RFC 8540
Stream Control Transmission Protocol: Errata and Issues in RFC 4960
Internet Engineering Task Force (IETF) R. Stewart Request for Comments: 8540 Netflix, Inc. Category: Informational M. Tuexen ISSN: 2070-1721 Muenster Univ. of Appl. Sciences M. Proshin Ericsson February 2019 Stream Control Transmission Protocol: Errata and Issues in RFC 4960Abstract
This document is a compilation of issues found since the publication of RFC 4960 in September 2007, based on experience with implementing, testing, and using the Stream Control Transmission Protocol (SCTP) along with the suggested fixes. This document provides deltas to RFC 4960 and is organized in a time-ordered way. The issues are listed in the order in which they were brought up. Because some text is changed several times, the last delta in the text is the one that should be applied. In addition to the deltas, a description of each problem and the details of the solution for each are also provided. Status of This Memo This document is not an Internet Standards Track specification; it is published for informational purposes. This document is a product of the Internet Engineering Task Force (IETF). It represents the consensus of the IETF community. It has received public review and has been approved for publication by the Internet Engineering Steering Group (IESG). Not all documents approved by the IESG are candidates for any level of Internet Standard; see Section 2 of RFC 7841. Information about the current status of this document, any errata, and how to provide feedback on it may be obtained at https://www.rfc-editor.org/info/rfc8540.
Copyright Notice Copyright (c) 2019 IETF Trust and the persons identified as the document authors. All rights reserved. This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License.Table of Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 3 2. Conventions . . . . . . . . . . . . . . . . . . . . . . . . . 4 3. Corrections to RFC 4960 . . . . . . . . . . . . . . . . . . . 4 3.1. Path Error Counter Threshold Handling . . . . . . . . . . 4 3.2. Upper-Layer Protocol Shutdown Request Handling . . . . . 5 3.3. Registration of New Chunk Types . . . . . . . . . . . . . 6 3.4. Variable Parameters for INIT Chunks . . . . . . . . . . . 7 3.5. CRC32c Sample Code on 64-Bit Platforms . . . . . . . . . 8 3.6. Endpoint Failure Detection . . . . . . . . . . . . . . . 9 3.7. Data Transmission Rules . . . . . . . . . . . . . . . . . 10 3.8. T1-Cookie Timer . . . . . . . . . . . . . . . . . . . . . 11 3.9. Miscellaneous Typos . . . . . . . . . . . . . . . . . . . 12 3.10. CRC32c Sample Code . . . . . . . . . . . . . . . . . . . 19 3.11. partial_bytes_acked after T3-rtx Expiration . . . . . . . 19 3.12. Order of Adjustments of partial_bytes_acked and cwnd . . 20 3.13. HEARTBEAT ACK and the Association Error Counter . . . . . 21 3.14. Path for Fast Retransmission . . . . . . . . . . . . . . 22 3.15. Transmittal in Fast Recovery . . . . . . . . . . . . . . 23 3.16. Initial Value of ssthresh . . . . . . . . . . . . . . . . 24 3.17. Automatically CONFIRMED Addresses . . . . . . . . . . . . 25 3.18. Only One Packet after Retransmission Timeout . . . . . . 26 3.19. INIT ACK Path for INIT in COOKIE-WAIT State . . . . . . . 27 3.20. Zero Window Probing and Unreachable Primary Path . . . . 28 3.21. Normative Language in Section 10 of RFC 4960 . . . . . . 29 3.22. Increase of partial_bytes_acked in Congestion Avoidance . 32 3.23. Inconsistent Handling of Notifications . . . . . . . . . 33 3.24. SACK.Delay Not Listed as a Protocol Parameter . . . . . . 37 3.25. Processing of Chunks in an Incoming SCTP Packet . . . . . 39 3.26. Increasing the cwnd in the Congestion Avoidance Phase . . 41 3.27. Refresh of cwnd and ssthresh after Idle Period . . . . . 43 3.28. Window Updates after Receiver Window Opens Up . . . . . . 45
3.29. Path of DATA and Reply Chunks . . . . . . . . . . . . . . 46
3.30. "Outstanding Data", "Flightsize", and "Data in Flight"
Key Terms . . . . . . . . . . . . . . . . . . . . . . . . 47
3.31. Degradation of cwnd due to Max.Burst . . . . . . . . . . 49
3.32. Reduction of RTO.Initial . . . . . . . . . . . . . . . . 50
3.33. Ordering of Bundled SACK and ERROR Chunks . . . . . . . . 51
3.34. Undefined Parameter Returned by RECEIVE Primitive . . . . 52
3.35. DSCP Changes . . . . . . . . . . . . . . . . . . . . . . 53
3.36. Inconsistent Handling of ICMPv4 and ICMPv6 Messages . . . 55
3.37. Handling of Soft Errors . . . . . . . . . . . . . . . . . 56
3.38. Honoring cwnd . . . . . . . . . . . . . . . . . . . . . . 57
3.39. Zero Window Probing . . . . . . . . . . . . . . . . . . . 58
3.40. Updating References regarding ECN . . . . . . . . . . . . 60
3.41. Host Name Address Parameter Deprecated . . . . . . . . . 62
3.42. Conflicting Text regarding the 'Supported Address Types'
Parameter . . . . . . . . . . . . . . . . . . . . . . . . 66
3.43. Integration of RFC 6096 . . . . . . . . . . . . . . . . . 67
3.44. Integration of RFC 6335 . . . . . . . . . . . . . . . . . 70
3.45. Integration of RFC 7053 . . . . . . . . . . . . . . . . . 72
3.46. CRC32c Code Improvements . . . . . . . . . . . . . . . . 76
3.47. Clarification of Gap Ack Blocks in SACK Chunks . . . . . 87
3.48. Handling of SSN Wraparounds . . . . . . . . . . . . . . . 89
3.49. Update to RFC 2119 Boilerplate Text . . . . . . . . . . . 90
3.50. Removal of Text (Previously Missed in RFC 4960) . . . . . 91
4. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 91
5. Security Considerations . . . . . . . . . . . . . . . . . . . 92
6. References . . . . . . . . . . . . . . . . . . . . . . . . . 92
6.1. Normative References . . . . . . . . . . . . . . . . . . 92
6.2. Informative References . . . . . . . . . . . . . . . . . 92
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . 94
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 94
1. Introduction
This document contains a compilation of all defects for [RFC4960]
("Stream Control Transmission Protocol") that were found up until the
publication of this document. These defects may be of an editorial
or technical nature. This document may be thought of as a companion
document to be used in the implementation of the Stream Control
Transmission Protocol (SCTP) to clarify errors in the original SCTP
document.
This document provides a history of the changes that will be compiled
into a bis document for [RFC4960]. It is structured similarly to
[RFC4460].
Each error will be detailed within this document in the form of: o The problem description, o The text quoted from [RFC4960], o The replacement text that should be placed into an upcoming bis document, and o A description of the solution. Note that when reading this document one must use care to ensure that a field or item is not updated later on within the document. Since this document is a historical record of the sequential changes that have been found necessary at various interop events and through discussion on the Transport Area Working Group mailing list, the last delta in the text is the one that should be applied.2. Conventions
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in BCP 14 [RFC2119] [RFC8174] when, and only when, they appear in all capitals, as shown here.3. Corrections to RFC 4960
3.1. Path Error Counter Threshold Handling
3.1.1. Description of the Problem
The handling of the 'Path.Max.Retrans' parameter is described in Sections 8.2 and 8.3 of [RFC4960] in an inconsistent way. Whereas Section 8.2 of [RFC4960] says that a path is marked inactive when the path error counter exceeds the threshold, Section 8.3 of [RFC4960] says that the path is marked inactive when the path error counter reaches the threshold. This issue was reported as an errata for [RFC4960] with Errata ID 1440.
3.1.2. Text Changes to the Document
--------- Old text: (Section 8.3) --------- When the value of this counter reaches the protocol parameter 'Path.Max.Retrans', the endpoint should mark the corresponding destination address as inactive if it is not so marked, and may also optionally report to the upper layer the change of reachability of this destination address. After this, the endpoint should continue HEARTBEAT on this destination address but should stop increasing the counter. --------- New text: (Section 8.3) --------- When the value of this counter exceeds the protocol parameter 'Path.Max.Retrans', the endpoint SHOULD mark the corresponding destination address as inactive if it is not so marked and MAY also optionally report to the upper layer the change in reachability of this destination address. After this, the endpoint SHOULD continue HEARTBEAT on this destination address but SHOULD stop increasing the counter. This text has been modified by multiple errata. It is further updated in Section 3.23.3.1.3. Solution Description
The intended state change should happen when the threshold is exceeded.3.2. Upper-Layer Protocol Shutdown Request Handling
3.2.1. Description of the Problem
Section 9.2 of [RFC4960] describes the handling of received SHUTDOWN chunks in the SHUTDOWN-RECEIVED state instead of the handling of shutdown requests from its upper layer in this state. This issue was reported as an errata for [RFC4960] with Errata ID 1574.
3.2.2. Text Changes to the Document
--------- Old text: (Section 9.2) --------- Once an endpoint has reached the SHUTDOWN-RECEIVED state, it MUST NOT send a SHUTDOWN in response to a ULP request, and should discard subsequent SHUTDOWN chunks. --------- New text: (Section 9.2) --------- Once an endpoint has reached the SHUTDOWN-RECEIVED state, it MUST ignore ULP shutdown requests but MUST continue responding to SHUTDOWN chunks from its peer. This text is in final form and is not further updated in this document.3.2.3. Solution Description
The text never intended that the SCTP endpoint ignore SHUTDOWN chunks from its peer. If it did, the endpoints could never gracefully terminate associations in some cases.3.3. Registration of New Chunk Types
3.3.1. Description of the Problem
Section 14.1 of [RFC4960] should deal with new chunk types; however, the text only refers to parameter types. This issue was reported as an errata for [RFC4960] with Errata ID 2592.3.3.2. Text Changes to the Document
--------- Old text: (Section 14.1) --------- The assignment of new chunk parameter type codes is done through an IETF Consensus action, as defined in [RFC2434]. Documentation of the chunk parameter MUST contain the following information:
--------- New text: (Section 14.1) --------- The assignment of new chunk type codes is done through an IETF Consensus action, as defined in [RFC8126]. Documentation for the chunk type MUST contain the following information: This text has been modified by multiple errata. It is further updated in Section 3.43.3.3.3. Solution Description
The new text refers to chunk types as intended and changes the reference to [RFC8126].3.4. Variable Parameters for INIT Chunks
3.4.1. Description of the Problem
In Section 3.3.2 of [RFC4960], newlines in wrong places break the layout of the table of variable parameters for the INIT chunk. This issue was reported as an errata for [RFC4960] with Errata ID 3291 and Errata ID 3804.3.4.2. Text Changes to the Document
--------- Old text: (Section 3.3.2) --------- Variable Parameters Status Type Value ------------------------------------------------------------- IPv4 Address (Note 1) Optional 5 IPv6 Address (Note 1) Optional 6 Cookie Preservative Optional 9 Reserved for ECN Capable (Note 2) Optional 32768 (0x8000) Host Name Address (Note 3) Optional 11 Supported Address Types (Note 4) Optional 12
--------- New text: (Section 3.3.2) --------- Variable Parameters Status Type Value ------------------------------------------------------------- IPv4 Address (Note 1) Optional 5 IPv6 Address (Note 1) Optional 6 Cookie Preservative Optional 9 Reserved for ECN Capable (Note 2) Optional 32768 (0x8000) Host Name Address (Note 3) Optional 11 Supported Address Types (Note 4) Optional 12 This text is in final form and is not further updated in this document.3.4.3. Solution Description
The formatting of the table is corrected.3.5. CRC32c Sample Code on 64-Bit Platforms
3.5.1. Description of the Problem
The sample code for CRC32c computation, as provided in [RFC4960], assumes that a variable of type unsigned long uses 32 bits. This is not true on some 64-bit platforms (for example, platforms that use LP64). This issue was reported as an errata for [RFC4960] with Errata ID 3423.3.5.2. Text Changes to the Document
--------- Old text: (Appendix C) --------- unsigned long generate_crc32c(unsigned char *buffer, unsigned int length) { unsigned int i; unsigned long crc32 = ~0L;
--------- New text: (Appendix C) --------- unsigned long generate_crc32c(unsigned char *buffer, unsigned int length) { unsigned int i; unsigned long crc32 = 0xffffffffL; This text has been modified by multiple errata. It is further updated in Section 3.10 and again in Section 3.46.3.5.3. Solution Description
The new text uses 0xffffffffL instead of ~0L; this gives the same value on platforms using 32 bits or 64 bits for variables of type unsigned long.3.6. Endpoint Failure Detection
3.6.1. Description of the Problem
The handling of the association error counter defined in Section 8.1 of [RFC4960] can result in an association failure even if the path used for data transmission is available (but idle). This issue was reported as an errata for [RFC4960] with Errata ID 3788.3.6.2. Text Changes to the Document
--------- Old text: (Section 8.1) --------- An endpoint shall keep a counter on the total number of consecutive retransmissions to its peer (this includes retransmissions to all the destination transport addresses of the peer if it is multi-homed), including unacknowledged HEARTBEAT chunks. --------- New text: (Section 8.1) --------- An endpoint SHOULD keep a counter on the total number of consecutive retransmissions to its peer (this includes data retransmissions to all the destination transport addresses of the peer if it is
multi-homed), including the number of unacknowledged HEARTBEAT chunks observed on the path that is currently used for data transfer. Unacknowledged HEARTBEAT chunks observed on paths different from the path currently used for data transfer SHOULD NOT increment the association error counter, as this could lead to association closure even if the path that is currently used for data transfer is available (but idle). This text has been modified by multiple errata. It is further updated in Section 3.23.3.6.3. Solution Description
A more refined handling of the association error counter is defined.3.7. Data Transmission Rules
3.7.1. Description of the Problem
When integrating the changes to Section 6.1 A) of [RFC2960] as described in Section 2.15.2 of [RFC4460], some text was duplicated and became the final paragraph of Section 6.1 A) of [RFC4960]. This issue was reported as an errata for [RFC4960] with Errata ID 4071.3.7.2. Text Changes to the Document
--------- Old text: (Section 6.1 A)) --------- The sender MUST also have an algorithm for sending new DATA chunks to avoid silly window syndrome (SWS) as described in [RFC0813]. The algorithm can be similar to the one described in Section 4.2.3.4 of [RFC1122]. However, regardless of the value of rwnd (including if it is 0), the data sender can always have one DATA chunk in flight to the receiver if allowed by cwnd (see rule B below). This rule allows the sender to probe for a change in rwnd that the sender missed due to the SACK having been lost in transit from the data receiver to the data sender.
--------- New text: (Section 6.1 A)) --------- The sender MUST also have an algorithm for sending new DATA chunks to avoid silly window syndrome (SWS) as described in [RFC1122]. The algorithm can be similar to the algorithm described in Section 4.2.3.4 of [RFC1122]. This text is in final form and is not further updated in this document.3.7.3. Solution Description
The last paragraph of Section 6.1 A) is removed, as had been intended in Section 2.15.2 of [RFC4460].3.8. T1-Cookie Timer
3.8.1. Description of the Problem
Figure 4 of [RFC4960] illustrates the SCTP association setup. However, it incorrectly shows that the T1-init timer is used in the COOKIE-ECHOED state, whereas the T1-cookie timer should have been used instead. This issue was reported as an errata for [RFC4960] with Errata ID 4400.3.8.2. Text Changes to the Document
--------- Old text: (Section 5.1.6, Figure 4) --------- COOKIE ECHO [Cookie_Z] ------\ (Start T1-init timer) \ (Enter COOKIE-ECHOED state) \---> (build TCB enter ESTABLISHED state) /---- COOKIE-ACK / (Cancel T1-init timer, <-----/ Enter ESTABLISHED state)
--------- New text: (Section 5.1.6, Figure 4) --------- COOKIE ECHO [Cookie_Z] ------\ (Start T1-cookie timer) \ (Enter COOKIE-ECHOED state) \---> (build TCB, enter ESTABLISHED state) /---- COOKIE-ACK / (Cancel T1-cookie timer, <---/ enter ESTABLISHED state) This text has been modified by multiple errata. It is further updated in Section 3.9.3.8.3. Solution Description
The figure is changed such that the T1-cookie timer is used instead of the T1-init timer.
(page 12 continued on part 2)
