 2088 is necessary to handle routing asymmetries between the edge-NAT/ 2089 edge-firewall and DR. It must be stressed that the semantics of the 2090 CREATE and EXT messages is not changed, i.e., each is processed as 2091 described earlier. 2093 DS Public Internet NAT/FW Private address NR 2094 No NI NF space NI+ 2095 NR+ 2097 | | EXT-PROXY[(DTInfo)] | 2098 | |<------------------------- | 2099 | | RESPONSE[Error/Success] | 2100 | | ---------------------- > | 2101 | | CREATE | 2102 | | ------------------------> | 2103 | | RESPONSE[Error/Success] | 2104 | | <---------------------- | 2105 | | | 2107 Figure 18: EXT Triggering Sending of CREATE Message 2109 A NATFW_NONCE object, carried in the EXT and CREATE message, is used 2110 to build the relationship between received CREATEs at the message 2111 initiator. An NI+ uses the presence of the NATFW_NONCE object to 2112 correlate it to the particular EXT-PROXY. The absence of a NONCE 2113 object indicates a CREATE initiated by the DS and not by the edge- 2114 NAT. Therefore, these processing rules of EXT-PROXY messages are 2115 added to the regular EXT processing: 2117 o NSLP initiator (NI+): The NI+ MUST choose a random value and place 2118 it in the NATFW_NONCE object. 2120 o NSLP forwarder being either edge-NAT or edge-firewall: When the NF 2121 accepts a EXT_PROXY message, it generates a successful RESPONSE 2122 message as if it were the NR and additionally, it generates a 2123 CREATE message as defined in Section 3.7.1 and includes a 2124 NATFW_NONCE object having the same value as of the received 2125 NATFW_NONCE object. The NF MUST not generate a CREATE-PROXY 2126 message. The NF MUST refresh the CREATE message signaling session 2127 only if a EXT-PROXY refresh message has been received first. This 2128 also includes tearing down signaling sessions, i.e., the NF must 2129 teardown the CREATE signaling session only if a EXT-PROXY message 2130 with lifetime set to 0 has been received first. 2132 The scenario described in this section challenges the data receiver 2133 because it must make a correct assumption about the data sender's 2134 ability to use NSIS NATFW NSLP signaling. It is possible for the DR 2135 to make the wrong assumption in two different ways: 2137 a) the DS is NSIS unaware but the DR assumes the DS to be NSIS 2138 aware and 2140 b) the DS is NSIS aware but the DR assumes the DS to be NSIS 2141 unaware. 2143 Case a) will result in middleboxes blocking the data traffic, since 2144 DS will never send the expected CREATE message. Case b) will result 2145 in the DR successfully requesting proxy mode support by the edge-NAT 2146 or edge-firewall. The edge-NAT/edge-firewall will send CREATE 2147 messages and DS will send CREATE messages as well. Both CREATE 2148 messages are handled as separated NATFW NSLP signaling sessions and 2149 therefore the common rules per NATFW NSLP signaling session apply; 2150 the NATFW_NONCE object is used to differentiate CREATE messages 2151 generated by the edge-NAT/edge-firewall from NI initiated CREATE 2152 messages. It is the NR's responsibility to decide whether to 2153 teardown the EXT-PROXY signaling sessions in the case where the data 2154 sender's side is NSIS aware, but was incorrectly assumed not to be so 2155 by the DR. It is RECOMMENDED that a DR behind NATs uses the proxy 2156 mode of operation by default, unless the DR knows that the DS is NSIS 2157 aware. The DR MAY cache information about data senders which it has 2158 found to be NSIS aware in past NATFW NSLP signaling sessions. 2160 There is a possible race condition between the RESPONSE message to 2161 the EXT-PROXY and the CREATE message generated by the edge-NAT. The 2162 CREATE message can arrive earlier than the RESPONSE message. An NI+ 2163 MUST accept CREATE messages generated by the edge-NAT even if the 2164 RESPONSE message to the EXT-PROXY was not received. 2166 3.7.6.2. Proxying for a Data Receiver 2168 As with data receivers behind middleboxes, data senders behind 2169 middleboxes can require proxy mode support. The issue here is that 2170 there is no NSIS support at the data receiver's side and, by default, 2171 there will be no response to CREATE messages. This scenario requires 2172 the last NSIS NATFW NSLP aware node to terminate the forwarding and 2173 to proxy the response to the CREATE message, meaning that this node 2174 is generating RESPONSE messages. This last node may be an edge-NAT/ 2175 edge-firewall, or any other NATFW NSLP peer, that detects that there 2176 is no NR available (probably as a result of GIST timeouts but there 2177 may be other triggers). 2179 DS Private Address NAT/FW Public Internet NR 2180 NI Space NF no NR 2182 | | | 2183 | CREATE-PROXY | | 2184 |------------------------------>| | 2185 | | | 2186 | RESPONSE[SUCCESS/ERROR] | | 2187 |<------------------------------| | 2188 | | | 2190 Figure 19: Proxy Mode CREATE Message Flow 2192 The processing of CREATE-PROXY messages and RESPONSE messages is 2193 similar to Section 3.7.1, except that forwarding is stopped at the 2194 edge-NAT/edge-firewall. The edge-NAT/edge-firewall responds back to 2195 NI according the situation (error/success) and will be the NR for 2196 future NATFW NSLP communication. 2198 The NI can choose the proxy mode of operation although the DR is NSIS 2199 aware. The CREATE-PROXY mode would not configure all NATs and 2200 firewalls along the data path, since it is terminated at the edge- 2201 device. Any device beyond this point will never receive any NATFW 2202 NSLP signaling for this flow. 2204 3.8. De-Multiplexing at NATs 2206 Section 3.7.2 describes how NSIS nodes behind NATs can obtain a 2207 public reachable IP address and port number at a NAT and and how the 2208 resulting mapping rule can be activated by using CREATE messages (see 2209 Section 3.7.1). The information about the public IP address/port 2210 number can be transmitted via an application level signaling protocol 2211 and/or third party to the communication partner that would like to 2212 send data toward the host behind the NAT. However, NSIS signaling 2213 flows are sent towards the address of the NAT at which this 2214 particular IP address and port number is allocated and not directly 2215 to the allocated IP address and port number. The NATFW NSLP 2216 forwarder at this NAT needs to know how the incoming NSLP CREATE 2217 messages are related to reserved addresses, meaning how to de- 2218 multiplex incoming NSIS CREATE messages. 2220 The de-multiplexing method uses information stored at the local NATFW 2221 NSLP node and the of the policy rule. The policy rule uses the LE- 2222 MRM MRI source-address (see [2]) as the flow destination IP address 2223 and the network-layer-version as IP version. The external IP address 2224 at the NAT is stored as the external flow destination IP address. 2225 All other parameters of the policy rule other than the flow 2226 destination IP address are wildcarded if no NATFW_DTINFO object is 2227 included in the EXT message. The LE-MRM MRI destination-address MUST 2228 NOT be used in the policy rule, since it is solely a signaling 2229 destination address. 2231 If the NATFW_DTINFO object is included in the EXT message, the policy 2232 rule is filled with further information. The 'dst port number' field 2233 of the NATFW_DTINFO is stored as the flow destination port number. 2234 The 'protocol' field is stored as the flow protocol. The 'src port 2235 number' field is stored as the flow source port number. The 'data 2236 sender's IPv4 address' is stored as the flow source IP address. Note 2237 that some of these field can contain wildcards. 2239 When receiving a CREATE message at the NATFW NSLP it uses the flow 2240 information stored in the MRI to do the matching process. This table 2241 shows the parameters to be compared against each others. Note that 2242 not all parameters can be present in a MRI at the same time. 2244 +-------------------------------+--------------------------------+ 2245 | Flow parameter (Policy Rule) | MRI parameter (CREATE message) | 2246 +-------------------------------+--------------------------------+ 2247 | IP version | network-layer-version | 2248 | | | 2249 | Protocol | IP-protocol | 2250 | | | 2251 | source IP address (w) | source-address (w) | 2252 | | | 2253 | external IP address | destination-address | 2254 | | | 2255 | destination IP address (n/u) | N/A | 2256 | | | 2257 | source port number (w) | L4-source-port (w) | 2258 | | | 2259 | external port number (w) | L4-destination-port (w) | 2260 | | | 2261 | destination port number (n/u) | N/A | 2262 | | | 2263 | IPsec SPI | ipsec-SPI | 2264 +-------------------------------+--------------------------------+ 2266 Table entries marked with (w) can be wildcarded and entries marked 2267 with (n/u) are not used for the matching. 2269 Table 1 2271 3.9. Reacting to Route Changes 2273 The NATFW NSLP needs to react to route changes in the data path. 2274 This assumes the capability to detect route changes, to perform NAT 2275 and firewall configuration on the new path and possibly to tear down 2276 NATFW NSLP signaling session state on the old path. The detection of 2277 route changes is described in Section 7 of [2] and the NATFW NSLP 2278 relies on notifications about route changes by the NTLP. This 2279 notification will be conveyed by the API between NTLP and NSLP, which 2280 is out of scope of this memo. 2282 A NATFW NSLP node other than the NI or NI+ detecting a route change, 2283 by means described in the NTLP specification or others, generates a 2284 NOTIFY message indicating this change and sends this inbound towards 2285 NI. Intermediate NFs on the way to the NI can use this information 2286 to decide later if their NATFW NSLP signaling session can be deleted 2287 locally, if they do not receive an update within a certain time 2288 period, as described in Section 3.2.3. It is important to consider 2289 the transport limitations of NOTIFY messages as mandated in 2290 Section 3.7.5. 2292 The NI receiving this NOTIFY message MAY generate a new CREATE or EXT 2293 message and sends it towards the NATFW NSLP signaling session's NI as 2294 for the initial message using the same session ID. All the remaining 2295 processing and message forwarding, such as NSLP next hop discovery, 2296 is subject to regular NSLP processing as described in the particular 2297 sections. Normal routing will guide the new CREATE or EXT message to 2298 the correct NFs along the changed route. NFs that were on the 2299 original path receiving these new CREATE or EXT messages (see also 2300 Section 3.10), can use the session ID to update the existing NATFW 2301 NSLP signaling session, whereas NFs that were not on the original 2302 path will create new state for this NATFW NSLP signaling session. 2303 The next section describes how policy rules are updated. 2305 3.10. Updating Policy Rules 2307 NSIS initiators can request an update of the installed/reserved 2308 policy rules at any time within a NATFW NSLP signaling session. 2309 Updates to policy rules can be required due to node mobility (NI is 2310 moving from one IP address to another), route changes (this can 2311 result in a different NAT mapping at a different NAT device), or the 2312 wish of the NI to simply change the rule. NIs can update policy 2313 rules in existing NATFW NSLP signaling sessions by sending an 2314 appropriate CREATE or EXT message (similar to Section 3.4) with 2315 modified message routing information (MRI) as compared with that 2316 installed previously, but using the existing session ID to identify 2317 the intended target of the update. With respect to authorization and 2318 authentication, this update CREATE or EXT message is treated in 2319 exactly the same way as any initial message. Therefore, any node 2320 along in the NATFW NSLP signaling session can reject the update with 2321 an error RESPONSE message, as defined in the previous sections. 2323 The message processing and forwarding is executed as defined in the 2324 particular sections. A NF or the NR receiving an update, simply 2325 replaces the installed policy rules installed in the firewall/NAT. 2326 The local procedures on how to update the MRI in the firewall/NAT is 2327 out of scope of this memo 2329 4. NATFW NSLP Message Components 2331 A NATFW NSLP message consists of a NSLP header and one or more 2332 objects following the header. The NSLP header is carried in all 2333 NATFW NSLP message and objects are Type-Length-Value (TLV) encoded 2334 using big endian (network ordered) binary data representations. 2335 Header and objects are aligned to 32 bit boundaries and object 2336 lengths that are not multiples of 32 bits must be padded to the next 2337 higher 32 bit multiple. 2339 The whole NSLP message is carried as payload of a NTLP message. 2341 Note that the notation 0x is used to indicate hexadecimal numbers. 2343 4.1. NSLP Header 2345 All GIST NSLP-Data objects for the NATFW NSLP MUST contain this 2346 common header as the first 32 bits of the object (this is not the 2347 same as the GIST Common Header). It contains two fields, the NSLP 2348 message type and a reserved field. The total length is 32 bits. The 2349 layout of the NSLP header is defined by Figure 20. 2351 0 1 2 3 2352 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2353 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2354 | Message type |P| reserved | reserved | 2355 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2357 Figure 20: Common NSLP header 2359 The reserved field MUST be set to zero in the NATFW NSLP header 2360 before sending and MUST be ignored during processing of the header. 2362 The defined messages types are: 2364 o IANA-TBD(1) : CREATE 2366 o IANA-TBD(2) : EXTERNAL(EXT) 2368 o IANA-TBD(3) : RESPONSE 2370 o IANA-TBD(4) : NOTIFY 2372 If a message with another type is received, an error RESPONSE of 2373 class 'Protocol error' (0x3) with response code 'Illegal message 2374 type' (0x01) MUST be generated. 2376 The P flag indicates the usage of proxy mode. If proxy mode is used 2377 it MUST be set to 1. Proxy mode usage is only allowed in combination 2378 with the message types CREATE and EXT, P=1 MUST NOT be set with 2379 message types other than CREATE and EXT. The P flag MUST be ignored 2380 when processing messages with type RESPONSE. An error RESPONSE 2381 message of class 'Protocol error' (0x3) and type 'Bad flags value' 2382 (0x03) MUST be generated, if the P flag is set in NOTIFY messages. 2384 4.2. NSLP Objects 2386 NATFW NSLP objects use a common header format defined by Figure 21. 2387 The object header contains two fields, the NSLP object type and the 2388 object length. Its total length is 32 bits. 2390 0 1 2 3 2391 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2392 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2393 |A|B|r|r| Object Type |r|r|r|r| Object Length | 2394 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2396 Figure 21: Common NSLP object header 2398 The object length field contains the total length of the object 2399 without the object header. The unit is a word, consisting of 4 2400 octets. The particular values of type and length for each NSLP 2401 object are listed in the subsequent sections that define the NSLP 2402 objects. An error RESPONSE of class 'Protocol error' (0x3) with 2403 response code 'Wrong object length' (0x07) MUST be generated if the 2404 length given for the object in the object header did not match the 2405 length of the object data present. The two leading bits of the NSLP 2406 object header are used to signal the desired treatment for objects 2407 whose treatment has not been defined in this memo (see [2], Section 2408 A.2.1), i.e., the Object Type has not been defined. NATFW NSLP uses 2409 a subset of the categories defined in GIST: 2411 o AB=00 ("Mandatory"): If the object is not understood, the entire 2412 message containing it MUST be rejected with an error RESPONSE of 2413 class 'Protocol error' (0x3) with response code 'Unknown object 2414 present' (0x06). 2416 o AB=01 ("Optional"): If the object is not understood, it should be 2417 deleted and then the rest of the message processed as usual. 2419 o AB=10 ("Forward"): If the object is not understood, it should be 2420 retained unchanged in any message forwarded as a result of message 2421 processing, but not stored locally. 2423 The combination AB=11 MUST NOT be used and an error RESPONSE of class 2424 'Protocol error' (0x3) with response code 'Invalid Flag-Field 2425 combination' (0x09) MUST be generated. 2427 The following sections do not repeat the common NSLP object header, 2428 they just list the type and the length. 2430 4.2.1. Signaling Session Lifetime Object 2432 The signaling session lifetime object carries the requested or 2433 granted lifetime of a NATFW NSLP signaling session measured in 2434 seconds. 2436 Type: NATFW_LT (IANA-TBD) 2438 Length: 1 2440 0 1 2 3 2441 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2442 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2443 | NATFW NSLP signaling session lifetime | 2444 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2446 Figure 22: Signaling Session Lifetime object 2448 4.2.2. External Address Object 2450 The external address object can be included in RESPONSE messages 2451 (Section 4.3.3) only. It carries the publicly reachable IP address, 2452 and if applicable port number, at an edge-NAT. 2454 Type: NATFW_EXT_IP (IANA-TBD) 2456 Length: 2 2458 0 1 2 3 2459 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2460 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2461 | port number | reserved | 2462 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2463 | IPv4 address | 2464 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2466 Figure 23: External Address Object for IPv4 addresses 2468 Please note that the field 'port number' MUST be set to 0 if only an 2469 IP address has been reserved, for instance, by a traditional NAT. A 2470 port number of 0 MUST be ignored in processing this object. 2472 4.2.3. Extended Flow Information Object 2474 In general, flow information is kept in the message routing 2475 information (MRI) of the NTLP. Nevertheless, some additional 2476 information may be required for NSLP operations. The 'extended flow 2477 information' object carries this additional information about the 2478 action of the policy rule for firewalls/NATs and contiguous port . 2480 Type: NATFW_EFI (IANA-TBD) 2482 Length: 1 2484 0 1 2 3 2485 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2486 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2487 | rule action | sub_ports | 2488 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2490 Figure 24: Extended Flow Information 2492 This object has two fields, 'rule action' and 'sub_ports'. The 'rule 2493 action' field has these meanings: 2495 o 0x0001: Allow: A policy rule with this action allows data traffic 2496 to traverse the middlebox and the NATFW NSLP MUST allow NSLP 2497 signaling to be forwarded. 2499 o 0x0002: Deny: A policy rule with this action blocks data traffic 2500 from traversing the middlebox and the NATFW NSLP MUST NOT allow 2501 NSLP signaling to be forwarded. 2503 If the 'rule action' field contains neither 0x0001 nor 0x0002, an 2504 error RESPONSE of class 'Signaling session error' (0x6) with response 2505 code 'Unknown policy rule action' (0x05) MUST be generated. 2507 The 'sub_ports' field contains the number of contiguous transport 2508 layer ports to which this rule applies. The default value of this 2509 field is 0, i.e., only the port specified in the NTLP's MRM or 2510 NATFW_DTINFO object is used for the policy rule. A value of 1 2511 indicates that additionally to the port specified in the NTLP's MRM 2512 (port1), a second port (port2) is used. This value of port 2 is 2513 calculated as: port2 = port1 + 1. Other values than 0 or 1 MUST NOT 2514 be used in this field and an error RESPONSE of class 'Signaling 2515 session error' (0x6) with response code 'Requested value in sub_ports 2516 field in NATFW_EFI not permitted' (0x08) MUST be generated. Further 2517 version of this memo may allow other values for the 'sub_ports' 2518 field. This two contiguous port numbered ports, can be used by 2519 legacy voice over IP equipment. This legacy equipment assumes that 2520 two adjacent port numbers for its RTP/RTCP flows respectively. 2522 4.2.4. Information Code Object 2524 This object carries the response code, which may be indications for 2525 either a successful or failed CREATE or EXT message depending on the 2526 value of the 'response code' field. 2528 Type: NATFW_INFO (IANA-TBD) 2530 Length: 1 2532 0 1 2 3 2533 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2534 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2535 | Resv. | Class | Response Code |r|r|r|r| Object Type | 2536 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2538 Figure 25: Information Code Object 2540 The field 'resv.' is reserved for future extensions and MUST be set 2541 to zero when generating such an object and MUST be ignored when 2542 receiving. The 'Object Type' field contains the type of the object 2543 causing the error. The value of 'Object Type' is set to 0, if no 2544 object is concerned and the leading fours bits marked with 'r' are 2545 always set to zero and ignored. The 4 bit class field contains the 2546 severity class. The following classes are defined: 2548 o 0x1: Informational (NOTIFY only) 2550 o 0x2: Success 2552 o 0x3: Protocol error 2554 o 0x4: Transient failure 2556 o 0x5: Permanent failure 2558 o 0x6: Signaling session failures 2560 Within each severity class a number of responses values are defined 2561 o Informational: 2563 * 0x01: Route change: possible route change on the outbound path. 2565 * 0x02: Re-authentication required. 2567 * 0x03: NATFW node is going down soon. 2569 o Success: 2571 * 0x01: All successfully processed. 2573 o Protocol error: 2575 * 0x01: Illegal message type: the type given in the Message Type 2576 field of the NSLP header is unknown. 2578 * 0x02: Wrong message length: the length given for the message in 2579 the NSLP header does not match the length of the message data. 2581 * 0x03: Bad flags value: an undefined flag or combination of 2582 flags was set in the NSLP header. 2584 * 0x04: Mandatory object missing: an object required in a message 2585 of this type was missing. 2587 * 0x05: Illegal object present: an object was present which must 2588 not be used in a message of this type. 2590 * 0x06: Unknown object present: an object of an unknown type was 2591 present in the message. 2593 * 0x07: Wrong object length: the length given for the object in 2594 the object header did not match the length of the object data 2595 present. 2597 * 0x08: Unknown object field value: a field in an object had an 2598 unknown value. 2600 * 0x09: Invalid Flag-Field combination: An object contains an 2601 invalid combination of flags and/or fields. 2603 * 0x0A: Duplicate object present. 2605 * 0x0B: Received EXT request message on external side. 2607 o Transient failure: 2609 * 0x01: Requested resources temporarily not available. 2611 o Permanent failure: 2613 * 0x01: Authentication failed. 2615 * 0x02: Authorization failed. 2617 * 0x03: Unable to agree transport security with peer. 2619 * 0x04: Internal or system error. 2621 * 0x05: No NAT here. 2623 * 0x06: No edge-device here. 2625 * 0x07: Did not reach the NR. 2627 o Signaling session failures: 2629 * 0x01: Session terminated asynchronously. 2631 * 0x02: Requested lifetime is too big. 2633 * 0x03: No reservation found matching the MRI of the CREATE 2634 request. 2636 * 0x04: Requested policy rule denied due to policy conflict. 2638 * 0x05: Unknown policy rule action. 2640 * 0x06: Requested rule action not applicable. 2642 * 0x07: NATFW_DTINFO object is required. 2644 * 0x08: Requested value in sub_ports field in NATFW_EFI not 2645 permitted. 2647 * 0x09: Requested IP protocol not supported. 2649 * 0x0A: Plain IP policy rules not permitted -- need transport 2650 layer information. 2652 * 0x0B: ICMP type value not permitted. 2654 * 0x0C: source IP address range is too large. 2656 * 0x0D: destination IP address range is too large. 2658 * 0x0E: source L4-port range is too large. 2660 * 0x0F: destination L4-port range is too large. 2662 * 0x10: Requested lifetime is too small. 2664 4.2.5. Nonce Object 2666 This object carries the nonce value as described in Section 3.7.6. 2668 Type: NATFW_NONCE (IANA-TBD) 2670 Length: 1 2672 0 1 2 3 2673 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2674 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2675 | nonce | 2676 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2678 Figure 26: Nonce Object 2680 4.2.6. Message Sequence Number Object 2682 This object carries the MSN value as described in Section 3.5. 2684 Type: NATFW_MSN (IANA-TBD) 2686 Length: 1 2688 0 1 2 3 2689 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2690 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2691 | message sequence number | 2692 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2694 Figure 27: Message Sequence Number Object 2696 4.2.7. Data Terminal Information Object 2698 The 'data terminal information' object carries additional information 2699 possibly needed during EXT operations. EXT messages are transported 2700 by the NTLP using the Loose-End message routing method (LE-MRM). The 2701 LE-MRM contains only DR's IP address and a signaling destination 2702 address (destination address). This destination address is used for 2703 message routing only and is not necessarily reflecting the address of 2704 the data sender. This object contains information about (if 2705 applicable) DR's port number (the destination port number), DS' port 2706 number (the source port number), the used transport protocol, the 2707 prefix length of the IP address, and DS' IP address. 2709 Type: NATFW_DTINFO (IANA-TBD) 2711 Length: variable. Maximum 3. 2713 0 1 2 3 2714 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2715 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2716 |I|P|S| reserved | sender prefix | protocol | 2717 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2718 : DR port number | DS port number : 2719 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2720 : IPsec SPI : 2721 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2722 | data sender's IPv4 address | 2723 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2725 Figure 28: Data Terminal IPv4 Address Object 2727 The flags are: 2729 o I: I=1 means that 'protocol' should be interpreted. 2731 o P: P=1 means that 'dst port number' and 'src port number' are 2732 present and should be interpreted. 2734 o S: S=1 means that SPI is present and should be interpreted. 2736 The SPI field is only present if S is set. The port numbers are only 2737 present if P is set. The flags P and S MUST NOT be set at the same 2738 time. An error RESPONSE of class 'Protocol error' (0x3) with 2739 response code 'Invalid Flag-Field combination' (0x09) MUST be 2740 generated if they are both set. If either P or S is set, I MUST be 2741 set as well and the protocol field MUST carry the particular 2742 protocol. An error RESPONSE of class 'Protocol error' (0x3) with 2743 response code 'Invalid Flag-Field combination' (0x09) MUST be 2744 generated if S or P is set but I is not set. 2746 The fields MUST be interpreted according these rules: 2748 o (data) sender prefix: This parameter indicates the prefix length 2749 of the 'data sender's IP address' in bits. For instance, a full 2750 IPv4 address requires 'sender prefix' to be set to 32. A value of 2751 0 indicates an IP address wildcard. 2753 o protocol: The IPv4 protocol field. This field MUST be interpreted 2754 if I=1, otherwise it MUST be set to 0 and MUST be ignored. 2756 o DR port number: The port number at the data receiver (DR), i.e., 2757 the destination port. A value of 0 indicates a port wildcard, 2758 i.e., the destination port number is not known. Any other value 2759 indicates the destination port number. 2761 o DS port number: The port number at the data sender (DS), i.e., the 2762 source port. A value of 0 indicates a port wildcard, i.e., the 2763 source port number is not known. Any other value indicates the 2764 source port number. 2766 o data sender's IPv4 address: The source IP address of the data 2767 sender. This field MUST be set to zero if no IP address is 2768 provided, i.e., a complete wildcard is desired (see dest prefix 2769 field above). 2771 4.2.8. ICMP Types Object 2773 The 'ICMP types' object contains additional information needed to 2774 configure a NAT of firewall with rules to control ICMP traffic. The 2775 object contains a number of values of the ICMP Type field for which a 2776 filter action should be set up: 2778 Type: NATFW_ICMP_TYPES (IANA-TBD) 2780 Length: Variable = ((Number of Types carried + 1) + 3) DIV 4 2782 Where DIV is an integer division. 2784 0 1 2 3 2785 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2786 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2787 | Count | Type | Type | ........ | 2788 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2789 | ................ | 2790 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2791 | ........ | Type | (Padding) | 2792 +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ 2794 Figure 29: ICMP Types Object 2796 The fields MUST be interpreted according these rules: 2798 count: 8 bit integer specifying the number of 'Type' entries in 2799 the object. 2801 type: 8 bit field specifying an ICMP Type value to which this rule 2802 applies. 2804 padding: Sufficient 0 bits to pad out the last word so that the 2805 total size of object is an even multiple of words. Ignored on 2806 reception. 2808 4.3. Message Formats 2810 This section defines the content of each NATFW NSLP message type. 2811 The message types are defined in Section 4.1. 2813 Basically, each message is constructed of NSLP header and one or more 2814 NSLP objects. The order of objects is not defined, meaning that 2815 objects may occur in any sequence. Objects are marked either with 2816 mandatory (M) or optional (O). Where (M) implies that this 2817 particular object MUST be included within the message and where (O) 2818 implies that this particular object is OPTIONAL within the message. 2819 Objects defined in this memo carry always the flag combination AB=00 2820 in the NSLP object header. An error RESPONSE message of class 2821 'Protocol error' (0x3) with response code 'Mandatory object missing' 2822 (0x02) MUST be generated if a mandatory declared object is missing. 2823 An error RESPONSE message of class 'Protocol error' (0x3) with 2824 response code 'Illegal object present' (0x05) MUST be generated if an 2825 object was present which must not be used in a message of this type. 2826 An error RESPONSE message of class 'Protocol error' (0x3) with 2827 response code 'Duplicate object present' (0x0A) MUST be generated if 2828 an object appears more than once in a message. 2830 Each section elaborates the required settings and parameters to be 2831 set by the NSLP for the NTLP, for instance, how the message routing 2832 information is set. 2834 4.3.1. CREATE 2836 The CREATE message is used to create NATFW NSLP signaling sessions 2837 and to create policy rules. Furthermore, CREATE messages are used to 2838 refresh NATFW NSLP signaling sessions and to delete them. 2840 The CREATE message carries these objects: 2842 o Signaling Session Lifetime object (M) 2844 o Extended flow information object (M) 2846 o Message sequence number object (M) 2848 o Nonce object (M) if P flag set to 1 in the NSLP header, otherwise 2849 (O) 2851 o ICMP Types Object (O) 2853 The message routing information in the NTLP MUST be set to DS as 2854 source address and DR as destination address. All other parameters 2855 MUST be set according the required policy rule. CREATE messages MUST 2856 be transported by using the path-coupled MRM with direction set to 2857 'downstream' (outbound). 2859 4.3.2. EXTERNAL (EXT) 2861 The EXTERNAL (EXT) message is used to a) reserve an external IP 2862 address/port at NATs, b) to notify firewalls about NSIS capable DRs, 2863 or c) to block incoming data traffic at inbound firewalls. 2865 The EXT message carries these objects: 2867 o Signaling Session Lifetime object (M) 2869 o Message sequence number object (M) 2871 o Extended flow information object (M) 2873 o Data terminal information object (M) 2875 o Nonce object [M if P flag set to 1 in the NSLP header, otherwise 2876 (O) 2878 o ICMP Types Object (O) 2879 The selected message routing method of the EXT message depends on a 2880 number of considerations. Section 3.7.2 describes it exhaustively 2881 how to select the correct method. EXT messages can be transported 2882 via the path-coupled message routing method (PC-MRM) or via the 2883 loose-end message routing method (LE-MRM). In the case of PC-MRM, 2884 the source-address is set to DS' address and the destination address 2885 is set to DR's address, the direction is set to inbound. In the case 2886 of LE-MRM, the destination-address is set to DR's address or to the 2887 signaling destination address. The source-address is set to DS's 2888 address. 2890 4.3.3. RESPONSE 2892 RESPONSE messages are responses to CREATE and EXT messages. RESPONSE 2893 messages MUST NOT be generated for any other message, such as NOTIFY 2894 and RESPONSE. 2896 The RESPONSE message for the class 'Success' (0x2) carries these 2897 objects: 2899 o Signaling Session Lifetime object (M) 2901 o Message sequence number object (M) 2903 o Information code object (M) 2905 o External address object (O) 2907 The RESPONSE message for other classes than 'Success' (0x2) carries 2908 these objects: 2910 o Message sequence number object (M) 2912 o Information code object (M) 2914 This message is routed towards the NI hop-by-hop, using existing NTLP 2915 messaging associations. The MRM used for this message MUST be the 2916 same as MRM used by the corresponding CREATE or EXT message. 2918 4.3.4. NOTIFY 2920 The NOTIFY messages is used to report asynchronous events happening 2921 along the signaled path to other NATFW NSLP nodes. 2923 The NOTIFY message carries this object: 2925 o Information code object (M). 2927 The NOTIFY message is routed towards the NI hop-by-hop using the 2928 existing inbound node messaging association entry within the node's 2929 Message Routing State table. The MRM used for this message MUST be 2930 the same as MRM used by the corresponding CREATE or EXT message. 2932 5. Security Considerations 2934 Security is of major concern particularly in case of firewall 2935 traversal. This section provides security considerations for the 2936 NAT/firewall traversal and is organized as follows. 2938 In Section 5.1 we describe the participating entities relate to each 2939 other from a security point of view. This subsection also motivates 2940 a particular authorization model. 2942 Security threats that focus on NSIS in general are described in [8] 2943 and they are applicable to this document as well. 2945 Finally, we illustrate how the security requirements that were 2946 created based on the security threats can be fulfilled by specific 2947 security mechanisms. These aspects will be elaborated in 2948 Section 5.2. 2950 5.1. Authorization Framework 2952 The NATFW NSLP is a protocol which may involve a number of NSIS nodes 2953 and is, as such, not a two-party protocol. Figure 1 and Figure 2 of 2954 [8] already depict the possible set of communication patterns. In 2955 this section we will re-evaluate these communication patters with 2956 respect to the NATFW NSLP protocol interaction. 2958 The security solutions for providing authorization have a direct 2959 impact on the treatment of different NSLPs. As it can be seen from 2960 the QoS NSLP [6] and the corresponding Diameter QoS work [19] 2961 accounting and charging seems to play an important role for QoS 2962 reservations, whereas monetary aspects might only indirectly effect 2963 authorization decisions for NAT and firewall signaling. Hence, there 2964 are differences in the semantic of authorization handling between QoS 2965 and NATFW signaling. A NATFW aware node will most likely want to 2966 authorize the entity (e.g., user or machine) requesting the 2967 establishment of pinholes or NAT bindings. The outcome of the 2968 authorization decision is either allowed or disallowed whereas a QoS 2969 authorization decision might indicate that a different set of QoS 2970 parameters is authorization (see [19] as an example). 2972 5.1.1. Peer-to-Peer Relationship 2974 Starting with the simplest scenario, it is assumed that neighboring 2975 nodes are able to authenticate each other and to establish keying 2976 material to protect the signaling message communication. The nodes 2977 will have to authorize each other, additionally to the 2978 authentication. We use the term 'Security Context' as a placeholder 2979 for referring to the entire security procedure, the necessary 2980 infrastructure that needs to be in place in order for this to work 2981 (e.g., key management) and the established security related state. 2982 The required long-term key (symmetric or asymmetric keys) used for 2983 authentication are either made available using an out-of-band 2984 mechanism between the two NSIS NATFW nodes or they are dynamically 2985 established using mechanisms not further specified in this document. 2986 Note that the deployment environment will most likely have an impact 2987 on the choice of credentials being used. The choice of these 2988 credentials used is also outside the scope of this document. 2990 +------------------------+ +-------------------------+ 2991 |Network A | | Network B| 2992 | +---------+ +---------+ | 2993 | +-///-+ Middle- +---///////----+ Middle- +-///-+ | 2994 | | | box 1 | Security | box 2 | | | 2995 | | +---------+ Context +---------+ | | 2996 | | Security | | Security | | 2997 | | Context | | Context | | 2998 | | | | | | 2999 | +--+---+ | | +--+---+ | 3000 | | Host | | | | Host | | 3001 | | A | | | | B | | 3002 | +------+ | | +------+ | 3003 +------------------------+ +-------------------------+ 3005 Figure 30: Peer-to-Peer Relationship 3007 Figure 30 shows a possible relationship between participating NSIS 3008 aware nodes. Host A might be, for example, a host in an enterprise 3009 network that has keying material established (e.g., a shared secret) 3010 with the company's firewall (Middlebox 1). The network administrator 3011 of Network A (company network) has created access control lists for 3012 Host A (or whatever identifiers a particular company wants to use). 3013 Exactly the same procedure might also be used between Host B and 3014 Middlebox 2 in Network B. For the communication between Middlebox 1 3015 and Middlebox 2 a security context is also assumed in order to allow 3016 authentication, authorization and signaling message protection to be 3017 successful. 3019 5.1.2. Intra-Domain Relationship 3021 In larger corporations, for example, a middlebox is used to protect 3022 individual departments. In many cases, the entire enterprise is 3023 controlled by a single (or a small number of) security department, 3024 which gives instructions to the department administrators. In such a 3025 scenario, the previously discussed peer-to-peer relationship might be 3026 prevalent. Sometimes it might be necessary to preserve 3027 authentication and authorization information within the network. As 3028 a possible solution, a centralized approach could be used, whereby an 3029 interaction between the individual middleboxes and a central entity 3030 (for example a policy decision point - PDP) takes place. As an 3031 alternative, individual middleboxes exchange the authorization 3032 decision with another middlebox within the same trust domain. 3033 Individual middleboxes within an administrative domain may exploit 3034 their relationship instead of requesting authentication and 3035 authorization of the signaling initiator again and again. Figure 31 3036 illustrates a network structure which uses a centralized entity. 3038 +-----------------------------------------------------------+ 3039 | Network A | 3040 | +---------+ +---------+ 3041 | +----///--------+ Middle- +------///------++ Middle- +--- 3042 | | Security | box 2 | Security | box 2 | 3043 | | Context +----+----+ Context +----+----+ 3044 | +----+----+ | | | 3045 | | Middle- +--------+ +---------+ | | 3046 | | box 1 | | | | | 3047 | +----+----+ | | | | 3048 | | Security | +----+-----+ | | 3049 | | Context | | Policy | | | 3050 | +--+---+ +-----------+ Decision +----------+ | 3051 | | Host | | Point | | 3052 | | A | +----------+ | 3053 | +------+ | 3054 +-----------------------------------------------------------+ 3056 Figure 31: Intra-domain Relationship 3058 The interaction between individual middleboxes and a policy decision 3059 point (or AAA server) is outside the scope of this document. 3061 5.1.3. End-to-Middle Relationship 3063 The peer-to-peer relationship between neighboring NSIS NATFW NSLP 3064 nodes might not always be sufficient. Network B might require 3065 additional authorization of the signaling message initiator (in 3066 addition to the authorization of the neighboring node). If 3067 authentication and authorization information is not attached to the 3068 initial signaling message then the signaling message arriving at 3069 Middlebox 2 would result in an error message being created, which 3070 indicates the additional authorization requirement. In many cases 3071 the signaling message initiator might already be aware of the 3072 additionally required authorization before the signaling message 3073 exchange is executed. 3075 Figure 32 shows this scenario. 3077 +--------------------+ +---------------------+ 3078 | Network A | |Network B | 3079 | | Security | | 3080 | +---------+ Context +---------+ | 3081 | +-///-+ Middle- +---///////----+ Middle- +-///-+ | 3082 | | | box 1 | +-------+ box 2 | | | 3083 | | +---------+ | +---------+ | | 3084 | |Security | | | Security | | 3085 | |Context | | | Context | 3086 | | | | | | | 3087 | +--+---+ | | | +--+---+ | 3088 | | Host +----///----+------+ | | Host | | 3089 | | A | | Security | | B | | 3090 | +------+ | Context | +------+ | 3091 +--------------------+ +---------------------+ 3093 Figure 32: End-to-Middle Relationship 3095 5.2. Security Framework for the NAT/Firewall NSLP 3097 The following list of security requirements has been created to 3098 ensure proper secure operation of the NATFW NSLP. 3100 5.2.1. Security Protection between neighboring NATFW NSLP Nodes 3102 Based on the analyzed threats it is RECOMMENDED to provide, between 3103 neighboring NATFW NSLP nodes, the following mechanism: 3105 o data origin authentication 3107 o replay protection 3109 o integrity protection and 3111 o optionally confidentiality protection 3113 It is RECOMMENDED to use the authentication and key exchange security 3114 mechanisms provided in [2] between neighboring nodes when sending 3115 NATFW signaling messages. The proposed security mechanisms of GIST 3116 provide support for authentication and key exchange in addition to 3117 denial of service protection. Depending on the chosen security 3118 protocol, support for multiple authentication protocols might be 3119 provided. If security between neighboring nodes is desired than the 3120 usage of C-MODE for the delivery of data packets and the usage of 3121 D-MODE only to discover the next NATFW NSLP aware node along the path 3122 is highly RECOMMENDED. Almost all security threats at the NATFW NSLP 3123 layer can be prevented by using a mutually authenticated Transport 3124 Layer secured connection and by relying on authorization by the 3125 neighboring NATFW NSLP entities. 3127 The NATFW NSLP relies on an established security association between 3128 neighboring peers to prevent unauthorized nodes to modify or delete 3129 installed state. Between non-neighboring nodes the session ID (SID) 3130 carried in the NTLP is used to show ownership of a NATFW NSLP 3131 signaling session. The session ID MUST be generated in a random way 3132 and thereby prevent an off-path adversary to mount targeted attacks. 3133 Hence, an adversary would have to learn the randomly generated 3134 session ID to perform an attack. In a mobility environment a former 3135 on-path node that is now off-path can perform an attack. Messages 3136 for a particular NATFW NSLP signaling session are handled by the NTLP 3137 to the NATFW NSLP for further processing. Messages carrying a 3138 different session ID not associated with any NATFW NSLP are subject 3139 to the regular processing for new NATFW NSLP signaling sessions. 3141 5.2.2. Security Protection between non-neighboring NATFW NSLP Nodes 3143 Based on the security threats and the listed requirements it was 3144 noted that some threats also demand authentication and authorization 3145 of a NATFW signaling entity (including the initiator) towards a non- 3146 neighboring node. This mechanism mainly demands entity 3147 authentication. Additionally, security protection of certain 3148 payloads may be required between non-neighboring signaling entities 3149 and the Cryptographic Message Syntax (CMS) [14] migh be a potential 3150 solution. Payload protection using CMS is not described in this 3151 document. The most important information exchanged at the NATFW NSLP 3152 is information related to the establishment for firewall pinholes and 3153 NAT bindings. This information can, however, not be protected over 3154 multiple NSIS NATFW NSLP hops since this information might change 3155 depending on the capability of each individual NATFW NSLP node. 3157 Some scenarios might also benefit from the usage of authorization 3158 tokens. Their purpose is to associate two different signaling 3159 protocols (e.g., SIP and NSIS) and their authorization decision. 3160 These tokens are obtained by non-NSIS protocols, such as SIP or as 3161 part of network access authentication. When a NAT or firewall along 3162 the path receives the token it might be verified locally or passed to 3163 the AAA infrastructure. Examples of authorization tokens can be 3164 found in RFC 3520 [17] and RFC 3521 [18]. Figure 33 shows an example 3165 of this protocol interaction. 3167 An authorization token is provided by the SIP proxy, which acts as 3168 the assertion generating entity and gets delivered to the end host 3169 with proper authentication and authorization. When the NATFW 3170 signaling message is transmitted towards the network, the 3171 authorization token is attached to the signaling messages to refer to 3172 the previous authorization decision. The assertion verifying entity 3173 needs to process the token or it might be necessary to interact with 3174 the assertion granting entity using HTTP (or other protocols). As a 3175 result of a successfully authorization by a NATFW NSLP node, the 3176 requested action is executed and later a RESPONSE message is 3177 generated. 3179 +----------------+ Trust Relationship +----------------+ 3180 | +------------+ |<.......................>| +------------+ | 3181 | | Protocol | | | | Assertion | | 3182 | | requesting | | HTTP, SIP Request | | Granting | | 3183 | | authz | |------------------------>| | Entity | | 3184 | | assertions | |<------------------------| +------------+ | 3185 | +------------+ | Artifact/Assertion | Entity Cecil | 3186 | ^ | +----------------+ 3187 | | | ^ ^| 3188 | | | . || HTTP, 3189 | | | Trust . || other 3190 | API Access | Relationship. || protocols 3191 | | | . || 3192 | | | . || 3193 | | | v |v 3194 | v | +----------------+ 3195 | +------------+ | | +------------+ | 3196 | | Protocol | | NSIS NATFW CREATE + | | Assertion | | 3197 | | using authz| | Assertion/Artifact | | Verifying | | 3198 | | assertion | | ----------------------- | | Entity | | 3199 | +------------+ | | +------------+ | 3200 | Entity Alice | <---------------------- | Entity Bob | 3201 +----------------+ RESPONSE +----------------+ 3203 Figure 33: Authorization Token Usage 3205 Threats against the usage of authorization tokens have been mentioned 3206 in [8]. Hence, it is required to provide confidentiality protection 3207 to avoid allowing an eavesdropper to learn the token and to use it in 3208 another NATFW NSLP signaling session (replay attack). The token 3209 itself also needs to be protected against tempering. 3211 To harmonize the usage of authorization tokens in NSLPs a separate 3212 document is available, see [20]. 3214 6. IAB Considerations on UNSAF 3216 UNilateral Self-Address Fixing (UNSAF) is described in [12] as a 3217 process at originating endpoints that attempt to determine or fix the 3218 address (and port) by which they are known to another endpoint. 3219 UNSAF proposals, such as STUN [15] are considered as a general class 3220 of workarounds for NAT traversal and as solutions for scenarios with 3221 no middlebox communication. 3223 This memo specifies a path-coupled middlebox communication protocol, 3224 i.e., the NSIS NATFW NSLP. NSIS in general and the NATFW NSLP are 3225 not intended as a short-term workaround, but more as a long-term 3226 solution for middlebox communication. In NSIS, endpoints are 3227 involved in allocating, maintaining, and deleting addresses and ports 3228 at the middlebox. However, the full control of addresses and ports 3229 at the middlebox is at the NATFW NSLP daemon located to the 3230 respective NAT. 3232 Therefore, this document addresses the UNSAF considerations in [12] 3233 by proposing a long-term alternative solution. 3235 7. IANA Considerations 3237 This section provides guidance to the Internet Assigned Numbers 3238 Authority (IANA) regarding registration of values related to the 3239 NATFW NSLP, in accordance with BCP 26 RFC 2434 [13]. 3241 The NATFW NSLP requires IANA to create a number of new registries. 3242 These registries may require further coordination with the registries 3243 of the NTLP [2] and the QoS NSLP [6]. 3245 NATFW NSLP Message Type Registry 3247 The NATFW NSLP Message Type is a 8 bit value. The allocation of 3248 values for new message types requires standards action. Updates and 3249 deletion of values from the registry is not possible. This 3250 specification defines four NATFW NSLP message types, which form the 3251 initial contents of this registry. IANA is requested to add these 3252 four NATFW NSLP Message Types: CREATE, EXT, RESPONSE, and NOTIFY. 3254 NATFW NSLP Header Flag Registry 3256 NATFW NSLP messages have a messages-specific 8 bit flags/reserved 3257 field in their header. The registration of flags is subject to IANA 3258 registration. The allocation of values for flag types requires 3259 standards action. Updates and deletion of values from the registry 3260 is not possible. This specification defines only one flag, the P 3261 flag in Figure 20. 3263 NSLP Object Type Registry 3265 [Delete this part if already done by another NSLP: 3267 A new registry is to be created for NSLP Message Objects. This is a 3268 12-bit field (giving values from 0 to 4095). This registry is shared 3269 between a number of NSLPs. Allocation policies are as follows: 3271 0-1023: Standards Action 3273 1024-1999: Specification Required 3275 2000-2047: Private/Experimental Use 3277 2048-4095: Reserved 3279 When a new object is defined, the extensbility bits (A/B) must also 3280 be defined.] 3282 This document defines 8 objects for the NATFW NSLP: NATFW_LT, 3283 NATFW_EXT_IP, NATFW_EFI, NATFW_INFO, NATFW_NONCE, NATFW_MSN, 3284 NATFW_DTINFO, NATFW_ICMP_TYPES. IANA is request to assigned values 3285 for them from NSLP Object Type registry and to replace the 3286 corresponding IANA-TBD tags with the numeric values. 3288 NSLP Response Code Registry 3290 In addition it defines a number of Response Codes for the NATFW NSLP. 3291 These can be found in Section 4.2.4 and are to be assigned values 3292 from NSLP Response Code registry. The allocation of values for 3293 Response Codes Codes requires standards action. IANA is request to 3294 assigned values for them from NSLP Response Code registry. 3296 Furthermore, IANA is requested to add a new value to the NSLP 3297 Identifiers (NSLPID) registry defined in [2] for the NATFW NSLP. 3299 8. Open Issues 3301 A more detailed list of open issue can be found at: 3302 https://kobe.netlab.nec.de/roundup/nsis-natfw-nslp/index 3304 9. Acknowledgments 3306 We would like to thank the following individuals for their 3307 contributions to this document at different stages: 3309 o Marcus Brunner and Henning Schulzrinne for work on work on IETF 3310 drafts which lead us to start with this document, 3312 o Miquel Martin for his help on the initial version of this 3313 document, 3315 o Srinath Thiruvengadam and Ali Fessi work for their work on the 3316 NAT/firewall threats draft, 3318 o Henning Peters for his comments and suggestions, 3320 o and the NSIS working group. 3322 10. References 3324 10.1. Normative References 3326 [1] Bradner, S., "Key words for use in RFCs to Indicate Requirement 3327 Levels", BCP 14, RFC 2119, March 1997. 3329 [2] Schulzrinne, H. and R. Hancock, "GIST: General Internet 3330 Signaling Transport", draft-ietf-nsis-ntlp-11 (work in 3331 progress), August 2006. 3333 [3] Elz, R. and R. Bush, "Serial Number Arithmetic", RFC 1982, 3334 August 1996. 3336 10.2. Informative References 3338 [4] Hancock, R., Karagiannis, G., Loughney, J., and S. Van den 3339 Bosch, "Next Steps in Signaling (NSIS): Framework", RFC 4080, 3340 June 2005. 3342 [5] Brunner, M., "Requirements for Signaling Protocols", RFC 3726, 3343 April 2004. 3345 [6] Manner, J., "NSLP for Quality-of-Service Signaling", 3346 draft-ietf-nsis-qos-nslp-12 (work in progress), October 2006. 3348 [7] Srisuresh, P., Kuthan, J., Rosenberg, J., Molitor, A., and A. 3349 Rayhan, "Middlebox communication architecture and framework", 3350 RFC 3303, August 2002. 3352 [8] Tschofenig, H. and D. Kroeselberg, "Security Threats for Next 3353 Steps in Signaling (NSIS)", RFC 4081, June 2005. 3355 [9] Srisuresh, P. and M. Holdrege, "IP Network Address Translator 3356 (NAT) Terminology and Considerations", RFC 2663, August 1999. 3358 [10] Carpenter, B. and S. Brim, "Middleboxes: Taxonomy and Issues", 3359 RFC 3234, February 2002. 3361 [11] Braden, B., Zhang, L., Berson, S., Herzog, S., and S. Jamin, 3362 "Resource ReSerVation Protocol (RSVP) -- Version 1 Functional 3363 Specification", RFC 2205, September 1997. 3365 [12] Daigle, L. and IAB, "IAB Considerations for UNilateral Self- 3366 Address Fixing (UNSAF) Across Network Address Translation", 3367 RFC 3424, November 2002. 3369 [13] Narten, T. and H. Alvestrand, "Guidelines for Writing an IANA 3370 Considerations Section in RFCs", BCP 26, RFC 2434, 3371 October 1998. 3373 [14] Housley, R., "Cryptographic Message Syntax (CMS)", RFC 3852, 3374 July 2004. 3376 [15] Rosenberg, J., Weinberger, J., Huitema, C., and R. Mahy, "STUN 3377 - Simple Traversal of User Datagram Protocol (UDP) Through 3378 Network Address Translators (NATs)", RFC 3489, March 2003. 3380 [16] Westerinen, A., Schnizlein, J., Strassner, J., Scherling, M., 3381 Quinn, B., Herzog, S., Huynh, A., Carlson, M., Perry, J., and 3382 S. Waldbusser, "Terminology for Policy-Based Management", 3383 RFC 3198, November 2001. 3385 [17] Hamer, L-N., Gage, B., Kosinski, B., and H. Shieh, "Session 3386 Authorization Policy Element", RFC 3520, April 2003. 3388 [18] Hamer, L-N., Gage, B., and H. Shieh, "Framework for Session 3389 Set-up with Media Authorization", RFC 3521, April 2003. 3391 [19] Zorn, G., "Diameter Quality of Service Application", 3392 draft-ietf-dime-diameter-qos-00 (work in progress), 3393 February 2007. 3395 [20] Manner, J., "Authorization for NSIS Signaling Layer Protocols", 3396 draft-manner-nsis-nslp-auth-02 (work in progress), 3397 October 2006. 3399 [21] Roedig, U., Goertz, M., Karten, M., and R. Steinmetz, "RSVP as 3400 firewall Signalling Protocol", Proceedings of the 6th IEEE 3401 Symposium on Computers and Communications, Hammamet, 3402 Tunisia pp. 57 to 62, IEEE Computer Society Press, July 2001. 3404 Appendix A. Selecting Signaling Destination Addresses for EXT 3406 As with all other message types, EXT messages need a reachable IP 3407 address of the data sender on the GIST level. For the path-coupled 3408 MRM the source-address of GIST is the reachable IP address (i.e., the 3409 real IP address of the data sender, or a wildcard). While this is 3410 straight forward, it is not necessarily so for the loose-end MRM. 3411 Many applications do not provide the IP address of the communication 3412 counterpart, i.e., either the data sender or both a data sender and 3413 receiver. For the EXT messages, the case of data sender is of 3414 interest only. The rest of this section is giving informational 3415 guidance about determining a good destination-address of the LE-MRM 3416 in GIST for EXT messages. 3418 This signaling destination address (SDA, the destination-address in 3419 GIST) can be the data sender, but for applications which do not 3420 provide an address upfront, the destination address has to be chosen 3421 independently, as it is unknown at the time when the NATFW NSLP 3422 signaling has to start. Choosing the 'correct' destination IP 3423 address may be difficult and it is possible that there is no 'right 3424 answer' for all applications relying on the NATFW NSLP. 3426 Whenever possible it is RECOMMENDED to chose the data sender's IP 3427 address as SDA. It necessary to differentiate between the received 3428 IP addresses on the data sender. Some application level signaling 3429 protocols (e.g., SIP) have the ability to transfer multiple contact 3430 IP addresses of the data sender. For instance, private IP address, 3431 public IP address at NAT, and public IP address at a relay. It is 3432 RECOMMENDED to use all non-private IP addresses as SDAs. 3434 A different SDA must be chosen, should the IP address of the data 3435 sender be unknown. This can have multiple reasons: The application 3436 level signaling protocol cannot determine any data sender IP address 3437 at this point of time or the data receiver is server behind a NAT, 3438 i.e., accepting inbound packets from any host. In this case, the 3439 NATFW NSLP can be instructed to use the public IP address of an 3440 application server or any other node. Choosing the SDA in this case 3441 is out of the scope of the NATFW NSLP and depends on the 3442 application's choice. The local network can provide a network-SDA, 3443 i.e., a SDA which is only meaningful to the local network. This will 3444 ensure that GIST packets with destination-address set to this 3445 network-SDA are going to be routed to a edge-NAT or edge-firewall. 3447 Appendix B. Applicability Statement on Data Receivers behind Firewalls 3449 Section 3.7.2 describes how data receivers behind middleboxes can 3450 instruct inbound firewalls/NATs to forward NATFW NSLP signaling 3451 towards them. Finding an inbound edge-NAT in address environment 3452 with NAT'ed addresses is quite easy. It is only required to find 3453 some edge-NAT, as the data traffic will be route-pinned to the NAT, 3454 which is done with the LE-MRM. Locating the appropriate edge- 3455 firewall with the PC-MRM, sent inbound is difficult. For cases with 3456 a single, symmetric route from the Internet to the data receiver, it 3457 is quite easy; simply follow the default route in the inbound 3458 direction. 3460 +------+ Data Flow 3461 +-------| EFW1 +----------+ <=========== 3462 | +------+ ,--+--. 3463 +--+--+ / \ 3464 NI+-----| FW1 | (Internet )----NR+/NI/DS 3465 NR +--+--+ \ / 3466 | +------+ `--+--' 3467 +-------| EFW2 +----------+ 3468 +------+ 3470 ~~~~~~~~~~~~~~~~~~~~~> 3471 Signaling Flow 3473 Figure 34: Data receiver behind multiple, parallel located firewalls 3475 When a data receiver, and thus NR, is located in a network site that 3476 is multihomed with several independently firewalled connections to 3477 the public Internet (as shown in Figure 34), the specific firewall 3478 through which the data traffic will be routed has to be ascertained. 3479 NATFW NSLP signaling messages sent from the NI+/NR during the EXT 3480 message exchange towards the NR+ must be routed by the NTLP to the 3481 edge-firewall that will be passed by the data traffic as well. The 3482 NTLP would need to be aware about the routing within the Internet to 3483 determine the path between DS and DR. Out of this, the NTLP could 3484 determine which of the edge-firewalls, either EFW1 or EFW2, must be 3485 selected to forward the NATFW NSLP signaling. Signaling to the wrong 3486 edge-firewall, as shown in Figure 34, would install the NATFW NSLP 3487 policy rules at the wrong device. This causes either a blocked data 3488 flow (when the policy rule is 'allow') or an ongoing attack (when the 3489 policy rule is 'deny'). Requiring the NTLP to know all about the 3490 routing within the Internet is definitely a tough challenge and 3491 usually not possible. In such described case, the NTLP must 3492 basically give up and return an error to the NSLP level, indicating 3493 that the next hop discovery is not possible. 3495 Appendix C. Firewall and NAT Resources 3497 This section gives some examples on how NATFW NSLP policy rules could 3498 be mapped to real firewall or NAT resources. The firewall rules and 3499 NAT bindings are described in a natural way, i.e., in a way one will 3500 find it in common implementation. 3502 C.1. Wildcarding of Policy Rules 3504 The policy rule/MRI to be installed can be wildcarded to some degree. 3505 Wildcarding applies to IP address, transport layer port numbers, and 3506 the IP payload (or next header in IPv6). Processing of wildcarding 3507 splits into the NTLP and the NATFW NSLP layer. The processing at the 3508 NTLP layer is independent of the NSLP layer processing and per layer 3509 constraints apply. For wildcarding in the NTLP see Section 5.8 of 3510 [2]. 3512 Wildcarding at the NATFW NSLP level is always a node local policy 3513 decision. A signaling message carrying a wildcarded MRI (and thus 3514 policy rule) arriving at an NSLP node can be rejected if the local 3515 policy does not allow the request. For instance, a MRI with IP 3516 addresses set (not wildcarded), transport protocol TCP, and TCP port 3517 numbers completely wildcarded. Now the local policy allows only 3518 requests for TCP with all ports set and not wildcarded. The request 3519 is going to be rejected. 3521 C.2. Mapping to Firewall Rules 3523 This section describes how a NSLP policy rule signaled with a CREATE 3524 message is mapped to a firewall rule. The MRI is set as follows: 3526 o network-layer-version=IPv4 3528 o source-address=192.0.2.100, prefix-length=32 3530 o destination-address=192.0.50.5, prefix-length=32 3532 o IP-protocol=UDP 3534 o L4-source-port=34543, L4-destination-port=23198 3536 The NATFW_EFI object is set to action=allow and sub_ports=0. 3538 The resulting policy rule (firewall rule) to be installed might look 3539 like: allow udp from 192.0.2.100 port=34543 to 192.0.50.5 port=23198 3541 C.3. Mapping to NAT Bindings 3543 This section describes how a NSLP policy rule signaled with a EXT 3544 message is mapped to a NAT binding. It is assumed that the EXT 3545 message is sent by a NI+ being located behind a NAT and does contain 3546 a NATFW_DTINFO object. The MRI is set following using the signaling 3547 destination address, since the IP address of the real data sender is 3548 not known: 3550 o network-layer-version=IPv4 3552 o source-address= 192.168.5.100 3554 o destination-address=SDA 3556 o IP-protocol=UDP 3558 The NATFW_EFI object is set to action=allow and sub_ports=0. The 3559 NATFW_DTINFO object contains these parameters: 3561 o P=1 3563 o dest prefix=0 3565 o protocol=UDP 3567 o dst port number = 20230, src port number=0 3569 o src IP=0.0.0.0 3571 The edge-NAT allocates the external IP 192.0.2.79 and port 45000. 3573 The resulting policy rule (NAT binding) to be installed could look 3574 like: translate from any to 192.0.2.79 port=45000 to 192.168.5.100 3575 port=20230 3577 C.4. NSLP Handling of Twice-NAT 3579 The dynamic configuration of twice-NATs requires application level 3580 support, as stated in Section 2.5. The NATFW NSLP cannot be used for 3581 configuring twice-NATs if application level support is needed. 3582 Assuming application level support performing the configuration of 3583 the twice-NAT and the NATFW NSLP being installed at this devices, the 3584 NATFW NSLP must be able to traverse it. The NSLP is probably able to 3585 traverse the twice-NAT, as any other data traffic, but the flow 3586 information stored in the NTLP's MRI will be invalidated through the 3587 translation of source and destination address. The NATFW NSLP 3588 implementation on the twice-NAT MUST intercept NATFW NSLP and NTLP 3589 signaling messages as any other NATFW NSLP node does. For the given 3590 signaling flow, the NATFW NSLP node MUST look up the corresponding IP 3591 address translation and modify the NTLP/NSLP signaling accordingly. 3592 The modification results in an updated MRI with respect to the source 3593 and destination IP addresses. 3595 Appendix D. Assigned Numbers for Testing 3597 NOTE: This section MUST be removed before publication. 3599 This section defines temporarily used values of the NATFW NSLP for 3600 testing the different implementations. 3602 Values for the NATFW NSLP message types: 3604 o CREATE: 0x01 3606 o EXT: 0x02 3608 o RESPONSE: 0x03 3610 o NOTIFY: 0x04 3612 Values for the NSLP object types 3614 o NATFW_LT: 0x00F1 3616 o NATFW_EXT_IP: 0x00F2 3618 o NATFW_EFI: 0x00F3 3620 o NATFW_INFO: 0x00F4 3622 o NATFW_NONCE: 0x00F5 3624 o NATFW_MSN: 0x00F6 3626 o NATFW_DTINFO: 0x00F7 3628 o NATFW_ICMP_TYPES: 0x00F9 3630 1345 3632 Authors' Addresses 3634 Martin Stiemerling 3635 NEC Europe Ltd. and University of Goettingen 3636 Kurfuersten-Anlage 36 3637 Heidelberg 69115 3638 Germany 3640 Phone: +49 (0) 6221 4342 113 3641 Email: stiemerling@netlab.nec.de 3642 URI: http://www.stiemerling.org 3644 Hannes Tschofenig 3645 Nokia Siemens Networks 3646 Otto-Hahn-Ring 6 3647 Munich 81739 3648 Germany 3650 Phone: 3651 Email: Hannes.Tschofenig@nsn.com 3652 URI: http://www.tschofenig.com 3654 Cedric Aoun 3655 Paris 3656 France 3658 Email: cedric@caoun.net 3660 Elwyn Davies 3661 Folly Consulting 3662 Soham 3663 UK 3665 Phone: +44 7889 488 335 3666 Email: elwynd@dial.pipex.com 3668 Full Copyright Statement 3670 Copyright (C) The IETF Trust (2007). 3672 This document is subject to the rights, licenses and restrictions 3673 contained in BCP 78, and except as set forth therein, the authors 3674 retain all their rights. 3676 This document and the information contained herein are provided on an 3677 "AS IS" basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS 3678 OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY, THE IETF TRUST AND 3679 THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS 3680 OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF 3681 THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED 3682 WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. 3684 Intellectual Property 3686 The IETF takes no position regarding the validity or scope of any 3687 Intellectual Property Rights or other rights that might be claimed to 3688 pertain to the implementation or use of the technology described in 3689 this document or the extent to which any license under such rights 3690 might or might not be available; nor does it represent that it has 3691 made any independent effort to identify any such rights. Information 3692 on the procedures with respect to rights in RFC documents can be 3693 found in BCP 78 and BCP 79. 3695 Copies of IPR disclosures made to the IETF Secretariat and any 3696 assurances of licenses to be made available, or the result of an 3697 attempt made to obtain a general license or permission for the use of 3698 such proprietary rights by implementers or users of this 3699 specification can be obtained from the IETF on-line IPR repository at 3700 http://www.ietf.org/ipr. 3702 The IETF invites any interested party to bring to its attention any 3703 copyrights, patents or patent applications, or other proprietary 3704 rights that may cover technology that may be required to implement 3705 this standard. Please address the information to the IETF at 3706 ietf-ipr@ietf.org. 3708 Acknowledgment 3710 Funding for the RFC Editor function is provided by the IETF 3711 Administrative Support Activity (IASA).