idnits 2.17.1 draft-ietf-nfsv4-rpcsec-gssv3-05.txt: Checking boilerplate required by RFC 5378 and the IETF Trust (see https://trustee.ietf.org/license-info): ---------------------------------------------------------------------------- No issues found here. 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 : ---------------------------------------------------------------------------- No issues found here. Miscellaneous warnings: ---------------------------------------------------------------------------- == The copyright year in the IETF Trust and authors Copyright Line does not match the current year == Line 314 has weird spacing: '... bool cri...' == Line 315 has weird spacing: '... opaque data<...' == Line 319 has weird spacing: '...ned int ver...' == Line 333 has weird spacing: '...ss3_lfs lfs...' == Line 351 has weird spacing: '...3_label label...' == (1 more instance...) -- The document date (October 17, 2013) is 3837 days in the past. Is this intentional? -- Found something which looks like a code comment -- if you have code sections in the document, please surround them with '' and '' lines. Checking references for intended status: Proposed Standard ---------------------------------------------------------------------------- (See RFCs 3967 and 4897 for information about using normative references to lower-maturity documents in RFCs) -- Possible downref: Non-RFC (?) normative reference: ref. '1' ** Obsolete normative reference: RFC 1831 (ref. '4') (Obsoleted by RFC 5531) == Outdated reference: A later version (-41) exists of draft-ietf-nfsv4-minorversion2-19 ** Obsolete normative reference: RFC 5226 (ref. '8') (Obsoleted by RFC 8126) == Outdated reference: A later version (-05) exists of draft-ietf-nfsv4-labreqs-03 Summary: 2 errors (**), 0 flaws (~~), 9 warnings (==), 3 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 NFSv4 W. Adamson 3 Internet-Draft NetApp 4 Intended status: Standards Track N. Williams 5 Expires: April 20, 2014 Cryptonector 6 October 17, 2013 8 Remote Procedure Call (RPC) Security Version 3 9 draft-ietf-nfsv4-rpcsec-gssv3-05.txt 11 Abstract 13 This document specifies version 3 of the Remote Procedure Call (RPC) 14 security protocol (RPCSEC_GSS). This protocol provides for compound 15 authentication of client hosts and users to server (constructed by 16 generic composition), security label assertions for multi-level and 17 type enforcement, structured privilege assertions, and channel 18 bindings. 20 Requirements Language 22 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 23 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 24 document are to be interpreted as described in RFC 2119 [1]. 26 Status of this Memo 28 This Internet-Draft is submitted in full conformance with the 29 provisions of BCP 78 and BCP 79. 31 Internet-Drafts are working documents of the Internet Engineering 32 Task Force (IETF). Note that other groups may also distribute 33 working documents as Internet-Drafts. The list of current Internet- 34 Drafts is at http://datatracker.ietf.org/drafts/current/. 36 Internet-Drafts are draft documents valid for a maximum of six months 37 and may be updated, replaced, or obsoleted by other documents at any 38 time. It is inappropriate to use Internet-Drafts as reference 39 material or to cite them other than as "work in progress." 41 This Internet-Draft will expire on April 20, 2014. 43 Copyright Notice 45 Copyright (c) 2013 IETF Trust and the persons identified as the 46 document authors. All rights reserved. 48 This document is subject to BCP 78 and the IETF Trust's Legal 49 Provisions Relating to IETF Documents 50 (http://trustee.ietf.org/license-info) in effect on the date of 51 publication of this document. Please review these documents 52 carefully, as they describe your rights and restrictions with respect 53 to this document. Code Components extracted from this document must 54 include Simplified BSD License text as described in Section 4.e of 55 the Trust Legal Provisions and are provided without warranty as 56 described in the Simplified BSD License. 58 Table of Contents 60 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3 61 1.1. Applications of RPCSEC_GSSv3 . . . . . . . . . . . . . . . 4 62 2. The RPCSEC_GSSv3 protocol . . . . . . . . . . . . . . . . . . 5 63 2.1. New auth_stat values . . . . . . . . . . . . . . . . . . . 9 64 2.2. RPC message credential and verifier . . . . . . . . . . . 10 65 2.3. Control Messages . . . . . . . . . . . . . . . . . . . . . 10 66 2.3.1. Create request . . . . . . . . . . . . . . . . . . . . 11 67 2.3.2. Destruction request . . . . . . . . . . . . . . . . . 15 68 2.3.3. List request . . . . . . . . . . . . . . . . . . . . . 16 69 2.3.4. Extensibility . . . . . . . . . . . . . . . . . . . . 16 70 2.4. Data Messages . . . . . . . . . . . . . . . . . . . . . . 17 71 3. Security Considerations . . . . . . . . . . . . . . . . . . . 17 72 4. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 18 73 5. References . . . . . . . . . . . . . . . . . . . . . . . . . . 19 74 5.1. Normative References . . . . . . . . . . . . . . . . . . . 19 75 5.2. Informative References . . . . . . . . . . . . . . . . . . 19 76 Appendix A. Acknowledgments . . . . . . . . . . . . . . . . . . . 20 77 Appendix B. RFC Editor Notes . . . . . . . . . . . . . . . . . . 20 78 Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 20 80 1. Introduction 82 The original RPCSEC_GSS protocol [2] provided for authentication of 83 RPC clients and servers to each other using the Generic Security 84 Services Application Programming Interface (GSS-API) [3]. The second 85 version of RPCSEC_GSS [4] added support for channel bindings [5]. 87 We find that GSS-API mechanisms are insufficient for communicating 88 certain aspects of a client's authority to a server. The GSS-API and 89 its mechanisms certainly could be extended to address this 90 shortcoming, but it seems be far simpler to address it at the 91 application layer, namely, in this case, RPCSEC_GSS. 93 The motivation for RPCSEC_GSSv3 is to add support for labeled 94 security and server-side copy for NFSv4 (see [6] and [9]). Both of 95 these features require assertions of authority from the client. 97 Assertions need to be verified. One party that can verify an 98 assertion is the client host, which can authenticate to the server 99 using its own credentials. We can also require users to verify an 100 assertion as well. This calls for compound authentication. 102 Because the design of RPCSEC_GSSv3 relies on either RPCSEC_GSS 103 version 1 (though version 2 can be used) to do the actual GSS-API 104 security context establishment, we add support for channel binding so 105 that implementors who have implemented RPCSEC_GSS version 1 but not 106 version 2 can provide a (simplified) channel binding implementation 107 using RPCSEC_GSSv3. 109 We therefore describe a new version of RPCSEC_GSS that allows for the 110 following: 112 o Client-side assertions of authority: 114 * Security labels for multi-level, type enforcement, and other 115 labeled security models. See [10], [11], [12], [6] and [9]. 117 * Application-specific structured privileges. For an example see 118 server-side copy [6]. 120 * Compound authentication of the client host and user to the 121 server done by binding two RPCSEC_GSS handles. 123 * Simplified channel binding. 125 Assertions of labels and privileges are evaluated by the server, 126 which may then map the asserted values to other values, all according 127 to server-side policy. 129 We also add an option for enumerating active server-side privileges 130 and supported label format specifiers (LFS). The LFS and Label 131 Format Registry are described in detail in [13]. 133 RPCSEC_GSSv3 is patterned as follows: 135 o A client uses an existing RPCSEC_GSSv1 (or RPCSEC_GSSv2) context 136 handle to protect RPCSEC_GSSv3 exchanges (this will be termed the 137 "parent" handle) 139 o The server issues a "child" RPCSEC_GSSv3 handle, but the 140 underlying GSS-API security context for the parent handle is used 141 in all subsequent exchanges using the child handle. This works 142 because the RPCSEC_GSS handle is included in the integrity 143 protected RPCSEC_GSS auth/verifier header for all versions of 144 RPCSEC_GSS. The child context, however, has its own sequence 145 number space and window, distinct from that of the parent. 147 [[Comment.1: RFC22203 states that when data integrity is used, the 148 seq_num in the rpc_gss_data_t must be the same as in the credential. 149 This means that using data integrity with GSS3 context's can not 150 simply construct it using the parent context as the seq_num must be 151 from the GSS3 context. --AA]] 153 This means that RPCSEC_GSSv3 depends on RPCSEC_GSS versions 1 and/or 154 2 for actual GSS-API security context establishment. This keeps the 155 specification of RPCSEC_GSSv3 simple by avoiding the need to 156 duplicate the core functionality of RPCSEC_GSS version 1. 158 1.1. Applications of RPCSEC_GSSv3 160 The common uses of RPCSEC_GSSv3, particularly for NFSv4 [6], are 161 expected to be: 163 a. labeled security: client-side process label assertion [+ 164 privilege assertion] + compound client host & user 165 authentication; 167 b. compound client host & user authentication [+ critical structured 168 privilege assertions] used in inter-server server-side copy; 170 Labeled NFS (see Section 8 of [6]) uses the subject label provided by 171 the client via the RPCSEC_GSSv3 layer to enforce MAC access to 172 objects owned by the server to enable server guest mode or full mode 173 labeled NFS. 175 [[Comment.2: check that this language states what NFSv4.2 labeled NFS 176 problem we are really solving. (setting labels on the server) --AA]] 177 A traditional inter-server file copy entails the user gaining access 178 to a file on the source, reading it, and writing it to a file on the 179 destination. In secure NFSv4 inter-server server-side copy (see 180 Section 3.4.1 of [6]), the user first secures access to both source 181 and destination files, and then uses RPCSEC_GSSv3 compound 182 authentication and structured privileges to authorize the destination 183 to copy the file from the source on behalf of the user. 185 2. The RPCSEC_GSSv3 protocol 187 This document contains the External Data Representation (XDR) ([7]) 188 definitions for the RPCSEC_GSSv3 protocol. 190 The XDR description is provided in this document in a way that makes 191 it simple for the reader to extract into ready to compile form. The 192 reader can feed this document in the following shell script to 193 produce the machine readable XDR description of RPCSEC_GSSv3: 195 #!/bin/sh 196 grep "^ *///" | sed 's?^ */// ??' | sed 's?^ *///$??' 198 I.e. if the above script is stored in a file called "extract.sh", and 199 this document is in a file called "spec.txt", then the reader can do: 201 sh extract.sh < spec.txt > rpcsec_gss_v3.x 203 The effect of the script is to remove leading white space from each 204 line, plus a sentinel sequence of "///". 206 The XDR description, with the sentinel sequence follows: 208 /// /* 209 /// * Copyright (c) 2013 IETF Trust and the persons 210 /// * identified as the document authors. All rights 211 /// * reserved. 212 /// * 213 /// * The document authors are identified in [RFC2203], 214 /// * [RFC5403], and [RFCxxxx]. 215 /// * 216 /// * Redistribution and use in source and binary forms, 217 /// * with or without modification, are permitted 218 /// * provided that the following conditions are met: 219 /// * 220 /// * o Redistributions of source code must retain the above 221 /// * copyright notice, this list of conditions and the 222 /// * following disclaimer. 223 /// * 224 /// * o Redistributions in binary form must reproduce the 225 /// * above copyright notice, this list of 226 /// * conditions and the following disclaimer in 227 /// * the documentation and/or other materials 228 /// * provided with the distribution. 229 /// * 230 /// * o Neither the name of Internet Society, IETF or IETF 231 /// * Trust, nor the names of specific contributors, may be 232 /// * used to endorse or promote products derived from this 233 /// * software without specific prior written permission. 234 /// * 235 /// * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS 236 /// * AND CONTRIBUTORS "AS IS" AND ANY EXPRESS OR IMPLIED 237 /// * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 238 /// * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS 239 /// * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO 240 /// * EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE 241 /// * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, 242 /// * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT 243 /// * NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR 244 /// * SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 245 /// * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF 246 /// * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, 247 /// * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING 248 /// * IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF 249 /// * ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 250 /// */ 251 /// 252 /// /* 253 /// * This code was derived from [RFC2203]. Please 254 /// * reproduce this note if possible. 255 /// */ 256 /// 257 /// /* 258 /// * rpcsec_gss_v3.x 259 /// */ 260 /// 261 /// enum rpc_gss_service_t { 262 /// /* Note: the enumerated value for 0 is reserved. */ 263 /// rpc_gss_svc_none = 1, 264 /// rpc_gss_svc_integrity = 2, 265 /// rpc_gss_svc_privacy = 3, 266 /// rpc_gss_svc_channel_prot = 4 267 /// }; 268 /// 269 /// enum rpc_gss_proc_t { 270 /// RPCSEC_GSS_DATA = 0, 271 /// RPCSEC_GSS_INIT = 1, 272 /// RPCSEC_GSS_CONTINUE_INIT = 2, 273 /// RPCSEC_GSS_DESTROY = 3, 274 /// RPCSEC_GSS_BIND_CHANNEL = 4 275 /// }; 276 /// 277 /// struct rpc_gss_cred_vers_1_t { 278 /// rpc_gss_proc_t gss_proc; /* control procedure */ 279 /// unsigned int seq_num; /* sequence number */ 280 /// rpc_gss_service_t service; /* service used */ 281 /// opaque handle<>; /* context handle */ 282 /// }; 283 /// 284 /// enum rpc_gss3_proc_t { 285 /// RPCSEC_GSS3_DATA = 0, 286 /// RPCSEC_GSS3_LIST = 5, 287 /// RPCSEC_GSS3_CREATE = 6, 288 /// RPCSEC_GSS3_DESTROY = 7 289 /// }; 290 /// 291 /// struct rpc_gss_cred_vers_3_t { 292 /// rpc_gss3_proc_t gss_proc; 293 /// unsigned int seq_num; 294 /// rpc_gss_service_t service; 295 /// opaque handle<>; 296 /// }; 297 /// 298 /// const RPCSEC_GSS_VERS_1 = 1; 299 /// const RPCSEC_GSS_VERS_2 = 2; 300 /// const RPCSEC_GSS_VERS_3 = 3; /* new */ 301 /// 302 /// union rpc_gss_cred_t switch (unsigned int rgc_version) { 303 /// case RPCSEC_GSS_VERS_1: 304 /// case RPCSEC_GSS_VERS_2: 305 /// rpc_gss_cred_vers_1_t rgc_cred_v1; 306 /// case RPCSEC_GSS_VERS_3: /* new */ 307 /// rpc_gss_cred_vers_3_t rgc_cred_v3; 308 /// }; 309 /// 310 /// const MAXSEQ = 0x80000000; 311 /// 312 /// struct rpc_gss3_extension { 313 /// int type; 314 /// bool critical; 315 /// opaque data<>; 316 /// }; 317 /// 318 /// struct rpc_gss3_gss_binding { 319 /// unsigned int vers; 320 /// opaque handle<>; 321 /// opaque nonce<>; 322 /// opaque mic<>; 323 /// }; 324 /// 325 /// typedef opaque rpc_gss3_chan_binding<>; 326 /// 327 /// struct rpc_gss3_lfs { 328 /// unsigned int lfs_id; 329 /// unsigned int pi_id; 330 /// }; 331 /// 332 /// struct rpc_gss3_label { 333 /// rpc_gss3_lfs lfs; 334 /// opaque label<>; 335 /// }; 336 /// 337 /// typedef string rpc_gss3_list_name<>; 338 /// struct rpc_gss3_privs { 339 /// rpc_gss3_list_name listname; 340 /// opaque privilege<>; 341 /// }; 342 /// 343 /// enum rpc_gss3_assertion_type { 344 /// LABEL = 0, 345 /// PRIVS = 1 346 /// }; 347 /// 348 /// union rpc_gss3_assertion_u 349 /// switch (rpc_gss3_assertion_type atype) { 350 /// case LABEL: 351 /// rpc_gss3_label label; 352 /// case PRIVILEGES: 353 /// rpc_gss3_privs privs; 354 /// default: 355 /// opaque ext<>; 356 /// }; 357 /// 358 /// struct rpc_gss3_assertion { 359 /// bool critical; 360 /// rpc_gss3_assertion_u assertion; 361 /// }; 362 /// 363 /// struct rpc_gss3_create_args { 364 /// rpc_gss3_gss_binding *compound_binding; 365 /// rpc_gss3_chan_binding *chan_binding_mic; 366 /// rpc_gss3_assertion assertions<>; 367 /// rpc_gss3_extension extensions<>; 368 /// }; 369 /// 370 /// struct rpc_gss3_create_res { 371 /// opaque handle<>; 372 /// rpc_gss3_chan_binding *chan_binding_mic; 373 /// rpc_gss3_assertion granted_assertions<>; 374 /// rpc_gss3_extension granted_extensions<>; 375 /// }; 376 /// 377 /// enum rpc_gss3_list_item { 378 /// LABEL = 0, 379 /// PRIV = 1, 380 /// }; 381 /// 382 /// struct rpc_gss3_list_args { 383 /// rpc_gss3_list_item list_what<>; 384 /// }; 385 /// 386 /// union rpc_gss3_list_item_u 387 /// switch (rpc_gss3_list_item itype) { 388 /// case LABEL: 389 /// rpc_gss3_lable labels<>; 390 /// case PRIV: 391 /// rpc_gss3_list_name privs<>; 392 /// default: 393 /// opaque ext<>; 394 /// }; 395 /// 396 /// typedef rpc_gss3_list_item_u rpc_gss3_list_res<>; 398 2.1. New auth_stat values 400 RPCSEC_GSSv3 requires the addition of several values to the auth_stat 401 enumerated type definition: 403 enum auth_stat { 404 ... 405 /* 406 * RPCSEC_GSS errors 407 */ 408 RPCSEC_GSS3_COMPOUND_PROBEM = <>, 409 RPCSEC_GSS3_LABEL_PROBLEM = <>, 410 RPCSEC_GSS3_UNKNOWN_ASSERTION = <> 411 RPCSEC_GSS3_UNKNOWN_EXTENSION = <> 412 RPCSEC_GSS3_UNKNOWN_MESSAGE = <> 413 }; 415 [[Comment.3: fix above into YYY. All the entries are TBD... --NW]] 417 2.2. RPC message credential and verifier 419 The rpc_gss_cred_vers_3_t type is used in much the same way that 420 rpc_gss_cred_vers_1_t is used in RPCSEC_GSSv1, that is: as the arm of 421 the rpc_gss_cred_t discriminated union in the RPC message header 422 opaque_auth structure corresponding to version 3 (RPCSEC_GSS_VERS_3). 423 It differs from rpc_gss_cred_vers_1_t in that: 425 a. the values for gss_proc corresponding to control messages are 426 different. 428 b. the handle field is the RPCSEC_GSSv3 (child) handle, except for 429 the RPCSEC_GSS3_CREATE control message where it is set to the 430 parent context handle. 432 For all RPCSEC_GSSv3 data and control messages, the verifier field in 433 the RPC message header is constructed in the RPCSEC_GSSv1 manner 434 using the parent GSS-API security context. 436 2.3. Control Messages 438 There are three RPCSEC_GSSv3 control messages: RPCSEC_GSS3_CREATE, 439 RPCSEC_GSS3_DESTROY, and RPCSEC_GSS3_LIST. 441 RPCSEC_GSSv3 control messages are similar to the RPCSEC_GSSv1 442 RPCSEC_GSS_DESTROY control message (see section 5.4 [2]) in that the 443 sequence number in the request must be valid, and the header checksum 444 in the verifier must be valid. In other words, they look a lot like 445 an RPCSEC_GSSv3 data message with the header procedure set to 446 NULLPROC. 448 As in RPCSEC_GSSv1, the RPCSEC_GSSv3 control messages may contain 449 information following the verifier in the body of the NULLPROC 450 procedure. 452 The client MUST use one of the following security services to protect 453 any RPCSEC_GSSv3 control message: 455 o rpc_gss_svc_channel_prot (see RPCSEC_GSSv2) 457 o rpc_gss_svc_integrity 459 o rpc_gss_svc_privacy 461 Specifically the client MUST NOT use rpc_gss_svc_none. 463 For RPCSEC_GSSv3 control messages the rpc_gss_cred_vers_3_t in the 464 RPC message opaque_auth structure is encoded as follows: 466 1. the union rpc_gss_cred_t version is set to 3 with the value being 467 of type rpc_gss_cred_vers_3_t instead of rpc_gss_cred_vers_1_t. 469 2. the gss_proc is set to one of RPCSEC_GSS3_CREATE, 470 RPCSEC_GSS3_DESTROY, or RPCSEC_GSS3_LIST. 472 3. the seq_num is a valid sequence number for the context in the 473 handle field. 475 4. the rpc_gss_service_t is one of rpc_gss_svc_integrity, 476 rpc_gss_svc_privacy, or rpc_gss_svc_channel_prot. 478 5. the rpc_gss_cred_vers_3_t handle field is either set to the 479 parent context handle for RPCSEC_GSS3_CREATE, or to the GSS3 480 child handle for RPCSEC_GSS3_LIST and RPCSEC_GSS3_DESTROY. 482 2.3.1. Create request 484 As noted in the introduction, RPCSEC_GSSv3 relies on the RPCSEC_GSS 485 version 1 parent context (though version 2 can be used) secure 486 connection to do the actual GSS-API GSS3 security context 487 establishment. As such, the rpc_gss_cred_vers_3_t fields in the RPC 488 Call opaque_auth use the parent context handle and seq_num stream. 490 The RPCSEC_GSS3_CREATE call message binds one or more items of 491 several kinds into a new RPCSEC_GSSv3 context handle: 493 o another RPCSEC_GSS (version 1, 2, or 3) context handle (compound 494 authentication) 496 o a channel binding 498 o authorization assertions (labels, privileges) 500 o extensions (see Section 2.3.4 ) 502 The reply to this message consists of either an error or an 503 rpc_gss3_create_res structure which includes a new RPCSEC_GSSv3 504 handle, termed the "child" which is used for subsequent control and 505 data messages. 507 Upon successful RPCSEC_GSS3_CREATE, both the client and the server 508 should associate the resultant GSSv3 child context handle with the 509 parent context handle in their GSS context caches so as to be able to 510 reference the parent context given the child context handle. 512 [[Comment.4: Destruction of the parent context => first destroy child 513 handle. IOW fail the RPCSEC_GSS_DESTROY of parent with new 514 RPCSEC_GSS3_CONTEXT_EXISTS error code: What about the lifetime of the 515 GSS3 context. Is this meant to be long lived?? --AA]] 517 Server policies should take into account the identity of the client 518 and/or user as authenticated via the GSS-API. Server implementation 519 and policy MAY result in labels, privileges, and identities being 520 mapped to concepts and values that are local to the server. 522 2.3.1.1. Compound authentication 524 RPCSEC_GSSv3 allows for compound authentication of client hosts and 525 users to servers. As in non-compound authentication, there is a 526 parent handle used to protect the RPCSEC_GSS3_CREATE call message, 527 and a resultant RPCSEC_GSSv3 child handle. In addition to the parent 528 handle, the compound authentication create control message has a 529 handle referenced via the compound_binding field of the 530 RPCSEC_GSS3_CREATE arguments structure (rpc_gss3_create_args) termed 531 the "inner" handle, as well as a nonce and a MIC of that nounce 532 created using the GSS-API security context associated with the 533 "inner" handle. 535 All uses of a child context handle that is bound to an inner context 536 MUST be treated as speaking for the initiator principal (as modified 537 by any assertions in the RPCSEC_GSS3_CREATE message) of the inner 538 context handle's GSS-API security context. 540 This feature is needed, for example, when a client wishes to use 541 authority assertions that the server may only grant if a user and a 542 client are authenticated together to the server. Thus a server may 543 refuse to grant requested authority to a user acting alone (e.g., via 544 an unprivileged user-space program), or to a client acting alone 545 (e.g. when a client is acting on behalf of a user) but may grant 546 requested authority to a client acting on behalf of a user if the 547 server identifies the user and trusts the client. 549 It is assumed that an unprivileged user-space program would not have 550 access to client host credentials needed to establish a GSS-API 551 security context authenticating the client to the server, therefore 552 an unprivileged user-space program could not create an RPCSEC_GSSv3 553 RPCSEC_GSS3_CREATE message that successfully binds a client and a 554 user security context. 556 Clients using RPCSEC_GSS context binding MUST use, as the parent 557 context handle, an RPCSEC_GSS context handle that corresponds to a 558 GSS-API security context that authenticates the client host, and for 559 the inner context handle it SHOULD use a context handle to 560 authenticate a user. The reverse (parent handle authenticates user, 561 inner authenticates client) MUST NOT be used. Other compounds might 562 eventually make sense. 564 An RPCSEC_GSSv3 context handle that is bound to another RPCSEC_GSS 565 context MUST be treated by servers as authenticating the GSS-API 566 initiator principal authenticated by the inner context handle's GSS- 567 API security context. This principal may be mapped to a server-side 568 notion of user or principal as modified by any identity assertions by 569 the client in the same RPCSEC_GSS3_CREATE request that the server 570 accepts. 572 2.3.1.2. Channel binding 574 RPCSEC_GSSv3 provides a different way to do channel binding than 575 RPCSEC_GSSv2. Specifically: 577 a. RPCSEC_GSSv3 builds on RPCSEC_GSSv1 by reusing existing, 578 established context handles rather than providing a different RPC 579 security flavor for establishing context handles, 581 b. channel bindings data are not hashed because the community now 582 agrees that it is the secure channel's responsibility to produce 583 channel bindings data of manageable size. 585 (a) is useful in keeping RPCSEC_GSSv3 simple in general, not just for 586 channel binding. (b) is useful in keeping RPCSEC_GSSv3 simple 587 specifically for channel binding. 589 Channel binding is accomplished as follows. The client prefixes the 590 channel bindings data octet string with the channel type as described 591 in [5], then the client calls GSS_GetMIC() to get a MIC of resulting 592 octet string, using the parent RPCSEC_GSS context handle's GSS-API 593 security context. The MIC is then placed in the chan_binding_mic 594 field of RPCSEC_GSS3_CREATE arguments (rpc_gss3_create_args). 596 If the chan_binding_mic field of the arguments of a 597 RPCSEC_GSS3_CREATE control message is set, then the server MUST 598 verify the client's channel binding MIC if the server supports this 599 feature. If channel binding verification succeeds then the server 600 MUST generate a new MIC of the same channel bindings and place it in 601 the chan_binding_mic field of the RPCSEC_GSS3_CREATE results. If 602 channel binding verification fails or the server doesn't support 603 channel binding then the server MUST indicate this in its reply by 604 not including a chan_binding_mic value (chan_binding_mic is an 605 optional field). 607 The client MUST verify the result's chan_binding_mic value, if the 608 server included it, by calling GSS_VerifyMIC() with the given MIC and 609 the channel bindings data (including the channel type prefix). If 610 client-side channel binding verification fails then the client MUST 611 call RPCSEC_GSS3_DESTROY. If the client requested channel binding 612 but the server did not include a chan_binding_mic field in the 613 results, then the client MAY continue to use the resulting context 614 handle as though channel binding had never been requested, otherwise 615 (if the client really wanted channel binding) it MUST call 616 RPCSEC_GSS3_DESTROY. 618 As per-RPCSEC_GSSv2 [4]: 620 "Once a successful [channel binding] procedure has been performed 621 on an [RPCSEC_GSSv3] context handle, the initiator's 622 implementation may map application requests for rpc_gss_svc_none 623 and rpc_gss_svc_integrity to rpc_gss_svc_channel_prot credentials. 624 And if the secure channel has privacy enabled, requests for 625 rpc_gss_svc_privacy can also be mapped to 626 rpc_gss_svc_channel_prot." 628 Any RPCSEC_GSSv3 context handle that has been bound to a secure 629 channel in this way SHOULD be used only with the 630 rpc_gss_svc_channel_prot, and SHOULD NOT be used with 631 rpc_gss_svc_none nor rpc_gss_svc_integrity -- if the secure channel 632 does not provide privacy protection then the client MAY use 633 rpc_gss_svc_privacy where privacy protection is needed or desired. 635 2.3.1.3. Label assertions 637 RPCSEC_GSSv3 clients MAY assert a security label in some LSF by 638 binding this assertion into an RPCSEC_GSSv3 context handle. This is 639 done by including an assertion of type rpc_gss3_label in the 640 'assertions' field (discriminant: 'LABEL') of the RPCSEC_GSS3_CREATE 641 arguments to the desired LSF and label. 643 Label encoding is specified to mirror the NFSv4 sec_label attribute 644 described in Section 12.2.2 of [6]. The label format specifier (LFS) 645 is an identifier used by the client to establish the syntactic format 646 of the security label and the semantic meaning of its components. 647 The policy identifier (PI) is an optional part of the definition of 648 an LFS which allows for clients and server to identify specific 649 security policies. The opaque label field of rpc_gss3_label is 650 dependent on the MAC model to interpret and enforce. 652 [[Comment.5: Check that this Label definition provides all the 653 required pieces to enable full mode when combined with NFSv4.2 LNFS. 654 Specifically, how does the client find out and respond if a server 655 has changed a label. --AA]] 657 If a label itself requires privacy protection (i.e., that the user 658 can assert that label is a secret) then the client MUST use the 659 rpc_gss_svc_privacy protection service for the RPCSEC_GSS3_CREATE 660 request or, if the parent handle is bound to a secure channel that 661 provides privacy protection, rpc_gss_svc_channel_prot. 663 If a client wants to ensure that the server understands the asserted 664 label then it MUST set the 'critical' field of the label assertion to 665 TRUE, otherwise it MUST set it to FALSE. 667 Servers that do not support labeling MUST ignore non-critical label 668 assertions. Servers that do not support the requested LFS MUST 669 either ignore non-critical label assertions or map them to a suitable 670 label in a supported LFS. Servers that do not support labeling or do 671 not support the requested LFS MUST return an error if the label 672 request is critical. Servers that support labeling in the requested 673 LFS MAY map the requested label to different label as a result of 674 server-side policy evaluation. 676 2.3.1.4. Structured privilege assertions 678 A structured privilege is an RPC application defined structure that 679 is opaque, and is encoded in the rpc_gss3_privs privilege field. 680 Encoding, server verification and any server policies for structured 681 privileges are described by the RPC application definition. The 682 listname field of rpc_gss3_privs is a description string used to list 683 the privilege. 685 A successful structured privilege assertion RPCSEC_GSS3_CREATE call 686 must return all accepted privileges in the rpc_gss3_privs 687 granted_assertions field. 689 Section 3.4.1.2. "Inter-Server Copy with RPCSEC_GSSv3" of [6] shows 690 an example of structured privilege definition and use. 692 2.3.2. Destruction request 694 The RPCSEC_GSS3_DESTROY control message is the same as the 695 RPCSEC_GSSv1 RPCSEC_GSS_DESTROY control message, but with the version 696 3 header. Specifically, the rpc_gss_cred_vers_3_t fields in the RPC 697 Call opaque_auth use the GSS3 context handle and seq_num stream. As 698 with all RPCSEC_GSSv3 messages, the header checksum uses the parent 699 context, and needs to be valid. 701 The server sends a response as it would to a data request. The 702 client and server must then destroy the context for the session. 704 2.3.3. List request 706 The RPCSEC_GSS3_LIST control message is similar to 707 RPCSEC_GSS3_DESTROY message. Specifically, the rpc_gss_cred_vers_3_t 708 fields in the RPC Call opaque_auth use the GSS3 context handle and 709 seq_num stream. As with all RPCSEC_GSSv3 messages, the header 710 checksum uses the parent context, and needs to be valid. 712 The RPCSEC_GSS3_LIST control message consists of a single integer 713 indicating what should be listed, and the reply consists of an error 714 or the requested list. The client may list LFSs or structured 715 privilege listnames. 717 The result is an opaque octet string containing a list of LFSs 718 [encoding TBD] or a list of active structured privileges [encoding 719 TBD]. 721 2.3.4. Extensibility 723 New fields may be added through the 'extensions' typed hole. All 724 such extensions have a 'critical' flag. 726 [[Comment.6: Should we keep the extensions types hole? I think 727 not... --AA]] 729 Assertion types may be added in the future by adding arms to the 730 'rpc_gss3_assertion_u' union. Every assertion has a 'critical' flag 731 that can be used to indicate criticality. Other assertion types are 732 described elsewhere and include: 734 o Client-side assertions of identity: 736 * Primary client/user identity 738 * Supplementary group memberships of the client/user, including 739 support for specifying deltas to the membership list as seen on 740 the server. 742 New control message types may be added. 744 Servers receiving unknown critical client assertions or unknown 745 RPCSEC_GSS_v3 extensions MUST return an error. 747 There is no IANA or other registry for RPCSEC_GSSv3 extensions. All 748 extensions MUST be done by IETF Protocol Action. 750 2.4. Data Messages 752 RPCSEC_GSS3_DATA messages differ from from RPCSEC_GSSv1 data messages 753 in that the version number used MUST be '3' instead of '1'. As noted 754 in Section 2.2 the RPCSEC_GSSv3 context handle is used along with 755 it's sequence number stream. 757 For RPCSEC_GSSv3 data messages the rpc_gss_cred_vers_3_t in the RPC 758 message opaque_auth structure is encoded as follows: 760 1. the union rpc_gss_cred_t version is set to 3 with the value being 761 of type rpc_gss_cred_vers_3_t instead of rpc_gss_cred_vers_1_t. 763 2. the gss_proc is set to RPCSEC_GSS3_DATA 765 3. the seq_num is a valid GSS3 context (child context) sequence 766 number. 768 4. just as in RPCSEC_GSSv1, the rpc_gss_service_t is one of 769 rpc_gss_svc_none, rpc_gss_svc_integrity, rpc_gss_svc_privacy, or 770 rpc_gss_svc_channel_prot. 772 5. the handle field is set to the (child) RPCSEC_GSSv3 context 773 handle 775 3. Security Considerations 777 This entire document deals with security issues. 779 The RPCSEC_GSSv3 protocol allows for client-side assertions of data 780 that is relevant to server-side authorization decisions. These 781 assertions must be evaludated by the server in the context of whether 782 the client and/or user are authenticated, whether compound 783 authentication was used, whether the client is trusted, what ranges 784 of assertions are allowed for the client and the user (separately or 785 together), and any relevant server-side policy. 787 The security semantics of assertions carried by RPCSEC_GSSv3 are 788 application protocol-specific. 790 RPCSEC_GSSv3 supports a notion of critical assertions (and 791 extensions), but there's no need for peers to tell each other what 792 assertions were granted, or what they were mapped to. 794 Note that RPSEC_GSSv3 is not a complete solution for labeling: it 795 conveys the labels of actors, but not the labels of objects. RPC 796 application protocols may require extending in order to carry object 797 label information. 799 There may be interactions with NFSv4's callback security scheme and 800 NFSv4.1's GSS-API "SSV" mechanisms. Specifically, the NFSv4 callback 801 scheme requires that the server initiate GSS-API security contexts, 802 which does not work well in practice, and in the context of client- 803 side processes running as the same user but with different privileges 804 and security labels the NFSv4 callback security scheme seems 805 particularly unlikely to work well. NFSv4.1 has the server use an 806 existing, client-initiated RPCSEC_GSS context handle to protect 807 server-initiated callback RPCs. The NFSv4.1 callback security scheme 808 lacks all the problems of the NFSv4 scheme, however, it is important 809 that the server pick an appropriate RPCSEC_GSS context handle to 810 protect any callbacks. Specifically, it is important that the server 811 use RPCSEC_GSS context handles which authenticate the client to 812 protect any callbacks relating to server state initiated by RPCs 813 protected by RPCSEC_GSSv3 contexts. 815 [[Comment.7: [Add text about interaction with GSS-SSV...] --NW]] 817 [[Comment.8: I see no reason to use RPCSEC_GSSv3 contexts for NFSv4.x 818 back channel. --AA]] 820 [[Comment.9: Since GSS3 requires an RPCSEC_GSSv1 or v2 context handle 821 to establish a GSS3 context, SSV can not be used as this draft is 822 written.]] 824 [[Comment.10: AFAICS the reason to use SSV is to avoid using a client 825 machine credential which means compound authentication can not be 826 used. Since GSS3 requires an RPCSEC_GSSv1 or v2 context handle to 827 establish a GSS3 context, SSV can not be used as the parent context 828 for GSSv3. --AA]] 830 4. IANA Considerations 832 This section uses terms that are defined in [8]. 834 There are no IANA considerations in this document. TBDs in this 835 document will be assigned by the ONC RPC registrar (which is not 836 IANA, XXX: verify). 838 5. References 839 5.1. Normative References 841 [1] Bradner, S., "Key words for use in RFCs to Indicate Requirement 842 Levels", March 1997. 844 [2] Eisler, M., Chiu, A., and L. Ling, "RPCSEC_GSS Protocol 845 Specification", RFC 2203, September 1997. 847 [3] Linn, J., "Generic Security Service Application Program 848 Interface Version 2, Update 1", RFC 2743, January 2000. 850 [4] Srinivasan, R., "RPC: Remote Procedure Call Protocol 851 Specification Version 2", RFC 1831, August 1995. 853 [5] Williams, N., "On the Use of Channel Bindings to Secure 854 Channels", RFC 5056, November 2007. 856 [6] Haynes, T., "NFS Version 4 Minor Version 2", 857 draft-ietf-nfsv4-minorversion2-19 (Work In Progress), 858 March 2013. 860 [7] Eisler, M., "XDR: External Data Representation Standard", 861 RFC 4506, May 2006. 863 [8] Narten, T. and H. Alvestrand, "Guidelines for Writing an IANA 864 Considerations Section in RFCs", BCP 26, RFC 5226, May 2008. 866 5.2. Informative References 868 [9] Haynes, T., "Requirements for Labeled NFS", 869 draft-ietf-nfsv4-labreqs-03 (work in progress). 871 [10] "Section 46.6. Multi-Level Security (MLS) of Deployment Guide: 872 Deployment, configuration and administration of Red Hat 873 Enterprise Linux 5, Edition 6", 2011. 875 [11] Smalley, S., "The Distributed Trusted Operating System (DTOS) 876 Home Page", 877 . 879 [12] Carter, J., "Implementing SELinux Support for NFS", 880 . 882 [13] Quigley, D. and J. Lu, "Registry Specification for MAC Security 883 Label Formats", draft-quigley-label-format-registry (work in 884 progress), 2011. 886 Appendix A. Acknowledgments 888 Appendix B. RFC Editor Notes 890 [RFC Editor: please remove this section prior to publishing this 891 document as an RFC] 893 [RFC Editor: prior to publishing this document as an RFC, please 894 replace all occurrences of RFCTBD10 with RFCxxxx where xxxx is the 895 RFC number of this document] 897 Authors' Addresses 899 William A. (Andy) Adamson 900 NetApp 901 3629 Wagner Ridge Ctt 902 Ann Arbor, MI 48103 903 USA 905 Phone: +1 734 665 1204 906 Email: andros@netapp.com 908 Nico Williams 909 cryptonector.com 910 13115 Tamayo Dr 911 Austin, TX 78729 912 USA 914 Email: nico@cryptonector.com