idnits 2.17.00 (12 Aug 2021) /tmp/idnits23376/draft-ietf-nfsv4-rfc5667bis-13.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 == The document seems to lack the recommended RFC 2119 boilerplate, even if it appears to use RFC 2119 keywords -- however, there's a paragraph with a matching beginning. Boilerplate error? (The document does seem to have the reference to RFC 2119 which the ID-Checklist requires). == The document seems to contain a disclaimer for pre-RFC5378 work, but was first submitted on or after 10 November 2008. The disclaimer is usually necessary only for documents that revise or obsolete older RFCs, and that take significant amounts of text from those RFCs. If you can contact all authors of the source material and they are willing to grant the BCP78 rights to the IETF Trust, you can and should remove the disclaimer. Otherwise, the disclaimer is needed and you can ignore this comment. (See the Legal Provisions document at https://trustee.ietf.org/license-info for more information.) -- The document date (August 9, 2017) is 1739 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) ** Obsolete normative reference: RFC 5661 (Obsoleted by RFC 8881) -- Obsolete informational reference (is this intentional?): RFC 5666 (Obsoleted by RFC 8166) -- Obsolete informational reference (is this intentional?): RFC 5667 (Obsoleted by RFC 8267) Summary: 1 error (**), 0 flaws (~~), 3 warnings (==), 3 comments (--). Run idnits with the --verbose option for more detailed information about the items above. -------------------------------------------------------------------------------- 2 Network File System Version 4 C. Lever 3 Internet-Draft Oracle 4 Obsoletes: 5667 (if approved) August 9, 2017 5 Intended status: Standards Track 6 Expires: February 10, 2018 8 Network File System (NFS) Upper Layer Binding To RPC-Over-RDMA Version 1 9 draft-ietf-nfsv4-rfc5667bis-13 11 Abstract 13 This document specifies Upper Layer Bindings of Network File System 14 (NFS) protocol versions to RPC-over-RDMA version 1, enabling the use 15 of Direct Data Placement. This document obsoletes RFC 5667. 17 Status of This Memo 19 This Internet-Draft is submitted in full conformance with the 20 provisions of BCP 78 and BCP 79. 22 Internet-Drafts are working documents of the Internet Engineering 23 Task Force (IETF). Note that other groups may also distribute 24 working documents as Internet-Drafts. The list of current Internet- 25 Drafts is at http://datatracker.ietf.org/drafts/current/. 27 Internet-Drafts are draft documents valid for a maximum of six months 28 and may be updated, replaced, or obsoleted by other documents at any 29 time. It is inappropriate to use Internet-Drafts as reference 30 material or to cite them other than as "work in progress." 32 This Internet-Draft will expire on February 10, 2018. 34 Copyright Notice 36 Copyright (c) 2017 IETF Trust and the persons identified as the 37 document authors. All rights reserved. 39 This document is subject to BCP 78 and the IETF Trust's Legal 40 Provisions Relating to IETF Documents 41 (http://trustee.ietf.org/license-info) in effect on the date of 42 publication of this document. Please review these documents 43 carefully, as they describe your rights and restrictions with respect 44 to this document. Code Components extracted from this document must 45 include Simplified BSD License text as described in Section 4.e of 46 the Trust Legal Provisions and are provided without warranty as 47 described in the Simplified BSD License. 49 This document may contain material from IETF Documents or IETF 50 Contributions published or made publicly available before November 51 10, 2008. The person(s) controlling the copyright in some of this 52 material may not have granted the IETF Trust the right to allow 53 modifications of such material outside the IETF Standards Process. 54 Without obtaining an adequate license from the person(s) controlling 55 the copyright in such materials, this document may not be modified 56 outside the IETF Standards Process, and derivative works of it may 57 not be created outside the IETF Standards Process, except to format 58 it for publication as an RFC or to translate it into languages other 59 than English. 61 Table of Contents 63 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 64 2. Requirements Language . . . . . . . . . . . . . . . . . . . . 3 65 3. Reply Size Estimation . . . . . . . . . . . . . . . . . . . . 3 66 3.1. Short Reply Chunk Retry . . . . . . . . . . . . . . . . . 4 67 4. Upper Layer Binding for NFS Versions 2 and 3 . . . . . . . . 5 68 4.1. Reply Size Estimation . . . . . . . . . . . . . . . . . . 5 69 4.2. RPC Binding Considerations . . . . . . . . . . . . . . . 5 70 5. Upper Layer Bindings for NFS Version 2 and 3 Auxiliary 71 Protocols . . . . . . . . . . . . . . . . . . . . . . . . . . 6 72 5.1. MOUNT, NLM, and NSM Protocols . . . . . . . . . . . . . . 6 73 5.2. NFSACL Protocol . . . . . . . . . . . . . . . . . . . . . 6 74 6. Upper Layer Binding For NFS Version 4 . . . . . . . . . . . . 7 75 6.1. DDP-Eligibility . . . . . . . . . . . . . . . . . . . . . 7 76 6.2. Reply Size Estimation . . . . . . . . . . . . . . . . . . 7 77 6.3. RPC Binding Considerations . . . . . . . . . . . . . . . 8 78 6.4. NFS COMPOUND Requests . . . . . . . . . . . . . . . . . . 9 79 6.5. NFS Callback Requests . . . . . . . . . . . . . . . . . . 11 80 6.6. Session-Related Considerations . . . . . . . . . . . . . 12 81 6.7. Transport Considerations . . . . . . . . . . . . . . . . 13 82 7. Extending NFS Upper Layer Bindings . . . . . . . . . . . . . 14 83 8. Security Considerations . . . . . . . . . . . . . . . . . . . 14 84 9. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 15 85 10. References . . . . . . . . . . . . . . . . . . . . . . . . . 15 86 10.1. Normative References . . . . . . . . . . . . . . . . . . 15 87 10.2. Informative References . . . . . . . . . . . . . . . . . 16 88 Appendix A. Changes Since RFC 5667 . . . . . . . . . . . . . . . 17 89 Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . 18 90 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 18 92 1. Introduction 94 The RPC-over-RDMA version 1 transport may employ direct data 95 placement to convey data payloads associated with RPC transactions 96 [RFC8166]. To enable successful interoperation, RPC client and 97 server implementations using RPC-over-RDMA version 1 must agree which 98 XDR data items and RPC procedures are eligible to use direct data 99 placement (DDP). 101 An Upper Layer Binding specifies this agreement for one version or 102 more versions of one RPC program. Other operational details, such as 103 RPC binding assignments, pairing Write chunks with result data items, 104 and reply size estimation, are also specified by this Binding. 106 This document contains material required of Upper Layer Bindings, as 107 specified in [RFC8166], for the following NFS protocol versions: 109 o NFS version 2 [RFC1094] 111 o NFS version 3 [RFC1813] 113 o NFS version 4.0 [RFC7530] 115 o NFS version 4.1 [RFC5661] 117 o NFS version 4.2 [RFC7862] 119 Upper Layer Bindings are also provided for auxiliary protocols used 120 with NFS versions 2 and 3 (see Section 5). 122 This document assumes the reader is already familiar with concepts 123 and terminology defined in [RFC8166] and the documents it references. 125 2. Requirements Language 127 The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", 128 "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this 129 document are to be interpreted as described in BCP 14 [RFC2119] 130 [RFC8174] when, and only when, they appear in all capitals, as shown 131 here. 133 3. Reply Size Estimation 135 During the construction of each RPC Call message, a requester is 136 responsible for allocating appropriate resources for receiving the 137 corresponding Reply message. If the requester expects the RPC Reply 138 message will be larger than its inline threshold, it provides Write 139 and/or Reply chunks wherein the responder can place results and the 140 reply's Payload stream. 142 A reply resource overrun occurs if the RPC Reply Payload stream does 143 not fit into the provided Reply chunk, or no Reply chunk was provided 144 and the Payload stream does not fit inline. This prevents the 145 responder from returning the Upper Layer reply to the requester. 146 Therefore reliable reply size estimation is necessary to ensure 147 successful interoperation. 149 In most cases, the NFS protocol's XDR definition provides enough 150 information to enable an NFS client to predict the maximum size of 151 the expected Reply message. If there are variable-size data items in 152 the result, the maximum size of the RPC Reply message can be 153 estimated as follows: 155 o The client requests only a specific portion of an object (for 156 example, using the "count" and "offset" fields in an NFS READ). 158 o The client limits the number of results (e.g. using the "count" 159 field of an NFS READDIR request). 161 o The client has already cached the size of the whole object it is 162 about to request (say, via a previous NFS GETATTR request). 164 o The client and server have negotiated a maximum size for all calls 165 and responses (using a CREATE_SESSION operation, for instance). 167 3.1. Short Reply Chunk Retry 169 In a few cases, either the size of one or more returned data items or 170 the number of returned data items cannot be known in advance of 171 forming an RPC Call. 173 If an NFS server finds that the NFS client provided inadequate 174 receive resources to return the whole reply, it returns an RPC level 175 error or a transport error, such as ERR_CHUNK. 177 In response to these errors, an NFS client can choose to: 179 o Terminate the RPC transaction immediately with an error, or 181 o Allocate a larger Reply chunk and send the same request as a new 182 RPC transaction (a new XID should be assigned to the retransmitted 183 request to avoid matching a cached RPC Reply that caches the 184 original error). The NFS client should avoid retrying the request 185 indefinitely because a responder may return ERR_CHUNK for a 186 variety of reasons. 188 Subsequent sections of this document discuss exactly which operations 189 might have ultimate difficulty with Reply size estimation. These 190 operations are eligible for "short Reply chunk retry." Unless 191 explicitly mentioned as applicable, short Reply chunk retry should 192 not be used since accurate reply size estimation is problematic in 193 only a few cases. In all other cases reply size underestimation is 194 considered a correctable implementation bug. 196 NFS server implementations can avoid connection loss by first 197 confirming that target RDMA segments are large enough to receive 198 results before initiating explicit RDMA operations. 200 4. Upper Layer Binding for NFS Versions 2 and 3 202 The Upper Layer Binding specification in this section applies to NFS 203 version 2 [RFC1094] and NFS version 3 [RFC1813]. For brevity, in 204 this document a "Legacy NFS client" refers to an NFS client using 205 version 2 or version 3 of the NFS RPC program (100003) to communicate 206 with an NFS server. Likewise, a "Legacy NFS server" is an NFS server 207 communicating with clients using NFS version 2 or NFS version 3. 209 The following XDR data items in NFS versions 2 and 3 are DDP- 210 eligible: 212 o The opaque file data argument in the NFS WRITE procedure 214 o The pathname argument in the NFS SYMLINK procedure 216 o The opaque file data result in the NFS READ procedure 218 o The pathname result in the NFS READLINK procedure 220 All other argument or result data items in NFS versions 2 and 3 are 221 not DDP-eligible. 223 A transport error does not give an indication of whether the server 224 has processed the arguments of the RPC Call, or whether the server 225 has accessed or modified client memory associated with that RPC. 227 4.1. Reply Size Estimation 229 A Legacy NFS client determines the maximum reply size for each 230 operation using the criteria outlined in Section 3. There are no 231 operations in NFS version 2 or 3 that benefit from short Reply chunk 232 retry. 234 4.2. RPC Binding Considerations 236 Legacy NFS servers traditionally listen for clients on UDP and TCP 237 port 2049. Additionally, they register these ports with a local 238 portmapper [RFC1833] service. 240 A Legacy NFS server supporting RPC-over-RDMA version 1 on such a 241 network and registering itself with the RPC portmapper MAY choose an 242 arbitrary port, or MAY use the alternative well-known port number for 243 its RPC-over-RDMA service (see Section 9). The chosen port MAY be 244 registered with the RPC portmapper under the netids assigned in 245 [RFC8166]. 247 5. Upper Layer Bindings for NFS Version 2 and 3 Auxiliary Protocols 249 NFS versions 2 and 3 are typically deployed with several other 250 protocols, sometimes referred to as "NFS auxiliary protocols." These 251 are distinct RPC programs that define procedures which are not part 252 of the NFS RPC program (100003). The Upper Layer Bindings in this 253 section apply to: 255 o Versions 2 and 3 of the MOUNT RPC program (100005) [RFC1813] 257 o Versions 1, 3, and 4 of the NLM RPC program (100021) [RFC1813] 259 o Version 1 of the NSM RPC program (100024), described in Chapter 11 260 of [XNFS] 262 o Version 1 of the NFSACL RPC program (100227), which does not have 263 a public definition. NFSACL is treated in this document as a de 264 facto standard, as there are several interoperating 265 implementations. 267 5.1. MOUNT, NLM, and NSM Protocols 269 Historically, NFS/RDMA implementations have chosen to convey the 270 MOUNT, NLM, and NSM protocols via TCP. To enable interoperation of 271 these protocols when NFS/RDMA is in use, a legacy NFS server MUST 272 provide support for these protocols via TCP. 274 5.2. NFSACL Protocol 276 Legacy clients and servers that support the NFSACL RPC program 277 typically convey NFSACL procedures on the same connection as the NFS 278 RPC program (100003). This obviates the need for separate rpcbind 279 queries to discover server support for this RPC program. 281 ACLs are typically small, but even large ACLs must be encoded and 282 decoded to some degree. Thus no data item in this Upper Layer 283 Protocol is DDP-eligible. 285 For procedures whose replies do not include an ACL object, the size 286 of a reply is determined directly from the NFSACL RPC program's XDR 287 definition. 289 There is no protocol-specified size limit for NFS version 3 ACLs, and 290 there is no mechanism in either the NFSACL or NFS RPC programs for a 291 Legacy client to ascertain the largest ACL a Legacy server can 292 return. Legacy client implementations should choose a maximum size 293 for ACLs based on their own internal limits. 295 Because an NFSACL client cannot know in advance how large a returned 296 ACL will be, it can use short Reply chunk retry when an NFSACL GETACL 297 operation encounters a transport error. 299 6. Upper Layer Binding For NFS Version 4 301 The Upper Layer Binding specification in this section applies to 302 versions of the NFS RPC program defined in NFS version 4.0 [RFC7530], 303 NFS version 4.1 [RFC5661], and NFS version 4.2 [RFC7862]. 305 6.1. DDP-Eligibility 307 Only the following XDR data items in the COMPOUND procedure of all 308 NFS version 4 minor versions are DDP-eligible: 310 o The opaque data field in the WRITE4args structure 312 o The linkdata field of the NF4LNK arm in the createtype4 union 314 o The opaque data field in the READ4resok structure 316 o The linkdata field in the READLINK4resok structure 318 6.2. Reply Size Estimation 320 Within NFS version 4, there are certain variable-length result data 321 items whose maximum size cannot be estimated by clients reliably 322 because there is no protocol-specified size limit on these arrays. 323 These include: 325 o The attrlist4 field 327 o Fields containing ACLs such as fattr4_acl, fattr4_dacl, 328 fattr4_sacl 330 o Fields in the fs_locations4 and fs_locations_info4 data structures 332 o Fields opaque to the NFS version 4 protocol which pertain to pNFS 333 layout metadata, such as loc_body, loh_body, da_addr_body, 334 lou_body, lrf_body, fattr_layout_types and fs_layout_types, 336 6.2.1. Reply Size Estimation for Minor Version 0 338 The NFS version 4.0 protocol itself does not impose any bound on the 339 size of NFS calls or responses. 341 Some of the data items enumerated in Section 6.2 (in particular, the 342 items related to ACLs and fs_locations) make it difficult to predict 343 the maximum size of NFS version 4.0 replies that interrogate 344 variable-length fattr4 attributes. Client implementations might rely 345 on their own internal architectural limits to constrain the reply 346 size, but such limits are not always guaranteed to be reliable. 348 When an especially large fattr4 result is expected, a Reply chunk 349 might be required. An NFS version 4.0 client can use short Reply 350 chunk retry when an NFS COMPOUND containing a GETATTR operation 351 encounters a transport error. 353 The use of NFS COMPOUND operations raises the possibility of requests 354 that combine a non-idempotent operation (e.g. RENAME) with a GETATTR 355 operation that requests one or more variable-length results. This 356 combination should be avoided by ensuring that any GETATTR operation 357 that requests a result of unpredictable length is sent in an NFS 358 COMPOUND by itself. 360 6.2.2. Reply Size Estimation for Minor Version 1 and Newer 362 In NFS version 4.1 and newer minor versions, the csa_fore_chan_attrs 363 argument of the CREATE_SESSION operation contains a 364 ca_maxresponsesize field. The value in this field can be taken as 365 the absolute maximum size of replies generated by an NFS version 4.1 366 server. 368 This value can be used in cases where it is not possible to estimate 369 a reply size upper bound precisely. In practice, objects such as 370 ACLs, named attributes, layout bodies, and security labels are much 371 smaller than this maximum. 373 6.3. RPC Binding Considerations 375 NFS version 4 servers are required to listen on TCP port 2049, and 376 they are not required to register with an rpcbind service [RFC7530]. 378 Therefore, an NFS version 4 server supporting RPC-over-RDMA version 1 379 MUST use the alternative well-known port number for its RPC-over-RDMA 380 service (see Section 9). Clients SHOULD connect to this well-known 381 port without consulting the RPC portmapper (as for NFS version 4 on 382 TCP transports). 384 6.4. NFS COMPOUND Requests 386 6.4.1. Multiple DDP-eligible Data Items 388 An NFS version 4 COMPOUND procedure can contain more than one 389 operation that carries a DDP-eligible data item. An NFS version 4 390 client provides XDR Position values in each Read chunk to 391 disambiguate which chunk is associated with which argument data item. 392 However NFS version 4 server and client implementations must agree in 393 advance on how to pair Write chunks with returned result data items. 395 In the following list, a "READ operation" refers to any NFS version 4 396 operation which has a DDP-eligible result data item. The mechanism 397 specified in Section 4.3.2 of [RFC8166] is applied to this class of 398 operations: 400 o If an NFS version 4 client wishes all DDP-eligible items in an NFS 401 reply to be conveyed inline, it leaves the Write list empty. 403 o The first chunk in the Write list MUST be used by the first READ 404 operation in an NFS version 4 COMPOUND procedure. The next Write 405 chunk is used by the next READ operation, and so on. 407 o If an NFS version 4 client has provided a matching non-empty Write 408 chunk, then the corresponding READ operation MUST return its DDP- 409 eligible data item using that chunk. 411 o If an NFS version 4 client has provided an empty matching Write 412 chunk, then the corresponding READ operation MUST return all of 413 its result data items inline. 415 o If a READ operation returns a union arm which does not contain a 416 DDP-eligible result, and the NFS version 4 client has provided a 417 matching non-empty Write chunk, an NFS version 4 server MUST 418 return an empty Write chunk in that Write list position. 420 o If there are more READ operations than Write chunks, then 421 remaining NFS Read operations in an NFS version 4 COMPOUND that 422 have no matching Write chunk MUST return their results inline. 424 6.4.2. Chunk List Complexity 426 The RPC-over-RDMA version 1 protocol does not place any limit on the 427 number of chunks or segments that may appear in Read or Write lists. 428 However, for various reasons NFS version 4 server implementations 429 often have practical limits on the number of chunks or segments they 430 are prepared to process in a single RPC transaction conveyed via RPC- 431 over-RDMA version 1. 433 These implementation limits are especially important when Kerberos 434 integrity or privacy is in use [RFC7861]. GSS services increase the 435 size of credential material in RPC headers, potentially requiring 436 more frequent use of Long messages. This can increase the complexity 437 of chunk lists independent of the NFS version 4 COMPOUND being 438 conveyed. 440 In the absence of explicit knowledge of the server's limits, NFS 441 version 4 clients SHOULD follow the prescriptions listed below when 442 constructing RPC-over-RDMA version 1 messages. NFS version 4 servers 443 MUST accept and process such requests. 445 o The Read list can contain either a Position-Zero Read chunk, one 446 Read chunk with a non-zero Position, or both. 448 o The Write list can contain no more than one Write chunk. 450 o Any chunk can contain up to sixteen RDMA segments. 452 NFS version 4 clients wishing to send more complex chunk lists can 453 provide configuration interfaces to bound the complexity of NFS 454 version 4 COMPOUNDs, limit the number of elements in scatter-gather 455 operations, and avoid other sources of chunk overruns at the 456 receiving peer. 458 An NFS version 4 server SHOULD return one of the following responses 459 to a client that has sent an RPC transaction via RPC-over-RDMA 460 version 1 which cannot be processed due to chunk list complexity 461 limits on the server: 463 o A problem is detected by the transport layer while parsing the 464 transport header in an RPC Call message. The server responds with 465 an RDMA_ERROR message with the err field set to ERR_CHUNK. 467 o A problem is detected during XDR decoding of the RPC Call message 468 while the RPC layer reassembles the call's XDR stream. The server 469 responds with an RPC reply with its "reply_stat" field set to 470 MSG_ACCEPTED and its "accept_stat" field set to GARBAGE_ARGS. 472 After receiving one of these errors, an NFS version 4 client SHOULD 473 NOT retransmit the failing request, as the result would be the same 474 error. It SHOULD immediately terminate the RPC transaction 475 associated with the XID in the reply. 477 6.4.3. NFS Version 4 COMPOUND Example 479 The following example shows a Write list with three Write chunks, A, 480 B, and C. The NFS version 4 server consumes the provided Write 481 chunks by writing the results of the designated operations in the 482 compound request (READ and READLINK) back to each chunk. 484 Write list: 486 A --> B --> C 488 NFS version 4 COMPOUND request: 490 PUTFH LOOKUP READ PUTFH LOOKUP READLINK PUTFH LOOKUP READ 491 | | | 492 v v v 493 A B C 495 If the NFS version 4 client does not want to have the READLINK result 496 returned via RDMA, it provides an empty Write chunk for buffer B to 497 indicate that the READLINK result must be returned inline. 499 6.5. NFS Callback Requests 501 The NFS version 4 family of protocols support server-initiated 502 callbacks to notify NFS version 4 clients of events such as recalled 503 delegations. 505 6.5.1. NFS Version 4.0 Callback 507 NFS version 4.0 implementations typically employ a separate TCP 508 connection to handle callback operations, even when the forward 509 channel uses an RPC-over-RDMA version 1 transport. 511 No operation in the NFS version 4.0 callback RPC program conveys a 512 significant data payload. Therefore, no XDR data items in this RPC 513 program is DDP-eligible. 515 A CB_RECALL reply is small and fixed in size. The CB_GETATTR reply 516 contains a variable-length fattr4 data item. See Section 6.2.1 for a 517 discussion of reply size prediction for this data item. 519 An NFS version 4.0 client advertises netids and ad hoc port addresses 520 for contacting its NFS version 4.0 callback service using the 521 SETCLIENTID operation. 523 6.5.2. NFS Version 4.1 Callback 525 In NFS version 4.1 and newer minor versions, callback operations may 526 appear on the same connection as is used for NFS version 4 forward 527 channel client requests. NFS version 4 clients and servers MUST use 528 the approach described in [RFC8167] when backchannel operations are 529 conveyed on RPC-over-RDMA version 1 transports. 531 The csa_back_chan_attrs argument of the CREATE_SESSION operation 532 contains a ca_maxresponsesize field. The value in this field can be 533 taken as the absolute maximum size of backchannel replies generated 534 by a replying NFS version 4 client. 536 There are no DDP-eligible data items in callback procedures defined 537 in NFS version 4.1 or NFS version 4.2. However, some callback 538 operations, such as messages that convey device ID information, can 539 be large, in which case a Long Call or Reply might be required. 541 When an NFS version 4.1 client can support Long Calls in its 542 backchannel, it reports a backchannel ca_maxrequestsize that is 543 larger than the connection's inline thresholds. Otherwise an NFS 544 version 4 server MUST use only Short messages to convey backchannel 545 operations. 547 6.6. Session-Related Considerations 549 The presence of an NFS session (defined in [RFC5661]) has no effect 550 on the operation of RPC-over-RDMA version 1. None of the operations 551 introduced to support NFS sessions (e.g. the SEQUENCE operation) 552 contain DDP-eligible data items. There is no need to match the 553 number of session slots with the number of available RPC-over-RDMA 554 credits. 556 However, there are a few new cases where an RPC transaction can fail. 557 For example, a requester might receive, in response to an RPC 558 request, an RDMA_ERROR message with an rdma_err value of ERR_CHUNK. 559 These situations are not different from existing RPC errors which an 560 NFS session implementation is already prepared to handle for other 561 transports. And as with other transports during such a failure, 562 there might be no SEQUENCE result available to the requester to 563 distinguish whether failure occurred before or after the requested 564 operations were executed on the responder. 566 When a transport error occurs (e.g. RDMA_ERROR), the requester 567 proceeds as usual to match the incoming XID value to a waiting RPC 568 Call. The RPC transaction is terminated, and the result status is 569 reported to the Upper Layer Protocol. The requester's session 570 implementation then determines the session ID and slot for the failed 571 request, and performs slot recovery to make that slot usable again. 572 If this were not done, that slot could be rendered permanently 573 unavailable. 575 When an NFS session is not present (for example, when NFS version 4.0 576 is in use), a transport error does not provide an indication of 577 whether the server has processed the arguments of the RPC Call, or 578 whether the server has accessed or modified client memory associated 579 with that RPC. 581 6.7. Transport Considerations 583 6.7.1. Congestion Avoidance 585 Section 3.1 of [RFC7530] states: 587 Where an NFS version 4 implementation supports operation over the 588 IP network protocol, the supported transport layer between NFS and 589 IP MUST be an IETF standardized transport protocol that is 590 specified to avoid network congestion; such transports include TCP 591 and the Stream Control Transmission Protocol (SCTP). 593 Section 2.9.1 of [RFC5661] also states: 595 Even if NFS version 4.1 is used over a non-IP network protocol, it 596 is RECOMMENDED that the transport support congestion control. 598 It is permissible for a connectionless transport to be used under 599 NFS version 4.1; however, reliable and in-order delivery of data 600 combined with congestion control by the connectionless transport 601 is REQUIRED. As a consequence, UDP by itself MUST NOT be used as 602 an NFS version 4.1 transport. 604 RPC-over-RDMA version 1 is constructed on a platform of RDMA Reliable 605 Connections [RFC8166] [RFC5041]. RDMA Reliable Connections are 606 reliable, connection-oriented transports that guarantee in-order 607 delivery, meeting all above requirements for NFS version 4 608 transports. 610 6.7.2. Retransmission and Keep-alive 612 NFS version 4 client implementations often rely on a transport-layer 613 keep-alive mechanism to detect when an NFS version 4 server has 614 become unresponsive. When an NFS server is no longer responsive, 615 client-side keep-alive terminates the connection, which in turn 616 triggers reconnection and RPC retransmission. 618 Some RDMA transports (such as Reliable Connections on InfiniBand) 619 have no keep-alive mechanism. Without a disconnect or new RPC 620 traffic, such connections can remain alive long after an NFS server 621 has become unresponsive. Once an NFS client has consumed all 622 available RPC-over-RDMA credits on that transport connection, it will 623 forever await a reply before sending another RPC request. 625 NFS version 4 clients SHOULD reserve one RPC-over-RDMA credit to use 626 for periodic server or connection health assessment. This credit can 627 be used to drive an RPC request on an otherwise idle connection, 628 triggering either a quick affirmative server response or immediate 629 connection termination. 631 In addition to network partition and request loss scenarios, RPC- 632 over-RDMA transport connections can be terminated when a Transport 633 header is malformed, Reply messages are larger than receive 634 resources, or when too many RPC-over-RDMA messages are sent at once. 635 In such cases: 637 o If there is a transport error indicated (ie, RDMA_ERROR) before 638 the disconnect or instead of a disconnect, the requester MUST 639 respond to that error as prescribed by the specification of the 640 RPC transport. Then the NFS version 4 rules for handling 641 retransmission apply. 643 o If there is a transport disconnect and the responder has provided 644 no other response for a request, then only the NFS version 4 rules 645 for handling retransmission apply. 647 7. Extending NFS Upper Layer Bindings 649 RPC programs such as NFS are required to have an Upper Layer Binding 650 specification to interoperate on RPC-over-RDMA version 1 transports 651 [RFC8166]. Via standards action, the Upper Layer Binding specified 652 in this document can be extended to cover versions of the NFS version 653 4 protocol specified after NFS version 4 minor version 2, or 654 separately published extensions to an existing NFS version 4 minor 655 version, as described in [RFC8178]. 657 8. Security Considerations 659 RPC-over-RDMA version 1 supports all RPC security models, including 660 RPCSEC_GSS security and transport-level security [RFC7861]. The 661 choice of what Direct Data Placement mechanism to convey RPC argument 662 and results does not affect this, since it changes only the method of 663 data transfer. Because this document defines only the binding of the 664 NFS protocols atop [RFC8166], all relevant security considerations 665 are therefore to be described at that layer. 667 9. IANA Considerations 669 The use of direct data placement in NFS introduces a need for an 670 additional port number assignment for networks that share traditional 671 UDP and TCP port spaces with RDMA services. The iWARP protocol is 672 such an example [RFC5041] [RFC5040]. 674 For this purpose, a set of transport protocol port number assignments 675 is specified by this document. IANA has assigned the following ports 676 for NFS/RDMA in the IANA port registry, according to the guidelines 677 described in [RFC6335]. 679 nfsrdma 20049/tcp Network File System (NFS) over RDMA 680 nfsrdma 20049/udp Network File System (NFS) over RDMA 681 nfsrdma 20049/sctp Network File System (NFS) over RDMA 683 This document should be listed as the reference for the nfsrdma port 684 assignments. This document does not alter these assignments. 686 10. References 688 10.1. Normative References 690 [RFC1833] Srinivasan, R., "Binding Protocols for ONC RPC Version 2", 691 RFC 1833, DOI 10.17487/RFC1833, August 1995, 692 . 694 [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate 695 Requirement Levels", BCP 14, RFC 2119, 696 DOI 10.17487/RFC2119, March 1997, 697 . 699 [RFC5661] Shepler, S., Ed., Eisler, M., Ed., and D. Noveck, Ed., 700 "Network File System (NFS) Version 4 Minor Version 1 701 Protocol", RFC 5661, DOI 10.17487/RFC5661, January 2010, 702 . 704 [RFC6335] Cotton, M., Eggert, L., Touch, J., Westerlund, M., and S. 705 Cheshire, "Internet Assigned Numbers Authority (IANA) 706 Procedures for the Management of the Service Name and 707 Transport Protocol Port Number Registry", BCP 165, 708 RFC 6335, DOI 10.17487/RFC6335, August 2011, 709 . 711 [RFC7530] Haynes, T., Ed. and D. Noveck, Ed., "Network File System 712 (NFS) Version 4 Protocol", RFC 7530, DOI 10.17487/RFC7530, 713 March 2015, . 715 [RFC7861] Adamson, A. and N. Williams, "Remote Procedure Call (RPC) 716 Security Version 3", RFC 7861, DOI 10.17487/RFC7861, 717 November 2016, . 719 [RFC7862] Haynes, T., "Network File System (NFS) Version 4 Minor 720 Version 2 Protocol", RFC 7862, DOI 10.17487/RFC7862, 721 November 2016, . 723 [RFC8166] Lever, C., Ed., Simpson, W., and T. Talpey, "Remote Direct 724 Memory Access Transport for Remote Procedure Call Version 725 1", RFC 8166, DOI 10.17487/RFC8166, June 2017, 726 . 728 [RFC8167] Lever, C., "Bidirectional Remote Procedure Call on RPC- 729 over-RDMA Transports", RFC 8167, DOI 10.17487/RFC8167, 730 June 2017, . 732 [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC 733 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 734 May 2017, . 736 10.2. Informative References 738 [RFC1094] Nowicki, B., "NFS: Network File System Protocol 739 specification", RFC 1094, DOI 10.17487/RFC1094, March 740 1989, . 742 [RFC1813] Callaghan, B., Pawlowski, B., and P. Staubach, "NFS 743 Version 3 Protocol Specification", RFC 1813, 744 DOI 10.17487/RFC1813, June 1995, 745 . 747 [RFC5040] Recio, R., Metzler, B., Culley, P., Hilland, J., and D. 748 Garcia, "A Remote Direct Memory Access Protocol 749 Specification", RFC 5040, DOI 10.17487/RFC5040, October 750 2007, . 752 [RFC5041] Shah, H., Pinkerton, J., Recio, R., and P. Culley, "Direct 753 Data Placement over Reliable Transports", RFC 5041, 754 DOI 10.17487/RFC5041, October 2007, 755 . 757 [RFC5666] Talpey, T. and B. Callaghan, "Remote Direct Memory Access 758 Transport for Remote Procedure Call", RFC 5666, 759 DOI 10.17487/RFC5666, January 2010, 760 . 762 [RFC5667] Talpey, T. and B. Callaghan, "Network File System (NFS) 763 Direct Data Placement", RFC 5667, DOI 10.17487/RFC5667, 764 January 2010, . 766 [RFC8178] Noveck, D., "Rules for NFSv4 Extensions and Minor 767 Versions", RFC 8178, DOI 10.17487/RFC8178, July 2017, 768 . 770 [XNFS] The Open Group, "Protocols for Interworking: XNFS, Version 771 3W", February 1998. 773 Appendix A. Changes Since RFC 5667 775 Corrections and updates made necessary by new language in [RFC8166] 776 have been introduced. For example, references to deprecated features 777 of RPC-over-RDMA version 1, such as RDMA_MSGP, and the use of the 778 Read list for handling RPC replies, have been removed. The term 779 "mapping" has been replaced with the term "binding" or "Upper Layer 780 Binding" throughout the document. Material that duplicates what is 781 in [RFC8166] has been deleted. 783 Material required by [RFC8166] for Upper Layer Bindings that was not 784 present in [RFC5667] has been added. A complete discussion of reply 785 size estimation has been introduced for all protocols covered by the 786 Upper Layer Bindings in this document. 788 Technical corrections have been made. For example, the mention of 789 12KB and 36KB inline thresholds have been removed. The reference to 790 a non-existant NFS version 4 SYMLINK operation has been replaced. 792 The discussion of NFS version 4 COMPOUND handling has been completed. 793 Some changes were made to the algorithm for matching DDP-eligible 794 results to Write chunks. 796 Requirements to ignore extra Read or Write chunks have been removed 797 from the NFS version 2 and 3 Upper Layer Binding, as they conflict 798 with [RFC8166]. 800 A section discussing NFS version 4 retransmission and connection loss 801 has been added. 803 The following additional improvements have been made, relative to 804 [RFC5667]: 806 o An explicit discussion of NFS version 4.0 and NFS version 4.1 807 backchannel operation has replaced the previous treatment of 808 callback operations. 810 o A section describing considerations when an NFS session is in use 811 has been added. 813 o An Upper Layer Binding for NFS version 4.2 has been added. 815 o A section suggesting a mechanism for periodically assessing 816 connection health has been introduced. 818 o Ambiguous or erroneous uses of RFC2119 terms have been corrected. 820 o References to obsolete RFCs have been updated. 822 o An IANA Considerations Section has been added, which specifies the 823 port assignments for NFS/RDMA. This replaces the example 824 assignment that appeared in [RFC5666]. 826 o Code excerpts have been removed, and figures have been modernized. 828 Acknowledgments 830 The author gratefully acknowledges the work of Brent Callaghan and 831 Tom Talpey on the original NFS Direct Data Placement specification 832 [RFC5667]. Tom contributed the text of Section 6.4.2. 834 Dave Noveck provided excellent review, constructive suggestions, and 835 consistent navigational guidance throughout the process of drafting 836 this document. Dave contributed the text of Section 6.6 and 837 Section 7, and insisted on precise discussion of reply size 838 estimation. 840 Thanks to Karen Deitke for her sharp observations about idempotency, 841 NFS COMPOUNDs, and NFS sessions. 843 Special thanks go to Transport Area Director Spencer Dawkins, NFSV4 844 Working Group Chair and Document Shepherd Spencer Shepler, and NFSV4 845 Working Group Secretary Thomas Haynes for their support. The author 846 also wishes to thank Bill Baker and Greg Marsden for their support of 847 this work. 849 Author's Address 850 Charles Lever 851 Oracle Corporation 852 1015 Granger Avenue 853 Ann Arbor, MI 48104 854 United States of America 856 Phone: +1 248 816 6463 857 Email: chuck.lever@oracle.com