< draft-ietf-6lo-fragment-recovery-08.txt   draft-ietf-6lo-fragment-recovery-09.txt >
6lo P. Thubert, Ed. 6lo P. Thubert, Ed.
Internet-Draft Cisco Systems Internet-Draft Cisco Systems
Updates: 4944 (if approved) 28 November 2019 Updates: 4944 (if approved) 4 February 2020
Intended status: Standards Track Intended status: Standards Track
Expires: 31 May 2020 Expires: 7 August 2020
6LoWPAN Selective Fragment Recovery 6LoWPAN Selective Fragment Recovery
draft-ietf-6lo-fragment-recovery-08 draft-ietf-6lo-fragment-recovery-09
Abstract Abstract
This draft updates RFC 4944 with a simple protocol to recover This draft updates RFC 4944 with a simple protocol to recover
individual fragments across a route-over mesh network, with a minimal individual fragments across a route-over mesh network, with a minimal
flow control to protect the network against bloat. flow control to protect the network against bloat.
Status of This Memo Status of This Memo
This Internet-Draft is submitted in full conformance with the This Internet-Draft is submitted in full conformance with the
skipping to change at page 1, line 33 skipping to change at page 1, line 33
Internet-Drafts are working documents of the Internet Engineering Internet-Drafts are working documents of the Internet Engineering
Task Force (IETF). Note that other groups may also distribute Task Force (IETF). Note that other groups may also distribute
working documents as Internet-Drafts. The list of current Internet- working documents as Internet-Drafts. The list of current Internet-
Drafts is at https://datatracker.ietf.org/drafts/current/. Drafts is at https://datatracker.ietf.org/drafts/current/.
Internet-Drafts are draft documents valid for a maximum of six months Internet-Drafts are draft documents valid for a maximum of six months
and may be updated, replaced, or obsoleted by other documents at any and may be updated, replaced, or obsoleted by other documents at any
time. It is inappropriate to use Internet-Drafts as reference time. It is inappropriate to use Internet-Drafts as reference
material or to cite them other than as "work in progress." material or to cite them other than as "work in progress."
This Internet-Draft will expire on 31 May 2020. This Internet-Draft will expire on 7 August 2020.
Copyright Notice Copyright Notice
Copyright (c) 2019 IETF Trust and the persons identified as the Copyright (c) 2020 IETF Trust and the persons identified as the
document authors. All rights reserved. document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal This document is subject to BCP 78 and the IETF Trust's Legal
Provisions Relating to IETF Documents (https://trustee.ietf.org/ Provisions Relating to IETF Documents (https://trustee.ietf.org/
license-info) in effect on the date of publication of this document. license-info) in effect on the date of publication of this document.
Please review these documents carefully, as they describe your rights Please review these documents carefully, as they describe your rights
and restrictions with respect to this document. Code Components and restrictions with respect to this document. Code Components
extracted from this document must include Simplified BSD License text extracted from this document must include Simplified BSD License text
as described in Section 4.e of the Trust Legal Provisions and are as described in Section 4.e of the Trust Legal Provisions and are
provided without warranty as described in the Simplified BSD License. provided without warranty as described in the Simplified BSD License.
skipping to change at page 2, line 22 skipping to change at page 2, line 22
3. Updating RFC 4944 . . . . . . . . . . . . . . . . . . . . . . 6 3. Updating RFC 4944 . . . . . . . . . . . . . . . . . . . . . . 6
4. Extending draft-ietf-6lo-minimal-fragment . . . . . . . . . . 6 4. Extending draft-ietf-6lo-minimal-fragment . . . . . . . . . . 6
4.1. Slack in the First Fragment . . . . . . . . . . . . . . . 7 4.1. Slack in the First Fragment . . . . . . . . . . . . . . . 7
4.2. Gap between frames . . . . . . . . . . . . . . . . . . . 7 4.2. Gap between frames . . . . . . . . . . . . . . . . . . . 7
4.3. Modifying the First Fragment . . . . . . . . . . . . . . 7 4.3. Modifying the First Fragment . . . . . . . . . . . . . . 7
5. New Dispatch types and headers . . . . . . . . . . . . . . . 8 5. New Dispatch types and headers . . . . . . . . . . . . . . . 8
5.1. Recoverable Fragment Dispatch type and Header . . . . . . 8 5.1. Recoverable Fragment Dispatch type and Header . . . . . . 8
5.2. RFRAG Acknowledgment Dispatch type and Header . . . . . . 11 5.2. RFRAG Acknowledgment Dispatch type and Header . . . . . . 11
6. Fragments Recovery . . . . . . . . . . . . . . . . . . . . . 12 6. Fragments Recovery . . . . . . . . . . . . . . . . . . . . . 12
6.1. Forwarding Fragments . . . . . . . . . . . . . . . . . . 14 6.1. Forwarding Fragments . . . . . . . . . . . . . . . . . . 14
6.1.1. Upon the first fragment . . . . . . . . . . . . . . . 14 6.1.1. Receiving the first fragment . . . . . . . . . . . . 15
6.1.2. Upon the next fragments . . . . . . . . . . . . . . . 15 6.1.2. Receiving the next fragments . . . . . . . . . . . . 15
6.2. Upon the RFRAG Acknowledgments . . . . . . . . . . . . . 16 6.2. Receiving RFRAG Acknowledgments . . . . . . . . . . . . . 16
6.3. Aborting the Transmission of a Fragmented Packet . . . . 16 6.3. Aborting the Transmission of a Fragmented Packet . . . . 16
6.4. Applying Recoverable Fragmentation along a Diverse 6.4. Applying Recoverable Fragmentation along a Diverse
Path . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Path . . . . . . . . . . . . . . . . . . . . . . . . . . 17
7. Management Considerations . . . . . . . . . . . . . . . . . . 17 7. Management Considerations . . . . . . . . . . . . . . . . . . 17
7.1. Protocol Parameters . . . . . . . . . . . . . . . . . . . 17 7.1. Protocol Parameters . . . . . . . . . . . . . . . . . . . 17
7.2. Observing the network . . . . . . . . . . . . . . . . . . 19 7.2. Observing the network . . . . . . . . . . . . . . . . . . 19
8. Security Considerations . . . . . . . . . . . . . . . . . . . 19 8. Security Considerations . . . . . . . . . . . . . . . . . . . 19
9. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 20 9. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 20
10. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 20 10. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . 20
11. Normative References . . . . . . . . . . . . . . . . . . . . 20 11. Normative References . . . . . . . . . . . . . . . . . . . . 20
12. Informative References . . . . . . . . . . . . . . . . . . . 21 12. Informative References . . . . . . . . . . . . . . . . . . . 21
Appendix A. Rationale . . . . . . . . . . . . . . . . . . . . . 24 Appendix A. Rationale . . . . . . . . . . . . . . . . . . . . . 24
Appendix B. Requirements . . . . . . . . . . . . . . . . . . . . 25 Appendix B. Requirements . . . . . . . . . . . . . . . . . . . . 25
Appendix C. Considerations On Flow Control . . . . . . . . . . . 26 Appendix C. Considerations on Flow Control . . . . . . . . . . . 26
Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 27 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 27
1. Introduction 1. Introduction
In most Low Power and Lossy Network (LLN) applications, the bulk of In most Low Power and Lossy Network (LLN) applications, the bulk of
the traffic consists of small chunks of data (in the order few bytes the traffic consists of small chunks of data (on the order of a few
to a few tens of bytes) at a time. Given that an IEEE Std. 802.15.4 bytes to a few tens of bytes) at a time. Given that an IEEE Std.
[IEEE.802.15.4] frame can carry a payload of 74 bytes or more, 802.15.4 [IEEE.802.15.4] frame can carry a payload of 74 bytes or
fragmentation is usually not required. However, and though this more, fragmentation is usually not required. However, and though
happens only occasionally, a number of mission critical applications this happens only occasionally, a number of mission critical
do require the capability to transfer larger chunks of data, for applications do require the capability to transfer larger chunks of
instance to support the firmware upgrade of the LLN nodes or the data, for instance to support the firmware upgrade of the LLN nodes
extraction of logs from LLN nodes. In the former case, the large or the extraction of logs from LLN nodes. In the former case, the
chunk of data is transferred to the LLN node, whereas in the latter, large chunk of data is transferred to the LLN node, whereas in the
the large chunk flows away from the LLN node. In both cases, the latter, the large chunk flows away from the LLN node. In both cases,
size can be on the order of 10 kilobytes or more and an end-to-end the size can be on the order of 10 kilobytes or more and an end-to-
reliable transport is required. end reliable transport is required.
"Transmission of IPv6 Packets over IEEE 802.15.4 Networks" [RFC4944] "Transmission of IPv6 Packets over IEEE 802.15.4 Networks" [RFC4944]
defines the original 6LoWPAN datagram fragmentation mechanism for defines the original 6LoWPAN datagram fragmentation mechanism for
LLNs. One critical issue with this original design is that routing LLNs. One critical issue with this original design is that routing
an IPv6 [RFC8200] packet across a route-over mesh requires to an IPv6 [RFC8200] packet across a route-over mesh requires
reassemble the full packet at each hop, which may cause latency along reassembling the full packet at each hop, which may cause latency
a path and an overall buffer bloat in the network. The "6TiSCH along a path and an overall buffer bloat in the network. The "6TiSCH
Architecture" [I-D.ietf-6tisch-architecture] recommends to use a Architecture" [I-D.ietf-6tisch-architecture] recommends using a
fragment forwarding (FF) technique to alleviate those undesirable fragment forwarding (FF) technique to alleviate those undesirable
effects. "LLN Minimal Fragment Forwarding" effects. "LLN Minimal Fragment Forwarding"
[I-D.ietf-6lo-minimal-fragment] specifies the general behavior that [I-D.ietf-6lo-minimal-fragment] specifies the general behavior that
all FF techniques including this specification follow, and presents all FF techniques including this specification follow, and presents
the associated caveats. In particular, the routing information is the associated caveats. In particular, the routing information is
fully indicated in the first fragment, which is always forwarded fully indicated in the first fragment, which is always forwarded
first. A state is formed and used to forward all the next fragments first. A state is formed and used to forward all the next fragments
along the same path. The datagram_tag is locally significant to the along the same path. The datagram_tag is locally significant to the
Layer-2 source of the packet and is swapped at each hop. Layer-2 source of the packet and is swapped at each hop.
skipping to change at page 3, line 37 skipping to change at page 3, line 37
technique that is compatible with [RFC4944] without the need to technique that is compatible with [RFC4944] without the need to
define a new protocol. However, adding that capability alone to the define a new protocol. However, adding that capability alone to the
local implementation of the original 6LoWPAN fragmentation would not local implementation of the original 6LoWPAN fragmentation would not
address the inherent fragility of fragmentation (see address the inherent fragility of fragmentation (see
[I-D.ietf-intarea-frag-fragile]) in particular the issues of [I-D.ietf-intarea-frag-fragile]) in particular the issues of
resources locked on the receiver and the wasted transmissions due to resources locked on the receiver and the wasted transmissions due to
the loss of a single fragment in a whole datagram. [Kent] compares the loss of a single fragment in a whole datagram. [Kent] compares
the unreliable delivery of fragments with a mechanism it calls the unreliable delivery of fragments with a mechanism it calls
"selective acknowledgements" that recovers the loss of a fragment "selective acknowledgements" that recovers the loss of a fragment
individually. The paper illustrates the benefits that can be derived individually. The paper illustrates the benefits that can be derived
from such a method in figures 1, 2 and 3, pages 6 and 7. [RFC4944] from such a method in figures 1, 2 and 3, on pages 6 and 7.
as no selective recovery and the whole datagram fails when one [RFC4944] as no selective recovery and the whole datagram fails when
fragment is not delivered to the destination 6LoWPAN endpoint. one fragment is not delivered to the destination 6LoWPAN endpoint.
Constrained memory resources are blocked on the receiver until the Constrained memory resources are blocked on the receiver until the
receiver times out, possibly causing the loss of subsequent packets receiver times out, possibly causing the loss of subsequent packets
that can not be received for the lack of buffers. that cannot be received for the lack of buffers.
That problem is exacerbated when forwarding fragments over multiple That problem is exacerbated when forwarding fragments over multiple
hops since a loss at an intermediate hop will not be discovered by hops since a loss at an intermediate hop will not be discovered by
either the source or the destination, and the source will keep on either the source or the destination, and the source will keep on
sending fragments, wasting even more resources in the network and sending fragments, wasting even more resources in the network and
possibly contributing to the condition that caused the loss to no possibly contributing to the condition that caused the loss to no
avail since the datagram cannot arrive in its entirety. RFC 4944 is avail since the datagram cannot arrive in its entirety. RFC 4944 is
also missing signaling to abort a multi-fragment transmission at any also missing signaling to abort a multi-fragment transmission at any
time and from either end, and, if the capability to forward fragments time and from either end, and, if the capability to forward fragments
is implemented, clean up the related state in the network. It is is implemented, clean up the related state in the network. It is
also lacking flow control capabilities to avoid participating to a also lacking flow control capabilities to avoid participating in
congestion that may in turn cause the loss of a fragment and congestion that may in turn cause the loss of a fragment and
potentially the retransmission of the full datagram. potentially the retransmission of the full datagram.
This specification provides a method to forward fragments across a This specification provides a method to forward fragments across a
multi-hop route-over mesh, and a selective acknowledgment to recover multi-hop route-over mesh, and a selective acknowledgment to recover
individual fragments between 6LoWPAN endpoints. The method is individual fragments between 6LoWPAN endpoints. The method is
designed to limit congestion loss in the network and addresses the designed to limit congestion loss in the network and addresses the
requirements that are detailed in Appendix B. requirements that are detailed in Appendix B.
2. Terminology 2. Terminology
skipping to change at page 4, line 39 skipping to change at page 4, line 39
Low-Power Wireless Personal Area Network (6LoWPAN) Routing" [RFC6606] Low-Power Wireless Personal Area Network (6LoWPAN) Routing" [RFC6606]
"LLN Minimal Fragment Forwarding" [I-D.ietf-6lo-minimal-fragment] "LLN Minimal Fragment Forwarding" [I-D.ietf-6lo-minimal-fragment]
introduces the generic concept of a Virtual Reassembly Buffer (VRB) introduces the generic concept of a Virtual Reassembly Buffer (VRB)
and specifies behaviours and caveats that are common to a large and specifies behaviours and caveats that are common to a large
family of FF techniques including this, which fully inherits from family of FF techniques including this, which fully inherits from
that specification. that specification.
Past experience with fragmentation has shown that misassociated or Past experience with fragmentation has shown that misassociated or
lost fragments can lead to poor network behavior and, occasionally, lost fragments can lead to poor network behavior and, occasionally,
trouble at application layer. The reader is encouraged to read "IPv4 trouble at the application layer. The reader is encouraged to read
Reassembly Errors at High Data Rates" [RFC4963] and follow the "IPv4 Reassembly Errors at High Data Rates" [RFC4963] and follow the
references for more information. references for more information.
That experience led to the definition of "Path MTU discovery" That experience led to the definition of "Path MTU discovery"
[RFC8201] (PMTUD) protocol that limits fragmentation over the [RFC8201] (PMTUD) protocol that limits fragmentation over the
Internet. Internet.
Specifically in the case of UDP, valuable additional information can Specifically in the case of UDP, valuable additional information can
be found in "UDP Usage Guidelines for Application Designers" be found in "UDP Usage Guidelines for Application Designers"
[RFC8085]. [RFC8085].
skipping to change at page 5, line 39 skipping to change at page 5, line 39
expanding a 6LoWPAN header from/to a full IPv6 packet. The expanding a 6LoWPAN header from/to a full IPv6 packet. The
6LoWPAN endpoints are the points where fragmentation and 6LoWPAN endpoints are the points where fragmentation and
reassembly take place. reassembly take place.
Compressed Form: This specification uses the generic term Compressed Compressed Form: This specification uses the generic term Compressed
Form to refer to the format of a datagram after the action of Form to refer to the format of a datagram after the action of
[RFC6282] and possibly [RFC8138] for RPL [RFC6550] artifacts. [RFC6282] and possibly [RFC8138] for RPL [RFC6550] artifacts.
datagram_size: The size of the datagram in its Compressed Form datagram_size: The size of the datagram in its Compressed Form
before it is fragmented. The datagram_size is expressed in a unit before it is fragmented. The datagram_size is expressed in a unit
that depends on the MAC layer technology, by default a byte. that depends on the MAC address layer technology, by default a
byte.
datagram_tag: An identifier of a datagram that is locally unique to datagram_tag: An identifier of a datagram that is locally unique to
the Layer-2 sender. Associated with the MAC address of the the Layer-2 sender. Associated with the MAC addressof the sender,
sender, this becomes a globally unique identifier for the this becomes a globally unique identifier for the datagram.
datagram.
fragment_offset: The offset of a particular fragment of a datagram fragment_offset: The offset of a particular fragment of a datagram
in its Compressed Form. The fragment_offset is expressed in a in its Compressed Form. The fragment_offset is expressed in a
unit that depends on the MAC layer technology and is by default a unit that depends on the MAC address layer technology and is by
byte. default a byte.
RFRAG: Recoverable Fragment RFRAG: Recoverable Fragment
RFRAG-ACK: Recoverable Fragment Acknowledgement RFRAG-ACK: Recoverable Fragment Acknowledgement
RFRAG Acknowledgment Request: An RFRAG with the Acknowledgement RFRAG Acknowledgment Request: An RFRAG with the Acknowledgement
Request flag ('X' flag) set. Request flag ('X' flag) set.
NULL bitmap: Refers to a bitmap with all bits set to zero. NULL bitmap: Refers to a bitmap with all bits set to zero.
FULL bitmap: Refers to a bitmap with all bits set to one. FULL bitmap: Refers to a bitmap with all bits set to one.
skipping to change at page 6, line 25 skipping to change at page 6, line 25
Reverse: The reverse direction of a LSP path, taken by the RFRAG- Reverse: The reverse direction of a LSP path, taken by the RFRAG-
ACK. ACK.
3. Updating RFC 4944 3. Updating RFC 4944
This specification updates the fragmentation mechanism that is This specification updates the fragmentation mechanism that is
specified in "Transmission of IPv6 Packets over IEEE 802.15.4 specified in "Transmission of IPv6 Packets over IEEE 802.15.4
Networks" [RFC4944] for use in route-over LLNs by providing a model Networks" [RFC4944] for use in route-over LLNs by providing a model
where fragments can be forwarded end-to-end across a 6LoWPAN LLN, and where fragments can be forwarded end-to-end across a 6LoWPAN LLN, and
where fragments that are lost on the way can be recovered where fragments that are lost on the way can be recovered
individually. A new format for fragment is introduced and new individually. A new format for fragments is introduced and new
dispatch types are defined in Section 5. dispatch types are defined in Section 5.
[RFC8138] allows to modify the size of a packet en-route by removing [RFC8138] allows modifying the size of a packet en route by removing
the consumed hops in a compressed Routing Header. It results that the consumed hops in a compressed Routing Header. This requires that
fragment_offset and datagram_size (see Section 2.3) must also be fragment_offset and datagram_size (see Section 2.3) are also modified
modified en-route, whcih is difficult to do in the uncompressed form. en route, which is difficult to do in the uncompressed form. This
This specification expresses those fields in the Compressed Form and specification expresses those fields in the Compressed Form and
allows to modify them en-route (see Section 4.3) easily. allows modifying them en route (see Section 4.3) easily.
Note that consistently with Section 2 of [RFC6282] for the Note that consistent with Section 2 of [RFC6282], for the
fragmentation mechanism described in Section 5.3 of [RFC4944], any fragmentation mechanism described in Section 5.3 of [RFC4944], any
header that cannot fit within the first fragment MUST NOT be header that cannot fit within the first fragment MUST NOT be
compressed when using the fragmentation mechanism described in this compressed when using the fragmentation mechanism described in this
specification. specification.
4. Extending draft-ietf-6lo-minimal-fragment 4. Extending draft-ietf-6lo-minimal-fragment
This specification implements the generic FF technique specified in This specification implements the generic FF technique specified in
"LLN Minimal Fragment Forwarding" [I-D.ietf-6lo-minimal-fragment] in "LLN Minimal Fragment Forwarding" [I-D.ietf-6lo-minimal-fragment] in
a fashion that enables end-to-end recovery of fragments and some a fashion that enables end-to-end recovery of fragments and some
degree of flow control. degree of flow control.
4.1. Slack in the First Fragment 4.1. Slack in the First Fragment
At the time of this writing, [I-D.ietf-6lo-minimal-fragment] allows [I-D.ietf-6lo-minimal-fragment] allows for refragmenting in
for refragmenting in intermediate nodes, meaning that some bytes from intermediate nodes, meaning that some bytes from a given fragment may
a given fragment may be left in the VRB to be added to the next be left in the VRB to be added to the next fragment. The reason for
fragment. The reason for this to happen would be the need for space this happening would be the need for space in the outgoing fragment
in the outgoing fragment that was not needed in the incoming that was not needed in the incoming fragment, for instance because
fragment, for instance because the 6LoWPAN Header Compression is not the 6LoWPAN Header Compression is not as efficient on the outgoing
as efficient on the outgoing link, e.g., if the Interface ID (IID) of link, e.g., if the Interface ID (IID) of the source IPv6 address is
the source IPv6 address is elided by the originator on the first hop elided by the originator on the first hop because it matches the
because it matches the source MAC address, but cannot be on the next source MAC address, but cannot be on the next hops because the source
hops because the source MAC address changes. MAC addresschanges.
This specification cannot allow this operation since fragments are This specification cannot allow this operation since fragments are
recovered end-to-end based on a sequence number. This means that the recovered end-to-end based on a sequence number. This means that the
fragments that contain a 6LoWPAN-compressed header MUST have enough fragments that contain a 6LoWPAN-compressed header MUST have enough
slack to enable a less efficient compression in the next hops that slack to enable a less efficient compression in the next hops that
still fits in one MAC frame. For instance, if the IID of the source still fits in one MAC address frame. For instance, if the IID of the
IPv6 address is elided by the originator, then it MUST compute the source IPv6 address is elided by the originator, then it MUST compute
fragment_size as if the MTU was 8 bytes less. This way, the next hop the fragment_size as if the MTU was 8 bytes less. This way, the next
can restore the source IID to the first fragment without impacting hop can restore the source IID to the first fragment without
the second fragment. impacting the second fragment.
4.2. Gap between frames 4.2. Gap between frames
This specification introduces a concept of Inter-Frame Gap, which is This specification introduces a concept of an inter-frame gap, which
a configurable interval of time between transmissions to a same next is a configurable interval of time between transmissions to a same
hop. In the case of half duplex interfaces, this InterFrameGap next hop. In the case of half duplex interfaces, this inter-frame
ensures that the next hop has progressed the previous frame and is gap ensures that the next hop has completed processing of the
capable of receiving the next one. previous frame and is capable of receiving the next one.
In the case of a mesh operating at a single frequency with In the case of a mesh operating at a single frequency with
omnidirectional antennas, a larger InterFrameGap is required to omnidirectional antennas, a larger inter-frame gap is required to
protect the frame against hidden terminal collisions with the protect the frame against hidden terminal collisions with the
previous frame of a same flow that is still progressing along a previous frame of a same flow that is still progressing along a
common path. common path.
The Inter-Frame Gap is useful even for unfragmented datagrams, but it The inter-frame gap is useful even for unfragmented datagrams, but it
becomes a necessity for fragments that are typically generated in a becomes a necessity for fragments that are typically generated in a
fast sequence and are all sent over the exact same path. fast sequence and are all sent over the exact same path.
4.3. Modifying the First Fragment 4.3. Modifying the First Fragment
The compression of the Hop Limit, of the source and destination The compression of the Hop Limit, of the source and destination
addresses in the IPv6 Header, and of the Routing Header, may change addresses in the IPv6 Header, and of the Routing Header may change en
en-route in a Route-Over mesh LLN. If the size of the first fragment route in a Route-Over mesh LLN. If the size of the first fragment is
is modified, then the intermediate node MUST adapt the datagram_size modified, then the intermediate node MUST adapt the datagram_size to
to reflect that difference. reflect that difference.
The intermediate node MUST also save the difference of datagram_size The intermediate node MUST also save the difference of datagram_size
of the first fragment in the VRB and add it to the datagram_size and of the first fragment in the VRB and add it to the datagram_size and
to the fragment_offset of all the subsequent fragments for that to the fragment_offset of all the subsequent fragments for that
datagram. datagram.
5. New Dispatch types and headers 5. New Dispatch types and headers
This specification enables the 6LoWPAN fragmentation sublayer to This specification enables the 6LoWPAN fragmentation sublayer to
provide an MTU up to 2048 bytes to the upper layer, which can be the provide an MTU up to 2048 bytes to the upper layer, which can be the
6LoWPAN Header Compression sublayer that is defined in the 6LoWPAN Header Compression sublayer that is defined in the
"Compression Format for IPv6 Datagrams" [RFC6282] specification. In "Compression Format for IPv6 Datagrams" [RFC6282] specification. In
order to achieve this, this specification enables the fragmentation order to achieve this, this specification enables the fragmentation
and the reliable transmission of fragments over a multihop 6LoWPAN and the reliable transmission of fragments over a multihop 6LoWPAN
mesh network. mesh network.
This specification provides a technique that is derived from MPLS to This specification provides a technique that is derived from MPLS to
forward individual fragments across a 6LoWPAN route-over mesh without forward individual fragments across a 6LoWPAN route-over mesh without
reassembly at each hop. The datagram_tag is used as a label; it is reassembly at each hop. The datagram_tag is used as a label; it is
locally unique to the node that owns the source MAC address of the locally unique to the node that owns the source MAC addressof the
fragment, so together the MAC address and the label can identify the fragment, so together the MAC addressand the label can identify the
fragment globally. A node may build the datagram_tag in its own fragment globally. A node may build the datagram_tag in its own
locally-significant way, as long as the chosen datagram_tag stays locally-significant way, as long as the chosen datagram_tag stays
unique to the particular datagram for the lifetime of that datagram. unique to the particular datagram for the lifetime of that datagram.
It results that the label does not need to be globally unique but It results that the label does not need to be globally unique but
also that it must be swapped at each hop as the source MAC address also that it must be swapped at each hop as the source MAC
changes. addresschanges.
This specification extends RFC 4944 [RFC4944] with 2 new Dispatch This specification extends RFC 4944 [RFC4944] with 2 new Dispatch
types, for Recoverable Fragment (RFRAG) and for the RFRAG types, for Recoverable Fragment (RFRAG) and for the RFRAG
Acknowledgment back. The new 6LoWPAN Dispatch types are taken from Acknowledgment back. The new 6LoWPAN Dispatch types are taken from
Page 0 [RFC8025] as indicated in Table 1 in Section 9. Page 0 [RFC8025] as indicated in Table 1 in Section 9.
In the following sections, a "datagram_tag" extends the semantics In the following sections, a "datagram_tag" extends the semantics
defined in [RFC4944] Section 5.3."Fragmentation Type and Header". defined in [RFC4944] Section 5.3."Fragmentation Type and Header".
The datagram_tag is a locally unique identifier for the datagram from The datagram_tag is a locally unique identifier for the datagram from
the perspective of the sender. This means that the datagram_tag the perspective of the sender. This means that the datagram_tag
identifies a datagram uniquely in the network when associated with identifies a datagram uniquely in the network when associated with
the source of the datagram. As the datagram gets forwarded, the the source of the datagram. As the datagram gets forwarded, the
source changes and the datagram_tag must be swapped as detailed in source changes and the datagram_tag must be swapped as detailed in
[I-D.ietf-6lo-minimal-fragment]. [I-D.ietf-6lo-minimal-fragment].
5.1. Recoverable Fragment Dispatch type and Header 5.1. Recoverable Fragment Dispatch type and Header
In this specification, if the packet is compressed then the size and In this specification, if the packet is compressed then the size and
offset of the fragments are expressed on the Compressed Form of the offset of the fragments are expressed with respect to the Compressed
packet form as opposed to the uncompressed - native - packet form. Form of the packet form as opposed to the uncompressed (native)
packet form.
The format of the fragment header is shown in Figure 1. It is the The format of the fragment header is shown in Figure 1. It is the
same for all fragments. The format has a length and an offset, as same for all fragments. The format has a length and an offset, as
well as a sequence field. This would be redundant if the offset was well as a sequence field. This would be redundant if the offset was
computed as the product of the sequence by the length, but this is computed as the product of the sequence by the length, but this is
not the case. The position of a fragment in the reassembly buffer is not the case. The position of a fragment in the reassembly buffer is
neither correlated with the value of the sequence field nor with the neither correlated with the value of the sequence field nor with the
order in which the fragments are received. This enables out-of- order in which the fragments are received. This enables out-of-
sequence subfragmenting, e.g., a fragment seq. 5 that is retried end- sequence subfragmenting, e.g., a fragment seq. 5 that is retried end-
to-end as smaller fragments seq. 5, 13 and 14 due to a change of MTU to-end as smaller fragments seq. 5, 13 and 14 due to a change of MTU
skipping to change at page 10, line 5 skipping to change at page 10, line 5
Acknowledgment to indicate the received fragments by setting the Acknowledgment to indicate the received fragments by setting the
individual bits that correspond to their sequence. individual bits that correspond to their sequence.
X: 1 bit; Ack-Request: when set, the sender requires an RFRAG X: 1 bit; Ack-Request: when set, the sender requires an RFRAG
Acknowledgment from the receiver. Acknowledgment from the receiver.
E: 1 bit; Explicit Congestion Notification; the "E" flag is reset by E: 1 bit; Explicit Congestion Notification; the "E" flag is reset by
the source of the fragment and set by intermediate routers to the source of the fragment and set by intermediate routers to
signal that this fragment experienced congestion along its path. signal that this fragment experienced congestion along its path.
Fragment_size: 10 bit unsigned integer; the size of this fragment in Fragment_size: 10-bit unsigned integer; the size of this fragment in
a unit that depends on the MAC layer technology. Unless a unit that depends on the MAC address layer technology. Unless
overridden by a more specific specification, that unit is the overridden by a more specific specification, that unit is the
octet which allows fragments up to 512 bytes. octet, which allows fragments up to 512 bytes.
datagram_tag: 8 bits; an identifier of the datagram that is locally datagram_tag: 8 bits; an identifier of the datagram that is locally
unique to the sender. unique to the sender.
Sequence: 5 bit unsigned integer; the sequence number of the Sequence: 5-bit unsigned integer; the sequence number of the
fragment in the acknowledgement bitmap. Fragments are numbered fragment in the acknowledgement bitmap. Fragments are numbered
[0..N] where N is in [0..31]. A Sequence of 0 indicates the first [0..N] where N is in [0..31]. A Sequence of 0 indicates the first
fragment in a datagram, but non-zero values are not indicative of fragment in a datagram, but non-zero values are not indicative of
the position in the reassembly buffer. the position in the reassembly buffer.
Fragment_offset: 16 bit unsigned integer. Fragment_offset: 16-bit unsigned integer.
When the Fragment_offset is set to a non-0 value, its semantics When the Fragment_offset is set to a non-0 value, its semantics
depend on the value of the Sequence field as follows: depend on the value of the Sequence field as follows:
* For a first fragment (i.e. with a Sequence of 0), this field * For a first fragment (i.e., with a Sequence of 0), this field
indicates the datagram_size of the compressed datagram, to help indicates the datagram_size of the compressed datagram, to help
the receiver allocate an adapted buffer for the reception and the receiver allocate an adapted buffer for the reception and
reassembly operations. The fragment may be stored for local reassembly operations. The fragment may be stored for local
reassembly. Alternatively, it may be routed based on the reassembly. Alternatively, it may be routed based on the
destination IPv6 address. In that case, a VRB state must be destination IPv6 address. In that case, a VRB state must be
installed as described in Section 6.1.1. installed as described in Section 6.1.1.
* When the Sequence is not 0, this field indicates the offset of * When the Sequence is not 0, this field indicates the offset of
the fragment in the Compressed Form of the datagram. The the fragment in the Compressed Form of the datagram. The
fragment may be added to a local reassembly buffer or forwarded fragment may be added to a local reassembly buffer or forwarded
based on an existing VRB as described in Section 6.1.2. based on an existing VRB as described in Section 6.1.2.
skipping to change at page 10, line 48 skipping to change at page 10, line 48
up once the processing of the fragment is complete; the processing up once the processing of the fragment is complete; the processing
of the fragment depends on whether there is a VRB already of the fragment depends on whether there is a VRB already
established for this datagram, and the next hop is still established for this datagram, and the next hop is still
reachable: reachable:
* if a VRB already exists and is not broken, the fragment is to * if a VRB already exists and is not broken, the fragment is to
be forwarded along the associated Label Switched Path (LSP) as be forwarded along the associated Label Switched Path (LSP) as
described in Section 6.1.2, but regardless of the value of the described in Section 6.1.2, but regardless of the value of the
Sequence field; Sequence field;
* else, if the Sequence is 0, then the fragment is to be routed * else, if the Sequence is 0, then the fragment is to be routed
as described in Section 6.1.1 but no state is conserved as described in Section 6.1.1, but no state is conserved
afterwards. In that case, the session if it exists is aborted afterwards. In that case, the session if it exists is aborted
and the packet is also forwarded in an attempt to clean up the and the packet is also forwarded in an attempt to clean up the
next hops as along the path indicated by the IPv6 header next hops along the path indicated by the IPv6 header (possibly
(possibly including a routing header). including a routing header).
If the fragment cannot be forwarded or routed, then an abort If the fragment cannot be forwarded or routed, then an abort
RFRAG-ACK is sent back to the source as described in RFRAG-ACK is sent back to the source as described in
Section 6.1.2. Section 6.1.2.
5.2. RFRAG Acknowledgment Dispatch type and Header 5.2. RFRAG Acknowledgment Dispatch type and Header
This specification also defines a 4-octet RFRAG Acknowledgment bitmap This specification also defines a 4-octet RFRAG Acknowledgment bitmap
that is used by the reassembling endpoint to confirm selectively the that is used by the reassembling endpoint to confirm selectively the
reception of individual fragments. A given offset in the bitmap maps reception of individual fragments. A given offset in the bitmap maps
one to one with a given sequence number and indicates which fragment one-to-one with a given sequence number and indicates which fragment
is acknowledged as follows: is acknowledged as follows:
1 2 3 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| RFRAG Acknowledgment Bitmap | | RFRAG Acknowledgment Bitmap |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
^ ^ ^ ^
| | bitmap indicating whether: | | bitmap indicating whether:
| +----- Fragment with sequence 9 was received | +----- Fragment with sequence 9 was received
+----------------------- Fragment with sequence 0 was received +----------------------- Fragment with sequence 0 was received
Figure 2: RFRAG Acknowledgment bitmap encoding Figure 2: RFRAG Acknowledgment Bitmap Encoding
Figure 3 shows an example Acknowledgment bitmap which indicates that Figure 3 shows an example Acknowledgment bitmap which indicates that
all fragments from sequence 0 to 20 were received, except for all fragments from sequence 0 to 20 were received, except for
fragments 1, 2 and 16 that were lost and must be retried. fragments 1, 2 and 16 were lost and must be retried.
1 2 3 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|1|0|0|1|1|1|1|1|1|1|1|1|1|1|1|1|0|1|1|1|1|0|0|0|0|0|0|0|0|0|0|0| |1|0|0|1|1|1|1|1|1|1|1|1|1|1|1|1|0|1|1|1|1|0|0|0|0|0|0|0|0|0|0|0|
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
Figure 3: Example RFRAG Acknowledgment Bitmap Figure 3: Example RFRAG Acknowledgment Bitmap
The RFRAG Acknowledgment Bitmap is included in a RFRAG Acknowledgment The RFRAG Acknowledgment Bitmap is included in an RFRAG
header, as follows: Acknowledgment header, as follows:
1 2 3 1 2 3
0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|1 1 1 0 1 0 1|E| datagram_tag | |1 1 1 0 1 0 1|E| datagram_tag |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| RFRAG Acknowledgment Bitmap (32 bits) | | RFRAG Acknowledgment Bitmap (32 bits) |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
Figure 4: RFRAG Acknowledgment Dispatch type and Header Figure 4: RFRAG Acknowledgment Dispatch type and Header
skipping to change at page 12, line 25 skipping to change at page 12, line 25
received, as shown in Figure 2. A NULL bitmap that indicates that received, as shown in Figure 2. A NULL bitmap that indicates that
the fragmentation process is aborted. A FULL bitmap that the fragmentation process is aborted. A FULL bitmap that
indicates that the fragmentation process is complete, all indicates that the fragmentation process is complete, all
fragments were received at the reassembly endpoint. fragments were received at the reassembly endpoint.
6. Fragments Recovery 6. Fragments Recovery
The Recoverable Fragment header RFRAG is used to transport a fragment The Recoverable Fragment header RFRAG is used to transport a fragment
and optionally request an RFRAG Acknowledgment that will confirm the and optionally request an RFRAG Acknowledgment that will confirm the
good reception of one or more fragments. An RFRAG Acknowledgment is good reception of one or more fragments. An RFRAG Acknowledgment is
carried as a standalone fragment header (i.e. with no 6LoWPAN carried as a standalone fragment header (i.e., with no 6LoWPAN
payload) in a message that is propagated back to the 6LoWPAN endpoint payload) in a message that is propagated back to the 6LoWPAN endpoint
that was the originator of the fragments. To achieve this, each hop that was the originator of the fragments. To achieve this, each hop
that performed an MPLS-like operation on fragments reverses that that performed an MPLS-like operation on fragments reverses that
operation for the RFRAG_ACK by sending a frame from the next hop to operation for the RFRAG_ACK by sending a frame from the next hop to
the previous hop as known by its MAC address in the VRB. The the previous hop as known by its MAC addressin the VRB. The
datagram_tag in the RFRAG_ACK is unique to the receiver and is enough datagram_tag in the RFRAG_ACK is unique to the receiver and is enough
information for an intermediate hop to locate the VRB that contains information for an intermediate hop to locate the VRB that contains
the datagram_tag used by the previous hop and the Layer-2 information the datagram_tag used by the previous hop and the Layer-2 information
associated to it (interface and MAC address). associated to it (interface and MAC address).
The 6LoWPAN endpoint that fragments the packets at 6LoWPAN level (the The 6LoWPAN endpoint that fragments the packets at the 6LoWPAN level
sender) also controls the amount of acknowledgments by setting the (the sender) also controls the amount of acknowledgments by setting
Ack-Request flag in the RFRAG packets. The sender may set the Ack- the Ack-Request flag in the RFRAG packets. The sender may set the
Request flag on any fragment to perform congestion control by Ack-Request flag on any fragment to perform congestion control by
limiting the number of outstanding fragments, which are the fragments limiting the number of outstanding fragments, which are the fragments
that have been sent but for which reception or loss was not that have been sent but for which reception or loss was not
positively confirmed by the reassembling endpoint. The maximum positively confirmed by the reassembling endpoint. The maximum
number of outstanding fragments is the Window-Size. It is number of outstanding fragments is the Window-Size. It is
configurable and may vary in case of ECN notification. When the configurable and may vary in case of ECN notification. When the
6LoWPAN endpoint that reassembles the packets at 6LoWPAN level (the 6LoWPAN endpoint that reassembles the packets at the 6LoWPAN level
receiver) receives a fragment with the Ack-Request flag set, it MUST (the receiver) receives a fragment with the Ack-Request flag set, it
send an RFRAG Acknowledgment back to the originator to confirm MUST send an RFRAG Acknowledgment back to the originator to confirm
reception of all the fragments it has received so far. reception of all the fragments it has received so far.
The Ack-Request ('X') set in an RFRAG marks the end of a window. The Ack-Request ('X') set in an RFRAG marks the end of a window.
This flag MUST be set on the last fragment if the sender wishes to This flag MUST be set on the last fragment if the sender wishes to
protect the datagram, and it MAY be set in any intermediate fragment protect the datagram, and it MAY be set in any intermediate fragment
for the purpose of flow control. This ARQ process MUST be protected for the purpose of flow control.
by a timer, and the fragment that carries the 'X' flag MAY be retried
upon time out a configurable amount of times (see Section 7.1). Upon This automatic repeat request (ARQ) process MUST be protected by a
exhaustion of the retries the sender may either abort the timer, and the fragment that carries the 'X' flag MAY be retried upon
a time out for a configurable number of times (see Section 7.1).
Upon exhaustion of the retries the sender may either abort the
transmission of the datagram or retry the datagram from the first transmission of the datagram or retry the datagram from the first
fragment with an 'X' flag set in order to reestablish a path and fragment with an 'X' flag set in order to reestablish a path and
discover which fragments were received over the old path in the discover which fragments were received over the old path in the
acknowledgment bitmap. When the sender of the fragment knows that an acknowledgment bitmap. When the sender of the fragment knows that an
underlying link-layer mechanism protects the fragments, it may underlying link-layer mechanism protects the fragments, it may
refrain from using the RFRAG Acknowledgment mechanism, and never set refrain from using the RFRAG Acknowledgment mechanism, and never set
the Ack-Request bit. the Ack-Request bit.
The receiver MAY issue unsolicited acknowledgments. An unsolicited
acknowledgment signals to the sender endpoint that it can resume
sending if it had reached its maximum number of outstanding
fragments. Another use is to inform the sender that the reassembling
endpoint aborted the processing of an individual datagram.
The RFRAG Acknowledgment can optionally carry an ECN indication for The RFRAG Acknowledgment can optionally carry an ECN indication for
flow control (see Appendix C). The receiver of a fragment with the flow control (see Appendix C). The receiver of a fragment with the
'E' (ECN) flag set MUST echo that information by setting the 'E' 'E' (ECN) flag set MUST echo that information by setting the 'E'
(ECN) flag in the next RFRAG Acknowledgment. (ECN) flag in the next RFRAG Acknowledgment.
In order to protect the datagram, the sender transfers a controlled In order to protect the datagram, the sender transfers a controlled
number of fragments and flags the last fragment of a window with an number of fragments and flags the last fragment of a window with an
RFRAG Acknowledgment Request. The receiver MUST acknowledge a RFRAG Acknowledgment Request. The receiver MUST acknowledge a
fragment with the acknowledgment request bit set. If any fragment fragment with the acknowledgment request bit set. If any fragment
immediately preceeding an acknowledgment request is still missing, immediately preceding an acknowledgment request is still missing, the
the receiver MAY intentionally delay its acknowledgment to allow in- receiver MAY intentionally delay its acknowledgment to allow in-
transit fragments to arrive. Because it might defeat the round trip transit fragments to arrive. Because it might defeat the round-trip
delay computation, delaying the acknowledgment should be configurable delay computation, delaying the acknowledgment should be configurable
and not enabled by default. and not enabled by default.
The receiver MAY issue unsolicited acknowledgments. An unsolicited
acknowledgment signals to the sender endpoint that it can resume
sending if it had reached its maximum number of outstanding
fragments. Another use is to inform that the reassembling endpoint
aborted the process of an individual datagram.
When all the fragments are received, the receiving endpoint When all the fragments are received, the receiving endpoint
reconstructs the packet, passes it to the upper layer, sends a RFRAG reconstructs the packet, passes it to the upper layer, sends an RFRAG
Acknowledgment on the reverse path with a FULL bitmap, and arms a Acknowledgment on the reverse path with a FULL bitmap, and arms a
short timer to absorb fragments that are still in flight for that short timer, e.g., in the order of an average round-trip delay in the
datagram without creating a new state and abort the communication if network. As the timer runs, the receiving endpoint absorbs the
it keeps going on beyond a reasonable time. fragments that were still in flight for that datagram without
creating a new state. The receiving endpoint abort the communication
if it keeps going on beyond the duration of the timer.
Note that acknowledgments might consume precious resources so the use Note that acknowledgments might consume precious resources so the use
of unsolicited acknowledgments should be configurable and not enabled of unsolicited acknowledgments should be configurable and not enabled
by default. by default.
An observation is that streamlining forwarding of fragments generally An observation is that streamlining forwarding of fragments generally
reduces the latency over the LLN mesh, providing room for retries reduces the latency over the LLN mesh, providing room for retries
within existing upper-layer reliability mechanisms. The sender within existing upper-layer reliability mechanisms. The sender
protects the transmission over the LLN mesh with a retry timer that protects the transmission over the LLN mesh with a retry timer that
is computed according to the method detailed in [RFC6298]. It is is computed according to the method detailed in [RFC6298]. It is
expected that the upper layer retries obey the recommendations in expected that the upper layer retries obey the recommendations in
"UDP Usage Guidelines" [RFC8085], in which case a single round of "UDP Usage Guidelines" [RFC8085], in which case a single round of
fragment recovery should fit within the upper layer recovery timers. fragment recovery should fit within the upper layer recovery timers.
Fragments are sent in a round robin fashion: the sender sends all the Fragments are sent in a round-robin fashion: the sender sends all the
fragments for a first time before it retries any lost fragment; lost fragments for a first time before it retries any lost fragment; lost
fragments are retried in sequence, oldest first. This mechanism fragments are retried in sequence, oldest first. This mechanism
enables the receiver to acknowledge fragments that were delayed in enables the receiver to acknowledge fragments that were delayed in
the network before they are retried. the network before they are retried.
When a single frequency is used by contiguous hops, the sender should When a single frequency is used by contiguous hops, the sender should
wait a reasonable amount of time between fragments so as to let a insert a delay between fragments of a same datagram that covers
fragment progress a few hops and avoid hidden terminal issues. This multiple transmissions so as to let a fragment progress a few hops
precaution is not required on channel hopping technologies such as and avoid hidden terminal issues. This precaution is not required on
Time Slotted Channel Hopping (TSCH) [RFC6554], where nodes that channel hopping technologies such as Time Slotted Channel Hopping
communicate at Layer-2 are scheduled to send and receive (TSCH) [RFC6554], where nodes that communicate at Layer-2 are
respectively, and different hops operate on different channels. scheduled to send and receive respectively, and different hops
operate on different channels.
6.1. Forwarding Fragments 6.1. Forwarding Fragments
It is assumed that the first Fragment is large enough to carry the It is assumed that the first fragment is large enough to carry the
IPv6 header and make routing decisions. If that is not so, then this IPv6 header and make routing decisions. If that is not so, then this
specification MUST NOT be used. specification MUST NOT be used.
This specification extends the Virtual Reassembly Buffer (VRB) This specification extends the Virtual Reassembly Buffer (VRB)
technique to forward fragments with no intermediate reconstruction of technique to forward fragments with no intermediate reconstruction of
the entire packet. It inherits operations like datagram_tag the entire packet. It inherits operations like datagram_tag
Switching and using a timer to clean the VRB when the traffic dries switching and using a timer to clean the VRB when the traffic dries
up. In more details, the first fragment carries the IP header and it up. The first fragment carries the IP header and it is routed all
is routed all the way from the fragmenting endpoint to the the way from the fragmenting endpoint to the reassembling endpoint.
reassembling endpoint. Upon the first fragment, the routers along Upon receiving the first fragment, the routers along the path install
the path install a label-switched path (LSP), and the following a label-switched path (LSP), and the following fragments are label-
fragments are label-switched along that path. As a consequence, the switched along that path. As a consequence, the next fragments can
next fragments can only follow the path that was set up by the first only follow the path that was set up by the first fragment and cannot
fragment and cannot follow an alternate route. The datagram_tag is follow an alternate route. The datagram_tag is used to carry the
used to carry the label, that is swapped at each hop. All fragments label, which is swapped in each hop. All fragments follow the same
follow the same path and fragments are delivered in the order at path and fragments are delivered in the order at which they are sent.
which they are sent.
6.1.1. Upon the first fragment 6.1.1. Receiving the first fragment
In Route-Over mode, the source and destination MAC addresses in a In Route-Over mode, the source and destination MAC addresses in a
frame change at each hop. The label that is formed and placed in the frame change at each hop. The label that is formed and placed in the
datagram_tag is associated to the source MAC and only valid (and datagram_tag is associated with the source MAC address and only valid
unique) for that source MAC. Upon a first fragment (i.e. with a (and unique) for that source MAC address. Upon a first fragment
sequence of zero), an intermediate router creates a VRB and the (i.e., with a sequence of zero), an intermediate router creates a VRB
associated LSP state for the tuple (source MAC address, datagram_tag) and the associated LSP state for the tuple (source MAC address,
and the fragment is forwarded along the IPv6 route that matches the datagram_tag) and the fragment is forwarded along the IPv6 route that
destination IPv6 address in the IPv6 header as prescribed by matches the destination IPv6 address in the IPv6 header as prescribed
[I-D.ietf-6lo-minimal-fragment], whereas the receiving endpoint by [I-D.ietf-6lo-minimal-fragment], where the receiving endpoint
allocates a reassembly buffer. allocates a reassembly buffer.
The LSP state enables to match the (previous MAC address, The LSP state enables to match the (previous MAC address,
datagram_tag) in an incoming fragment to the tuple (next MAC address, datagram_tag) in an incoming fragment to the tuple (next MAC address,
swapped datagram_tag) used in the forwarded fragment and points at swapped datagram_tag) used in the forwarded fragment and points at
the VRB. In addition, the router also forms a Reverse LSP state the VRB. In addition, the router also forms a reverse LSP state
indexed by the MAC address of the next hop and the swapped indexed by the MAC address address of the next hop and the swapped
datagram_tag. This reverse LSP state also points at the VRB and datagram_tag. This reverse LSP state also points at the VRB and
enables to match the (next MAC address, swapped_datagram_tag) found enables matching the (next MAC address, swapped_datagram_tag) found
in an RFRAG Acknowledgment to the tuple (previous MAC address, in an RFRAG Acknowledgment to the tuple (previous MAC address,
datagram_tag) used when forwarding a Fragment Acknowledgment (RFRAG- datagram_tag) used when forwarding a Fragment Acknowledgment (RFRAG-
ACK) back to the sender endpoint. ACK) back to the sender endpoint.
The first fragment may be received a second time, indicating that it The first fragment may be received a second time, indicating that it
did not reach the destination and was retried. In that case, it did not reach the destination and was retried. In that case, it
SHOULD follow the same path as the first occurrence. It is up to SHOULD follow the same path as the first occurrence. It is up to
sending endpoint to abort a transmission and then retry it from sending endpoint to determine whether to abort a transmission and
scratch, which may build an entirely new path. then retry it from scratch, which may build an entirely new path.
6.1.2. Upon the next fragments 6.1.2. Receiving the next fragments
Upon a next fragment (i.e. with a non-zero sequence), an intermediate Upon receiving a next fragment (i.e., with a non-zero sequence), an
router looks up a LSP indexed by the tuple (MAC address, intermediate router looks up a LSP indexed by the tuple (MAC address,
datagram_tag) found in the fragment. If it is found, the router datagram_tag) found in the fragment. If it is found, the router
forwards the fragment using the associated VRB as prescribed by forwards the fragment using the associated VRB as prescribed by
[I-D.ietf-6lo-minimal-fragment]. [I-D.ietf-6lo-minimal-fragment].
if the VRB for the tuple is not found, the router builds an RFRAG-ACK If the VRB for the tuple is not found, the router builds an RFRAG-ACK
to abort the transmission of the packet. The resulting message has to abort the transmission of the packet. The resulting message has
the following information: the following information:
* The source and destination MAC addresses are swapped from those * The source and destination MAC addresses are swapped from those
found in the fragment found in the fragment
* The datagram_tag set to the datagram_tag found in the fragment * The datagram_tag is set to the datagram_tag found in the fragment
* A NULL bitmap is used to signal the abort condition * A NULL bitmap is used to signal the abort condition
At this point the router is all set and can send the RFRAG-ACK back At this point the router is all set and can send the RFRAG-ACK back
to the previous router. The RFRAG-ACK should normally be forwarded to the previous router. The RFRAG-ACK should normally be forwarded
all the way to the source using the reverse LSP state in the VRBs in all the way to the source using the reverse LSP state in the VRBs in
the intermediate routers as described in the next section. the intermediate routers as described in the next section.
[I-D.ietf-6lo-minimal-fragment] indicates that the receiving endpoint [I-D.ietf-6lo-minimal-fragment] indicates that the receiving endpoint
stores "the actual packet data from the fragments received so far, in stores "the actual packet data from the fragments received so far, in
a form that makes it possible to detect when the whole packet has a form that makes it possible to detect when the whole packet has
been received and can be processed or forwarded". How this is been received and can be processed or forwarded". How this is
computed in implementation specific but relies on receiving all the computed in implementation specific but relies on receiving all the
bytes up to the datagram_size indicated in the first fragment. An bytes up to the datagram_size indicated in the first fragment. An
implementation may receive overlapping fragments as the result of implementation may receive overlapping fragments as the result of
retries after an MTU change. retries after an MTU change.
6.2. Upon the RFRAG Acknowledgments 6.2. Receiving RFRAG Acknowledgments
Upon an RFRAG-ACK, the router looks up a Reverse LSP indexed by the Upon receipt of an RFRAG-ACK, the router looks up a reverse LSP
tuple (MAC address, datagram_tag), which are respectively the source indexed by the tuple (MAC address, datagram_tag), which are
MAC address of the received frame and the received datagram_tag. If respectively the source MAC address of the received frame and the
it is found, the router forwards the fragment using the associated received datagram_tag. If it is found, the router forwards the
VRB as prescribed by [I-D.ietf-6lo-minimal-fragment], but using the fragment using the associated VRB as prescribed by
Reverse LSP so that the RFRAG-ACK flows back to the sender endpoint. [I-D.ietf-6lo-minimal-fragment], but using the reverse LSP so that
the RFRAG-ACK flows back to the sender endpoint.
If the Reverse LSP is not found, the router MUST silently drop the If the reverse LSP is not found, the router MUST silently drop the
RFRAG-ACK message. RFRAG-ACK message.
Either way, if the RFRAG-ACK indicates that the fragment was entirely Either way, if the RFRAG-ACK indicates that the fragment was entirely
received (FULL bitmap), it arms a short timer, and upon timeout, the received (FULL bitmap), it arms a short timer, and upon timeout, the
VRB and all the associated state are destroyed. Until the timer VRB and all the associated state are destroyed. Until the timer
elapses, fragments of that datagram may still be received, e.g. if elapses, fragments of that datagram may still be received, e.g. if
the RFRAG-ACK was lost on the way back and the source retried the the RFRAG-ACK was lost on the way back and the source retried the
last fragment. In that case, the router forwards the fragment last fragment. In that case, the router forwards the fragment
according to the state in the VRB. according to the state in the VRB.
skipping to change at page 16, line 41 skipping to change at page 16, line 50
of hops or the minimal value of MTU along those hops. But should the of hops or the minimal value of MTU along those hops. But should the
minimal MTU decrease, it is possible to retry a long fragment (say minimal MTU decrease, it is possible to retry a long fragment (say
sequence of 5) with first a shorter fragment of the same sequence (5 sequence of 5) with first a shorter fragment of the same sequence (5
again) and then one or more other fragments with a sequence that was again) and then one or more other fragments with a sequence that was
not used before (e.g., 13 and 14). Note that Path MTU Discovery is not used before (e.g., 13 and 14). Note that Path MTU Discovery is
out of scope for this document. out of scope for this document.
6.3. Aborting the Transmission of a Fragmented Packet 6.3. Aborting the Transmission of a Fragmented Packet
A reset is signaled on the forward path with a pseudo fragment that A reset is signaled on the forward path with a pseudo fragment that
has the fragment_offset, sequence and fragment_size all set to 0, and has the fragment_offset, sequence, and fragment_size all set to 0,
no data. and no data.
When the sender or a router on the way decides that a packet should When the sender or a router on the way decides that a packet should
be dropped and the fragmentation process aborted, it generates a be dropped and the fragmentation process aborted, it generates a
reset pseudo fragment and forwards it down the fragment path. reset pseudo fragment and forwards it down the fragment path.
Each router next along the path the way forwards the pseudo fragment Each router next along the path the way forwards the pseudo fragment
based on the VRB state. If an acknowledgment is not requested, the based on the VRB state. If an acknowledgment is not requested, the
VRB and all associated state are destroyed. VRB and all associated state are destroyed.
Upon reception of the pseudo fragment, the receiver cleans up all Upon reception of the pseudo fragment, the receiver cleans up all
resources for the packet associated to the datagram_tag. If an resources for the packet associated with the datagram_tag. If an
acknowledgment is requested, the receiver responds with a NULL acknowledgment is requested, the receiver responds with a NULL
bitmap. bitmap.
The other way around, the receiver might need to abort the process of The other way around, the receiver might need to abort the process of
a fragmented packet for internal reasons, for instance if it is out a fragmented packet for internal reasons, for instance if it is out
of reassembly buffers, already uses all 256 possible values of the of reassembly buffers, already uses all 256 possible values of the
datagram_tag, or if it keeps receiving fragments beyond a reasonable datagram_tag, or if it keeps receiving fragments beyond a reasonable
time while it considers that this packet is already fully reassembled time while it considers that this packet is already fully reassembled
and was passed to the upper layer. In that case, the receiver SHOULD and was passed to the upper layer. In that case, the receiver SHOULD
indicate so to the sender with a NULL bitmap in a RFRAG indicate so to the sender with a NULL bitmap in an RFRAG
Acknowledgment. The RFRAG Acknowledgment is frowarded all the way Acknowledgment. The RFRAG Acknowledgment is forwarded all the way
back to the source of the packet and cleans up all resources on the back to the source of the packet and cleans up all resources on the
way. Upon an acknowledgment with a NULL bitmap, the sender endpoint way. Upon an acknowledgment with a NULL bitmap, the sender endpoint
MUST abort the transmission of the fragmented datagram with one MUST abort the transmission of the fragmented datagram with one
exception: In the particular case of the first fragment, it MAY exception: In the particular case of the first fragment, it MAY
decide to retry via an alternate next hop instead. decide to retry via an alternate next hop instead.
6.4. Applying Recoverable Fragmentation along a Diverse Path 6.4. Applying Recoverable Fragmentation along a Diverse Path
The text above can be read with the assumption of a serial path The text above can be read with the assumption of a serial path
between a source and a destination. Section 4.5.3 of the "6TiSCH between a source and a destination. Section 4.5.3 of the "6TiSCH
skipping to change at page 18, line 24 skipping to change at page 18, line 30
MinWindowSize: The minimum value of Window_Size that the sender can MinWindowSize: The minimum value of Window_Size that the sender can
use. use.
OptWindowSize: The OptWindowSize is the value for the Window_Size OptWindowSize: The OptWindowSize is the value for the Window_Size
that the sender should use to start with. that the sender should use to start with.
MaxWindowSize: The maximum value of Window_Size that the sender can MaxWindowSize: The maximum value of Window_Size that the sender can
use. The value MUSt be less than 32. use. The value MUSt be less than 32.
InterFrameGap: Indicates a minimum amount of time between inter-frame gap: Indicates a minimum amount of time between
transmissions. All packets to a same destination, and in transmissions. All packets to a same destination, and in
particular fragments, may be subject to receive while transmitting particular fragments, may be subject to receive while transmitting
and hidden terminal collisions with the next or the previous and hidden terminal collisions with the next or the previous
transmission as the fragments progress along a same path. The transmission as the fragments progress along a same path. The
InterFrameGap protects the propagation of one transmission before inter-frame gap protects the propagation of one transmission
the next one is triggered and creates a duty cycle that controls before the next one is triggered and creates a duty cycle that
the ratio of air time and memory in intermediate nodes that a controls the ratio of air time and memory in intermediate nodes
particular datagram will use. that a particular datagram will use.
MinARQTimeOut: The maximum amount of time a node should wait for an MinARQTimeOut: The maximum amount of time a node should wait for an
RFRAG Acknowledgment before it takes a next action. RFRAG Acknowledgment before it takes a next action.
OptARQTimeOut: The starting point of the value of the amount that a OptARQTimeOut: The starting point of the value of the amount of time
sender should wait for an RFRAG Acknowledgment before it takes a that a sender should wait for an RFRAG Acknowledgment before it
next action. takes a next action.
MaxARQTimeOut: The maximum amount of time a node should wait for an MaxARQTimeOut: The maximum amount of time a node should wait for an
RFRAG Acknowledgment before it takes a next action. RFRAG Acknowledgment before it takes a next action.
MaxFragRetries: The maximum number of retries for a particular MaxFragRetries: The maximum number of retries for a particular
Fragment. fragment.
MaxDatagramRetries: The maximum number of retries from scratch for a MaxDatagramRetries: The maximum number of retries from scratch for a
particular Datagram. particular datagram.
7.2. Observing the network 7.2. Observing the network
The management system should monitor the amount of retries and of ECN The management system should monitor the amount of retries and of ECN
settings that can be observed from the perspective of the both the settings that can be observed from the perspective of both the sender
sender and the receiver, and may tune the optimum size of and the receiver, and may tune the optimum size of Fragment_Size and
Fragment_Size and of the Window_Size, OptWindowSize and OptWindowSize of the Window_Size, OptDatagramSize and OptWindowSize respectively,
respectively, at the sender. The values should be bounded by the at the sender. The values should be bounded by the expected number
expected number of hops and reduced beyond that when the number of of hops and reduced beyond that when the number of datagrams that can
datagrams that can traverse an intermediate point may exceed its traverse an intermediate point may exceed its capacity and cause a
capacity and cause a congestion loss. The InterFrameGap is another congestion loss. The inter-frame gap is another tool that can be
tool that can be used to increase the spacing between fragments of a used to increase the spacing between fragments of the same datagram
same datagram and reduce the ratio of time when a particular and reduce the ratio of time when a particular intermediate node
intermediate node holds a fragment of that datagram. holds a fragment of that datagram.
8. Security Considerations 8. Security Considerations
The considerations in the Security sections of [I-D.ietf-core-cocoa] The considerations in the Security sections of [I-D.ietf-core-cocoa]
and [I-D.ietf-6lo-minimal-fragment] apply equally to this and [I-D.ietf-6lo-minimal-fragment] apply equally to this
specification. specification.
The process of recovering fragments does not appear to create any The process of recovering fragments does not appear to create any
opening for new threat compared to "Transmission of IPv6 Packets over opening for new threat compared to "Transmission of IPv6 Packets over
IEEE 802.15.4 Networks" [RFC4944] beyond the change of size of the IEEE 802.15.4 Networks" [RFC4944] beyond the change of size of the
datagram_tag. By reducing to 8 bits, the tag will wrap faster than datagram_tag. By being reduced to 8 bits, the tag will wrap faster
with [RFC4944]. But for a constrained network where a node is than with [RFC4944]. But for a constrained network where a node is
expected to be able to hold only one or a few large packets in expected to be able to hold only one or a few large packets in
memory, 256 is still a large number. Also, the acknowledgement memory, 256 is still a large number. Also, the acknowledgement
mechanism allows ot clean up the state rapidly once the packet is mechanism allows cleaning up the state rapidly once the packet is
fully transmitted or aborted. fully transmitted or aborted.
The abstract Virtual Recovery Buffer inherited from The abstract Virtual Recovery Buffer inherited from
[I-D.ietf-6lo-minimal-fragment] may be used to perform a Denial-of- [I-D.ietf-6lo-minimal-fragment] may be used to perform a Denial-of-
Service (DoS) attack against the intermediate Routers since the Service (DoS) attack against the intermediate Routers since the
routers need to maintain a state per flow. The particular VRB routers need to maintain a state per flow. The particular VRB
implementation technique described in implementation technique described in
[I-D.ietf-lwig-6lowpan-virtual-reassembly] allows to realign which [I-D.ietf-lwig-6lowpan-virtual-reassembly] allows realigning which
data goes in which fragment which causes the intermediate node to data goes in which fragment, which causes the intermediate node to
store a portion of the data, which adds an attack vector that is not store a portion of the data, which adds an attack vector that is not
present with this specification. With this specification, the data present with this specification. With this specification, the data
that is transported in each fragment is conserved and the state to that is transported in each fragment is conserved and the state to
keep does not include any data that would not fit in the previous keep does not include any data that would not fit in the previous
fragment. fragment.
9. IANA Considerations 9. IANA Considerations
This document allocates 4 values in Page 0 for recoverable fragments This document allocates 2 patterns for a total of 4 dispatch values
from the "Dispatch Type Field" registry that was created by in Page 0 for recoverable fragments from the "Dispatch Type Field"
"Transmission of IPv6 Packets over IEEE 802.15.4 Networks" [RFC4944] registry that was created by "Transmission of IPv6 Packets over IEEE
and reformatted by "6LoWPAN Paging Dispatch" [RFC8025]. 802.15.4 Networks" [RFC4944] and reformatted by "6LoWPAN Paging
Dispatch" [RFC8025].
The suggested values (to be confirmed by IANA) are indicated in The suggested patterns (to be confirmed by IANA) are indicated in
Table 1. Table 1.
+-------------+------+----------------------------------+-----------+ +-------------+------+----------------------------------+-----------+
| Bit Pattern | Page | Header Type | Reference | | Bit Pattern | Page | Header Type | Reference |
+=============+======+==================================+===========+ +=============+======+==================================+===========+
| 11 10100x | 0 | RFRAG - Recoverable Fragment | THIS RFC | | 11 10100x | 0 | RFRAG - Recoverable Fragment | THIS RFC |
+-------------+------+----------------------------------+-----------+ +-------------+------+----------------------------------+-----------+
| 11 10100x | 1-14 | Unassigned | |
+-------------+------+----------------------------------+-----------+
| 11 10100x | 15 | Reserved for Experimental Use | RFC 8025 |
+-------------+------+----------------------------------+-----------+
| 11 10101x | 0 | RFRAG-ACK - RFRAG | THIS RFC | | 11 10101x | 0 | RFRAG-ACK - RFRAG | THIS RFC |
| | | Acknowledgment | | | | | Acknowledgment | |
+-------------+------+----------------------------------+-----------+ +-------------+------+----------------------------------+-----------+
| 11 10101x | 1-14 | Unassigned | |
+-------------+------+----------------------------------+-----------+
| 11 10101x | 15 | Reserved for Experimental Use | RFC 8025 |
+-------------+------+----------------------------------+-----------+
Table 1: Additional Dispatch Value Bit Patterns Table 1: Additional Dispatch Value Bit Patterns
10. Acknowledgments 10. Acknowledgments
The author wishes to thank Michel Veillette, Dario Tedeschi, Laurent The author wishes to thank Michel Veillette, Dario Tedeschi, Laurent
Toutain, Carles Gomez Montenegro, Thomas Watteyne and Michael Toutain, Carles Gomez Montenegro, Thomas Watteyne, and Michael
Richardson for in-depth reviews and comments. Also many thanks to Richardson for in-depth reviews and comments. Also many thanks to
Jonathan Hui, Jay Werb, Christos Polyzois, Soumitri Kolavennu, Pat Peter Yee and Erik Nordmark for their careful reviews and for helping
Kinney, Margaret Wasserman, Richard Kelsey, Carsten Bormann and Harry through the IESG review process, and to Jonathan Hui, Jay Werb,
Courtice for their various contributions. Christos Polyzois, Soumitri Kolavennu, Pat Kinney, Margaret
Wasserman, Richard Kelsey, Carsten Bormann, and Harry Courtice for
their various contributions.
11. Normative References 11. Normative References
[RFC2119] Bradner, S., "Key words for use in RFCs to Indicate [RFC2119] Bradner, S., "Key words for use in RFCs to Indicate
Requirement Levels", BCP 14, RFC 2119, Requirement Levels", BCP 14, RFC 2119,
DOI 10.17487/RFC2119, March 1997, DOI 10.17487/RFC2119, March 1997,
<https://www.rfc-editor.org/info/rfc2119>. <https://www.rfc-editor.org/info/rfc2119>.
[RFC4944] Montenegro, G., Kushalnagar, N., Hui, J., and D. Culler, [RFC4944] Montenegro, G., Kushalnagar, N., Hui, J., and D. Culler,
"Transmission of IPv6 Packets over IEEE 802.15.4 "Transmission of IPv6 Packets over IEEE 802.15.4
skipping to change at page 21, line 26 skipping to change at page 21, line 36
[RFC8138] Thubert, P., Ed., Bormann, C., Toutain, L., and R. Cragie, [RFC8138] Thubert, P., Ed., Bormann, C., Toutain, L., and R. Cragie,
"IPv6 over Low-Power Wireless Personal Area Network "IPv6 over Low-Power Wireless Personal Area Network
(6LoWPAN) Routing Header", RFC 8138, DOI 10.17487/RFC8138, (6LoWPAN) Routing Header", RFC 8138, DOI 10.17487/RFC8138,
April 2017, <https://www.rfc-editor.org/info/rfc8138>. April 2017, <https://www.rfc-editor.org/info/rfc8138>.
[RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC [RFC8174] Leiba, B., "Ambiguity of Uppercase vs Lowercase in RFC
2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174, 2119 Key Words", BCP 14, RFC 8174, DOI 10.17487/RFC8174,
May 2017, <https://www.rfc-editor.org/info/rfc8174>. May 2017, <https://www.rfc-editor.org/info/rfc8174>.
[I-D.ietf-6lo-minimal-fragment] [I-D.ietf-6lo-minimal-fragment]
Watteyne, T., Bormann, C., and P. Thubert, "6LoWPAN Watteyne, T., Thubert, P., and C. Bormann, "On Forwarding
Fragment Forwarding", Work in Progress, Internet-Draft, 6LoWPAN Fragments over a Multihop IPv6 Network", Work in
draft-ietf-6lo-minimal-fragment-04, 2 September 2019, Progress, Internet-Draft, draft-ietf-6lo-minimal-fragment-
<https://tools.ietf.org/html/draft-ietf-6lo-minimal- 10, 1 February 2020, <https://tools.ietf.org/html/draft-
fragment-04>. ietf-6lo-minimal-fragment-10>.
12. Informative References 12. Informative References
[RFC8201] McCann, J., Deering, S., Mogul, J., and R. Hinden, Ed., [RFC8201] McCann, J., Deering, S., Mogul, J., and R. Hinden, Ed.,
"Path MTU Discovery for IP version 6", STD 87, RFC 8201, "Path MTU Discovery for IP version 6", STD 87, RFC 8201,
DOI 10.17487/RFC8201, July 2017, DOI 10.17487/RFC8201, July 2017,
<https://www.rfc-editor.org/info/rfc8201>. <https://www.rfc-editor.org/info/rfc8201>.
[RFC7567] Baker, F., Ed. and G. Fairhurst, Ed., "IETF [RFC7567] Baker, F., Ed. and G. Fairhurst, Ed., "IETF
Recommendations Regarding Active Queue Management", Recommendations Regarding Active Queue Management",
skipping to change at page 25, line 12 skipping to change at page 25, line 21
caused the initial loss, and potentially leading to congestion caused the initial loss, and potentially leading to congestion
collapse. collapse.
This saturation may lead to excessive radio interference, or random This saturation may lead to excessive radio interference, or random
early discard (leaky bucket) in relaying nodes. Additional queuing early discard (leaky bucket) in relaying nodes. Additional queuing
and memory congestion may result while waiting for a low power next and memory congestion may result while waiting for a low power next
hop to emerge from its sleeping state. hop to emerge from its sleeping state.
Considering that RFC 4944 defines an MTU is 1280 bytes and that in Considering that RFC 4944 defines an MTU is 1280 bytes and that in
most incarnations (but 802.15.4g) a IEEE Std. 802.15.4 frame can most incarnations (but 802.15.4g) a IEEE Std. 802.15.4 frame can
limit the MAC payload to as few as 74 bytes, a packet might be limit the MAC address payload to as few as 74 bytes, a packet might
fragmented into at least 18 fragments at the 6LoWPAN shim layer. be fragmented into at least 18 fragments at the 6LoWPAN shim layer.
Taking into account the worst-case header overhead for 6LoWPAN Taking into account the worst-case header overhead for 6LoWPAN
Fragmentation and Mesh Addressing headers will increase the number of Fragmentation and Mesh Addressing headers will increase the number of
required fragments to around 32. This level of fragmentation is much required fragments to around 32. This level of fragmentation is much
higher than that traditionally experienced over the Internet with higher than that traditionally experienced over the Internet with
IPv4 fragments. At the same time, the use of radios increases the IPv4 fragments. At the same time, the use of radios increases the
probability of transmission loss and Mesh-Under techniques compound probability of transmission loss and Mesh-Under techniques compound
that risk over multiple hops. that risk over multiple hops.
Mechanisms such as TCP or application-layer segmentation could be Mechanisms such as TCP or application-layer segmentation could be
used to support end-to-end reliable transport. One option to support used to support end-to-end reliable transport. One option to support
skipping to change at page 25, line 38 skipping to change at page 25, line 47
that the end-to-end transport is aware of the delivery properties of that the end-to-end transport is aware of the delivery properties of
the underlying LLN, which is a layer violation, and difficult to the underlying LLN, which is a layer violation, and difficult to
achieve from the far end of the IPv6 network. achieve from the far end of the IPv6 network.
Appendix B. Requirements Appendix B. Requirements
For one-hop communications, a number of Low Power and Lossy Network For one-hop communications, a number of Low Power and Lossy Network
(LLN) link-layers propose a local acknowledgment mechanism that is (LLN) link-layers propose a local acknowledgment mechanism that is
enough to detect and recover the loss of fragments. In a multihop enough to detect and recover the loss of fragments. In a multihop
environment, an end-to-end fragment recovery mechanism might be a environment, an end-to-end fragment recovery mechanism might be a
good complement to a hop-by-hop MAC level recovery. This draft good complement to a hop-by-hop MAC address level recovery. This
introduces a simple protocol to recover individual fragments between draft introduces a simple protocol to recover individual fragments
6LoWPAN endpoints that may be multiple hops away. The method between 6LoWPAN endpoints that may be multiple hops away. The method
addresses the following requirements of a LLN: addresses the following requirements of an LLN:
Number of fragments The recovery mechanism must support highly Number of fragments: The recovery mechanism must support highly
fragmented packets, with a maximum of 32 fragments per packet. fragmented packets, with a maximum of 32 fragments per packet.
Minimum acknowledgment overhead Because the radio is half duplex, Minimum acknowledgment overhead: Because the radio is half duplex,
and because of silent time spent in the various medium access and because of silent time spent in the various medium access
mechanisms, an acknowledgment consumes roughly as many resources mechanisms, an acknowledgment consumes roughly as many resources
as data fragment. as a data fragment.
The new end-to-end fragment recovery mechanism should be able to The new end-to-end fragment recovery mechanism should be able to
acknowledge multiple fragments in a single message and not require acknowledge multiple fragments in a single message and not require
an acknowledgment at all if fragments are already protected at a an acknowledgment at all if fragments are already protected at a
lower layer. lower layer.
Controlled latency The recovery mechanism must succeed or give up Controlled latency: The recovery mechanism must succeed or give up
within the time boundary imposed by the recovery process of the within the time boundary imposed by the recovery process of the
Upper Layer Protocols. Upper Layer Protocols.
Optional congestion control The aggregation of multiple concurrent Optional congestion control: The aggregation of multiple concurrent
flows may lead to the saturation of the radio network and flows may lead to the saturation of the radio network and
congestion collapse. congestion collapse.
The recovery mechanism should provide means for controlling the The recovery mechanism should provide means for controlling the
number of fragments in transit over the LLN. number of fragments in transit over the LLN.
Appendix C. Considerations On Flow Control Appendix C. Considerations on Flow Control
Considering that a multi-hop LLN can be a very sensitive environment Considering that a multi-hop LLN can be a very sensitive environment
due to the limited queuing capabilities of a large population of its due to the limited queuing capabilities of a large population of its
nodes, this draft recommends a simple and conservative approach to nodes, this draft recommends a simple and conservative approach to
Congestion Control, based on TCP congestion avoidance. Congestion Control, based on TCP congestion avoidance.
Congestion on the forward path is assumed in case of packet loss, and Congestion on the forward path is assumed in case of packet loss, and
packet loss is assumed upon time out. The draft allows to control packet loss is assumed upon time out. The draft allows controlling
the number of outstanding fragments, that have been transmitted but the number of outstanding fragments that have been transmitted but
for which an acknowledgment was not received yet. It must be noted for which an acknowledgment was not received yet. It must be noted
that the number of outstanding fragments should not exceed the number that the number of outstanding fragments should not exceed the number
of hops in the network, but the way to figure the number of hops is of hops in the network, but the way to figure the number of hops is
out of scope for this document. out of scope for this document.
Congestion on the forward path can also be indicated by an Explicit Congestion on the forward path can also be indicated by an Explicit
Congestion Notification (ECN) mechanism. Though whether and how ECN Congestion Notification (ECN) mechanism. Though whether and how ECN
[RFC3168] is carried out over the LoWPAN is out of scope, this draft [RFC3168] is carried out over the LoWPAN is out of scope, this draft
provides a way for the destination endpoint to echo an ECN indication provides a way for the destination endpoint to echo an ECN indication
back to the source endpoint in an acknowledgment message as back to the source endpoint in an acknowledgment message as
represented in Figure 4 in Section 5.2. represented in Figure 4 in Section 5.2.
It must be noted that congestion and collision are different topics. It must be noted that congestion and collision are different topics.
In particular, when a mesh operates on a same channel over multiple In particular, when a mesh operates on a same channel over multiple
hops, then the forwarding of a fragment over a certain hop may hops, then the forwarding of a fragment over a certain hop may
collide with the forwarding of a next fragment that is following over collide with the forwarding of a next fragment that is following over
a previous hop but in a same interference domain. This draft enables a previous hop but in a same interference domain. This draft enables
an end-to-end flow control, but leaves it to the sender stack to pace end-to-end flow control, but leaves it to the sender stack to pace
individual fragments within a transmit window, so that a given individual fragments within a transmit window, so that a given
fragment is sent only when the previous fragment has had a chance to fragment is sent only when the previous fragment has had a chance to
progress beyond the interference domain of this hop. In the case of progress beyond the interference domain of this hop. In the case of
6TiSCH [I-D.ietf-6tisch-architecture], which operates over the 6TiSCH [I-D.ietf-6tisch-architecture], which operates over the
TimeSlotted Channel Hopping [RFC7554] (TSCH) mode of operation of TimeSlotted Channel Hopping [RFC7554] (TSCH) mode of operation of
IEEE802.14.5, a fragment is forwarded over a different channel at a IEEE802.14.5, a fragment is forwarded over a different channel at a
different time and it makes full sense to transmit the next fragment different time and it makes full sense to transmit the next fragment
as soon as the previous fragment has had its chance to be forwarded as soon as the previous fragment has had its chance to be forwarded
at the next hop. at the next hop.
skipping to change at page 27, line 20 skipping to change at page 27, line 32
fragment is a fragment that was sent but for which no explicit fragment is a fragment that was sent but for which no explicit
acknowledgment was received yet. This means that the fragment might acknowledgment was received yet. This means that the fragment might
be on the way, received but not yet acknowledged, or the be on the way, received but not yet acknowledged, or the
acknowledgment might be on the way back. It is also possible that acknowledgment might be on the way back. It is also possible that
either the fragment or the acknowledgment was lost on the way. either the fragment or the acknowledgment was lost on the way.
From the sender standpoint, all outstanding fragments might still be From the sender standpoint, all outstanding fragments might still be
in the network and contribute to its congestion. There is an in the network and contribute to its congestion. There is an
assumption, though, that after a certain amount of time, a frame is assumption, though, that after a certain amount of time, a frame is
either received or lost, so it is not causing congestion anymore. either received or lost, so it is not causing congestion anymore.
This amount of time can be estimated based on the round trip delay This amount of time can be estimated based on the round-trip delay
between the 6LoWPAN endpoints. The method detailed in [RFC6298] is between the 6LoWPAN endpoints. The method detailed in [RFC6298] is
recommended for that computation. recommended for that computation.
The reader is encouraged to read through "Congestion Control The reader is encouraged to read through "Congestion Control
Principles" [RFC2914]. Additionally [RFC7567] and [RFC5681] provide Principles" [RFC2914]. Additionally [RFC7567] and [RFC5681] provide
deeper information on why this mechanism is needed and how TCP deeper information on why this mechanism is needed and how TCP
handles Congestion Control. Basically, the goal here is to manage handles Congestion Control. Basically, the goal here is to manage
the amount of fragments present in the network; this is achieved by the amount of fragments present in the network; this is achieved by
to reducing the number of outstanding fragments over a congested path to reducing the number of outstanding fragments over a congested path
by throttling the sources. by throttling the sources.
Section 6 describes how the sender decides how many fragments are Section 6 describes how the sender decides how many fragments are
(re)sent before an acknowledgment is required, and how the sender (re)sent before an acknowledgment is required, and how the sender
adapts that number to the network conditions. adapts that number to the network conditions.
Author's Address Author's Address
Pascal Thubert (editor) Pascal Thubert (editor)
Cisco Systems, Inc Cisco Systems, Inc
Building D, 45 Allee des Ormes - BP1200 Building D
45 Allee des Ormes - BP1200
06254 MOUGINS - Sophia Antipolis 06254 MOUGINS - Sophia Antipolis
France France
Phone: +33 497 23 26 34 Phone: +33 497 23 26 34
Email: pthubert@cisco.com Email: pthubert@cisco.com
 End of changes. 97 change blocks. 
225 lines changed or deleted 242 lines changed or added

This html diff was produced by rfcdiff 1.48. The latest version is available from http://tools.ietf.org/tools/rfcdiff/