draft-ietf-manet-olsrv2-03.txt   draft-ietf-manet-olsrv2-04.txt 
Mobile Ad hoc Networking (MANET) T. Clausen Mobile Ad hoc Networking (MANET) T. Clausen
Internet-Draft LIX, Ecole Polytechnique, France Internet-Draft LIX, Ecole Polytechnique, France
Expires: August 5, 2007 C. Dearlove Intended status: Standards Track C. Dearlove
BAE Systems Advanced Technology Expires: January 10, 2008 BAE Systems Advanced Technology
Centre Centre
P. Jacquet P. Jacquet
Project Hipercom, INRIA Project Hipercom, INRIA
The OLSRv2 Design Team The OLSRv2 Design Team
MANET Working Group MANET Working Group
February 2007 July 9, 2007
The Optimized Link State Routing Protocol version 2 The Optimized Link State Routing Protocol version 2
draft-ietf-manet-olsrv2-03 draft-ietf-manet-olsrv2-04
Status of this Memo Status of this Memo
By submitting this Internet-Draft, each author represents that any By submitting this Internet-Draft, each author represents that any
applicable patent or other IPR claims of which he or she is aware applicable patent or other IPR claims of which he or she is aware
have been or will be disclosed, and any of which he or she becomes have been or will be disclosed, and any of which he or she becomes
aware will be disclosed, in accordance with Section 6 of BCP 79. aware will be disclosed, in accordance with Section 6 of BCP 79.
Internet-Drafts are working documents of the Internet Engineering Internet-Drafts are working documents of the Internet Engineering
Task Force (IETF), its areas, and its working groups. Note that Task Force (IETF), its areas, and its working groups. Note that
skipping to change at page 1, line 40 skipping to change at page 1, line 40
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."
The list of current Internet-Drafts can be accessed at The list of current Internet-Drafts can be accessed at
http://www.ietf.org/ietf/1id-abstracts.txt. http://www.ietf.org/ietf/1id-abstracts.txt.
The list of Internet-Draft Shadow Directories can be accessed at The list of Internet-Draft Shadow Directories can be accessed at
http://www.ietf.org/shadow.html. http://www.ietf.org/shadow.html.
This Internet-Draft will expire on August 5, 2007. This Internet-Draft will expire on January 10, 2008.
Copyright Notice Copyright Notice
Copyright (C) The IETF Trust (2007). Copyright (C) The IETF Trust (2007).
Abstract Abstract
This document describes version 2 of the Optimized Link State Routing This document describes version 2 of the Optimized Link State Routing
(OLSRv2) protocol for mobile ad hoc networks. The protocol embodies (OLSRv2) protocol for mobile ad hoc networks. The protocol embodies
an optimization of the classical link state algorithm tailored to the an optimization of the classical link state algorithm tailored to the
skipping to change at page 3, line 11 skipping to change at page 3, line 11
OLSRv2 is particularly suitable for large and dense networks as the OLSRv2 is particularly suitable for large and dense networks as the
technique of MPRs works well in this context. technique of MPRs works well in this context.
Table of Contents Table of Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 6 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 6
2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 7 2. Terminology . . . . . . . . . . . . . . . . . . . . . . . . . 7
3. Applicability Statement . . . . . . . . . . . . . . . . . . . 8 3. Applicability Statement . . . . . . . . . . . . . . . . . . . 8
4. Protocol Overview and Functioning . . . . . . . . . . . . . . 9 4. Protocol Overview and Functioning . . . . . . . . . . . . . . 9
5. Local Information Base . . . . . . . . . . . . . . . . . . . . 11 5. Protocol Parameters and Constants . . . . . . . . . . . . . . 11
5.1. Local Attached Network Set . . . . . . . . . . . . . . . . 11 5.1. Message Intervals . . . . . . . . . . . . . . . . . . . . 11
6. Processing and Forwarding Repositories . . . . . . . . . . . . 12 5.2. Advertised Information Validity Times . . . . . . . . . . 12
6.1. Received Set . . . . . . . . . . . . . . . . . . . . . . . 12 5.3. Received Message Validity Times . . . . . . . . . . . . . 12
6.2. Processed Set . . . . . . . . . . . . . . . . . . . . . . 12 5.4. Jitter . . . . . . . . . . . . . . . . . . . . . . . . . . 13
6.3. Forwarded Set . . . . . . . . . . . . . . . . . . . . . . 13 5.5. Hop Limit Parameter . . . . . . . . . . . . . . . . . . . 14
6.4. Relay Set . . . . . . . . . . . . . . . . . . . . . . . . 13 5.6. Willingness . . . . . . . . . . . . . . . . . . . . . . . 14
7. Packet Processing and Message Forwarding . . . . . . . . . . . 14 5.7. Parameter Change Constraints . . . . . . . . . . . . . . . 14
7.1. Actions when Receiving an OLSRv2 Packet . . . . . . . . . 14 6. Information Repositories . . . . . . . . . . . . . . . . . . . 17
7.2. Actions when Receiving an OLSRv2 Message . . . . . . . . . 14 6.1. Local Information Base . . . . . . . . . . . . . . . . . . 17
7.3. Message Considered for Processing . . . . . . . . . . . . 15 6.1.1. Local Attached Network Set . . . . . . . . . . . . . . 17
7.4. Message Considered for Forwarding . . . . . . . . . . . . 15 6.2. Neighborhood Information Base . . . . . . . . . . . . . . 18
8. Information Repositories . . . . . . . . . . . . . . . . . . . 18 6.3. Topology Information Base . . . . . . . . . . . . . . . . 18
8.1. Neighborhood Information Base . . . . . . . . . . . . . . 18 6.3.1. Advertised Neighbor Set . . . . . . . . . . . . . . . 19
8.1.1. Link Set . . . . . . . . . . . . . . . . . . . . . . . 18 6.3.2. Advertising Remote Node Set . . . . . . . . . . . . . 19
8.1.2. MPR Set . . . . . . . . . . . . . . . . . . . . . . . 18 6.3.3. Topology Set . . . . . . . . . . . . . . . . . . . . . 19
8.1.3. MPR Selector Set . . . . . . . . . . . . . . . . . . . 19 6.3.4. Attached Network Set . . . . . . . . . . . . . . . . . 20
8.2. Topology Information Base . . . . . . . . . . . . . . . . 19 6.3.5. Routing Set . . . . . . . . . . . . . . . . . . . . . 21
8.2.1. Advertised Neighbor Set . . . . . . . . . . . . . . . 19 6.4. Processing and Forwarding Information Base . . . . . . . . 21
8.2.2. ANSN History Set . . . . . . . . . . . . . . . . . . . 20 6.4.1. Received Set . . . . . . . . . . . . . . . . . . . . . 21
8.2.3. Topology Set . . . . . . . . . . . . . . . . . . . . . 20 6.4.2. Processed Set . . . . . . . . . . . . . . . . . . . . 22
8.2.4. Attached Network Set . . . . . . . . . . . . . . . . . 20 6.4.3. Forwarded Set . . . . . . . . . . . . . . . . . . . . 22
8.2.5. Routing Set . . . . . . . . . . . . . . . . . . . . . 21 6.4.4. Relay Set . . . . . . . . . . . . . . . . . . . . . . 23
9. Control Message Structures . . . . . . . . . . . . . . . . . . 22 7. Packet Processing and Message Forwarding . . . . . . . . . . . 24
9.1. HELLO Messages . . . . . . . . . . . . . . . . . . . . . . 22 7.1. Actions when Receiving an OLSRv2 Packet . . . . . . . . . 24
9.1.1. HELLO Message TLVs . . . . . . . . . . . . . . . . . . 23 7.2. Actions when Receiving an OLSRv2 Message . . . . . . . . . 24
9.1.2. HELLO Message Address Block TLVs . . . . . . . . . . . 23 7.3. Message Considered for Processing . . . . . . . . . . . . 25
9.2. TC Messages . . . . . . . . . . . . . . . . . . . . . . . 24 7.4. Message Considered for Forwarding . . . . . . . . . . . . 26
9.2.1. TC Message TLVs . . . . . . . . . . . . . . . . . . . 24 8. Packets and Messages . . . . . . . . . . . . . . . . . . . . . 29
9.2.2. TC Message Address Block TLVs . . . . . . . . . . . . 25 8.1. HELLO Messages . . . . . . . . . . . . . . . . . . . . . . 30
10. HELLO Message Generation . . . . . . . . . . . . . . . . . . . 26 8.1.1. HELLO Message TLVs . . . . . . . . . . . . . . . . . . 30
10.1. HELLO Message: Transmission . . . . . . . . . . . . . . . 26 8.1.2. HELLO Message Address Block TLVs . . . . . . . . . . . 31
11. HELLO Message Processing . . . . . . . . . . . . . . . . . . . 27 8.2. TC Messages . . . . . . . . . . . . . . . . . . . . . . . 31
11.1. Populating the MPR Selector Set . . . . . . . . . . . . . 27 8.2.1. TC Message TLVs . . . . . . . . . . . . . . . . . . . 32
11.2. Symmetric Neighborhood and 2-Hop Neighborhood Changes . . 28 8.2.2. TC Message Address Block TLVs . . . . . . . . . . . . 32
12. TC Message Generation . . . . . . . . . . . . . . . . . . . . 29 9. HELLO Message Generation . . . . . . . . . . . . . . . . . . . 33
12.1. TC Message: Transmission . . . . . . . . . . . . . . . . . 30 9.1. HELLO Message: Transmission . . . . . . . . . . . . . . . 33
13. TC Message Processing . . . . . . . . . . . . . . . . . . . . 32 10. HELLO Message Processing . . . . . . . . . . . . . . . . . . . 34
13.1. Initial TC Message Processing . . . . . . . . . . . . . . 32 10.1. Updating Willingness . . . . . . . . . . . . . . . . . . . 34
13.1.1. Populating the ANSN History Set . . . . . . . . . . . 32 10.2. Updating MPR Selectors . . . . . . . . . . . . . . . . . . 34
13.1.2. Populating the Topology Set . . . . . . . . . . . . . 33 10.3. Symmetric 1-Hop and 2-Hop Neighborhood Changes . . . . . . 34
13.1.3. Populating the Attached Network Set . . . . . . . . . 34 11. TC Message Generation . . . . . . . . . . . . . . . . . . . . 36
13.2. Completing TC Message Processing . . . . . . . . . . . . . 34 11.1. TC Message: Transmission . . . . . . . . . . . . . . . . . 37
13.2.1. Purging the Topology Set . . . . . . . . . . . . . . . 35 12. TC Message Processing . . . . . . . . . . . . . . . . . . . . 39
13.2.2. Purging the Attached Network Set . . . . . . . . . . . 35 12.1. Initial TC Message Processing . . . . . . . . . . . . . . 39
14. Populating the MPR Set . . . . . . . . . . . . . . . . . . . . 36 12.1.1. Populating the Advertising Remote Node Set . . . . . . 40
15. Populating Derived Sets . . . . . . . . . . . . . . . . . . . 37 12.1.2. Populating the Topology Set . . . . . . . . . . . . . 41
15.1. Populating the Relay Set . . . . . . . . . . . . . . . . . 37 12.1.3. Populating the Attached Network Set . . . . . . . . . 41
15.2. Populating the Advertised Neighbor Set . . . . . . . . . . 37 12.2. Completing TC Message Processing . . . . . . . . . . . . . 42
16. Routing Table Calculation . . . . . . . . . . . . . . . . . . 38 12.2.1. Purging the Topology Set . . . . . . . . . . . . . . . 42
17. Proposed Values for Constants . . . . . . . . . . . . . . . . 42 12.2.2. Purging the Attached Network Set . . . . . . . . . . . 42
17.1. Neighborhood Discovery Constants . . . . . . . . . . . . . 42 13. Selecting MPRs . . . . . . . . . . . . . . . . . . . . . . . . 43
17.2. Message Intervals . . . . . . . . . . . . . . . . . . . . 42 14. Populating Derived Sets . . . . . . . . . . . . . . . . . . . 45
17.3. Holding Times . . . . . . . . . . . . . . . . . . . . . . 42 14.1. Populating the Relay Set . . . . . . . . . . . . . . . . . 45
17.4. Jitter Times . . . . . . . . . . . . . . . . . . . . . . . 42 14.2. Populating the Advertised Neighbor Set . . . . . . . . . . 45
17.5. Willingness . . . . . . . . . . . . . . . . . . . . . . . 42 15. Routing Set Calculation . . . . . . . . . . . . . . . . . . . 46
18. Sequence Numbers . . . . . . . . . . . . . . . . . . . . . . . 43 15.1. Network Topology Graph . . . . . . . . . . . . . . . . . . 46
19. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 44 15.2. Populating the Routing Set . . . . . . . . . . . . . . . . 47
19.1. Message Types . . . . . . . . . . . . . . . . . . . . . . 44 15.3. Routing Set Updates . . . . . . . . . . . . . . . . . . . 48
19.2. TLV Types . . . . . . . . . . . . . . . . . . . . . . . . 44 16. Proposed Values for Parameters and Constants . . . . . . . . . 49
20. References . . . . . . . . . . . . . . . . . . . . . . . . . . 46 16.1. Message Interval Parameters . . . . . . . . . . . . . . . 49
20.1. Normative References . . . . . . . . . . . . . . . . . . . 46 16.2. Advertised Information Validity Time Parameters . . . . . 49
20.2. Informative References . . . . . . . . . . . . . . . . . . 46 16.3. Received Message Validity Time Parameters . . . . . . . . 49
Appendix A. Node Configuration . . . . . . . . . . . . . . . . . 47 16.4. Jitter Time Parameters . . . . . . . . . . . . . . . . . . 49
Appendix B. Protocol and Port Number . . . . . . . . . . . . . . 48 16.5. Hop Limit Parameter . . . . . . . . . . . . . . . . . . . 49
Appendix C. Example Heuristic for Calculating MPRs . . . . . . . 49 16.6. Willingness Parameter and Constants . . . . . . . . . . . 49
Appendix D. Packet and Message Layout . . . . . . . . . . . . . 52 17. Sequence Numbers . . . . . . . . . . . . . . . . . . . . . . . 51
Appendix D.1. Packet and Message Options . . . . . . . . . . . . . 52 18. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 52
Appendix D.2. Example HELLO Message . . . . . . . . . . . . . . . 54 18.1. Message Types . . . . . . . . . . . . . . . . . . . . . . 52
Appendix D.3. Example TC Message . . . . . . . . . . . . . . . . . 55 18.2. TLV Types . . . . . . . . . . . . . . . . . . . . . . . . 52
Appendix E. Time TLVs . . . . . . . . . . . . . . . . . . . . . 58 19. References . . . . . . . . . . . . . . . . . . . . . . . . . . 54
E.1. Representing Time . . . . . . . . . . . . . . . . . . . . 58 19.1. Normative References . . . . . . . . . . . . . . . . . . . 54
E.2. General Time TLV Structure . . . . . . . . . . . . . . . . 58 19.2. Informative References . . . . . . . . . . . . . . . . . . 54
E.3. Message TLVs . . . . . . . . . . . . . . . . . . . . . . . 60 Appendix A. Node Configuration . . . . . . . . . . . . . . . . . 56
E.3.1. VALIDITY_TIME TLV . . . . . . . . . . . . . . . . . . 60 Appendix B. Example Algorithm for Calculating MPRs . . . . . . . 57
E.3.2. INTERVAL_TIME TLV . . . . . . . . . . . . . . . . . . 60 B.1. Terminology . . . . . . . . . . . . . . . . . . . . . . . 57
Appendix F. Message Jitter . . . . . . . . . . . . . . . . . . . 61 B.2. MPR Selection Algorithm for each OLSRv2 Interface . . . . 58
F.1. Jitter . . . . . . . . . . . . . . . . . . . . . . . . . . 61 Appendix C. Example Algorithm for Calculating the Routing Set . 59
F.1.1. Periodic message generation . . . . . . . . . . . . . 61 C.1. Add Local Symmetric Links . . . . . . . . . . . . . . . . 59
F.1.2. Externally triggered message generation . . . . . . . 62 C.2. Add Remote Symmetric Links . . . . . . . . . . . . . . . . 60
F.1.3. Message forwarding . . . . . . . . . . . . . . . . . . 63 C.3. Add Attached Networks . . . . . . . . . . . . . . . . . . 61
F.1.4. Maximum Jitter Determination . . . . . . . . . . . . . 64 Appendix D. Packet and Message Layout . . . . . . . . . . . . . 62
Appendix G. Security Considerations . . . . . . . . . . . . . . 65 Appendix D.1. Packet and Message Options . . . . . . . . . . . . . 62
Appendix G.1. Confidentiality . . . . . . . . . . . . . . . . . . 65 Appendix D.2. Example HELLO Message . . . . . . . . . . . . . . . 64
Appendix G.2. Integrity . . . . . . . . . . . . . . . . . . . . . 65 Appendix D.3. Example TC Message . . . . . . . . . . . . . . . . . 65
Appendix G.3. Interaction with External Routing Domains . . . . . 66 Appendix E. Constraints . . . . . . . . . . . . . . . . . . . . 68
Appendix G.4. Node Identity . . . . . . . . . . . . . . . . . . . 67 Appendix F. Security Considerations . . . . . . . . . . . . . . 72
Appendix H. Flow and Congestion Control . . . . . . . . . . . . 68 Appendix F.1. Confidentiality . . . . . . . . . . . . . . . . . . 72
Appendix I. Contributors . . . . . . . . . . . . . . . . . . . . 69 Appendix F.2. Integrity . . . . . . . . . . . . . . . . . . . . . 72
Appendix J. Acknowledgements . . . . . . . . . . . . . . . . . . 70 Appendix F.3. Interaction with External Routing Domains . . . . . 73
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 71 Appendix F.4. Node Identity . . . . . . . . . . . . . . . . . . . 74
Intellectual Property and Copyright Statements . . . . . . . . . . 72 Appendix G. Flow and Congestion Control . . . . . . . . . . . . 75
Appendix H. Contributors . . . . . . . . . . . . . . . . . . . . 76
Appendix I. Acknowledgements . . . . . . . . . . . . . . . . . . 77
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . . 78
Intellectual Property and Copyright Statements . . . . . . . . . . 79
1. Introduction 1. Introduction
The Optimized Link State Routing protocol version 2 (OLSRv2) is an The Optimized Link State Routing protocol version 2 (OLSRv2) is an
update to OLSRv1 as published in RFC3626 [1]. Compared to RFC3626, update to OLSRv1 as published in RFC3626 [8]. Compared to RFC3626,
OLSRv2 retains the same basic mechanisms and algorithms, while OLSRv2 retains the same basic mechanisms and algorithms, while
providing a more flexible signaling framework and some simplification providing a more flexible signaling framework and some simplification
of the messages being exchanged. Also, OLSRv2 accommodates both IPv4 of the messages being exchanged. Also, OLSRv2 accommodates both IPv4
and IPv6 addresses in a compact manner. and IPv6 addresses in a compact manner.
OLSRv2 is developed for mobile ad hoc networks. It operates as a OLSRv2 is developed for mobile ad hoc networks. It operates as a
table driven, proactive protocol, i.e. it exchanges topology table driven, proactive protocol, i.e. it exchanges topology
information with other nodes in the network regularly. Each node information with other nodes in the network regularly. Each node
selects a set of its neighbor nodes as "MultiPoint Relays" (MPRs). selects a set of its neighbor nodes as "MultiPoint Relays" (MPRs).
Control traffic may be diffused through the network using hop by hop Control traffic may be diffused through the network using hop by hop
skipping to change at page 6, line 42 skipping to change at page 6, line 42
used as intermediate nodes in multi-hop route calculations. used as intermediate nodes in multi-hop route calculations.
A node selects MPRs from among its one hop neighbors connected by A node selects MPRs from among its one hop neighbors connected by
"symmetric", i.e. bi-directional, links. Therefore, selecting routes "symmetric", i.e. bi-directional, links. Therefore, selecting routes
through MPRs automatically avoids the problems associated with data through MPRs automatically avoids the problems associated with data
packet transfer over uni-directional links (such as the problem of packet transfer over uni-directional links (such as the problem of
not getting link layer acknowledgments at each hop, for link layers not getting link layer acknowledgments at each hop, for link layers
employing this technique). employing this technique).
OLSRv2 is developed to work independently from other protocols. OLSRv2 is developed to work independently from other protocols.
(Parts of OLSRv2 have been published separately as [3] and [4] for (Parts of OLSRv2 have been published separately as [1], [2], [3] and
wider use.) Likewise, OLSRv2 makes no assumptions about the [4] for wider use.) Likewise, OLSRv2 makes no assumptions about the
underlying link layer. However, OLSRv2 may use link layer underlying link layer. However, OLSRv2 may use link layer
information and notifications when available and applicable, as information and notifications when available and applicable, as
described in [4]. described in [4].
OLSRv2, as OLSRv1, inherits its concept of forwarding and relaying OLSRv2, as OLSRv1, inherits its concept of forwarding and relaying
from HIPERLAN (a MAC layer protocol) which is standardized by ETSI from HIPERLAN (a MAC layer protocol) which is standardized by ETSI
[6], [7]. [10], [11].
2. Terminology 2. Terminology
The keywords "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", The keywords "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
"SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
document are to be interpreted as described in RFC2119 [2]. document are to be interpreted as described in RFC2119 [5].
MANET specific terminology is to be interpreted as described in [3] MANET specific terminology is to be interpreted as described in [1]
and [4]. and [4].
Additionally, this document uses the following terminology: Additionally, this document uses the following terminology:
Node - A MANET router which implements the Optimized Link State Node - A MANET router which implements the Optimized Link State
Routing protocol version 2 as specified in this document. Routing protocol version 2 as specified in this document.
OLSRv2 interface - A MANET interface, running OLSRv2. OLSRv2 interface - A MANET interface, running OLSRv2.
Symmetric strict 2-hop neighbor - A symmetric 2-hop neighbor which Symmetric strict 2-hop neighbor - A symmetric 2-hop neighbor which
is not a symmetric 1-hop neighbor and is not a 2-hop neighbor only is not a symmetric 1-hop neighbor and is not a 2-hop neighbor only
through a symmetric 1-hop neighbor with willingness WILL_NEVER. through a symmetric 1-hop neighbor with willingness WILL_NEVER. A
node Z is a symmetric strict 2-hop neighbor of a node X if it is
not a symmetric 1-hop neighbor of node X and if there is a node Y
with willingness not equal to WILL_NEVER and such that there is a
symmetric link from node X to node Y, and a symmetric link from
node Y to node Z. A node Z is a symmetric strict 2-hop neighbor of
a node X by an OLSRv2 interface I of node X if in addition the
link from node X to node Y uses interface I.
Symmetric strict 2-hop neighborhood - The set of the symmetric Symmetric strict 2-hop neighborhood - The set of the symmetric
strict 2-hop neighbors of a node. strict 2-hop neighbors of a node.
Multipoint relay (MPR) - A node which is selected by its symmetric Multipoint relay (MPR) - A node which is selected by its symmetric
1-hop neighbor, node X, to "re-transmit" all the broadcast 1-hop neighbor, node X, to "re-transmit" all the broadcast
messages that it receives from node X, provided that the message messages that it receives from node X, provided that the message
is not a duplicate, and that the hop limit field of the message is is not a duplicate, and that the hop limit field of the message is
greater than one. greater than one.
MPR selector - A node which has selected its symmetric 1-hop MPR selector - A node which has selected its symmetric 1-hop
neighbor, node X, as one of its MPRs is an MPR selector of node X. neighbor, node X, as one of its MPRs is an MPR selector of node X.
3. Applicability Statement 3. Applicability Statement
OLSRv2 is a proactive routing protocol for mobile ad hoc networks OLSRv2 is a proactive routing protocol for mobile ad hoc networks
(MANETs). The larger and more dense a network, the more optimization (MANETs) [13]. The larger and more dense a network, the more
can be achieved by using MPRs compared to the classic link state optimization can be achieved by using MPRs compared to the classic
algorithm. OLSRv2 enables hop-by-hop routing, i.e. each node using link state algorithm. OLSRv2 enables hop-by-hop routing, i.e. each
its local information provided by OLSRv2 to route packets. node using its local information provided by OLSRv2 to route packets.
As OLSRv2 continuously maintains routes to all destinations in the As OLSRv2 continuously maintains routes to all destinations in the
network, the protocol is beneficial for traffic patterns where the network, the protocol is beneficial for traffic patterns where the
traffic is random and sporadic between a large subset of nodes, and traffic is random and sporadic between a large subset of nodes, and
where the [source, destination] pairs are changing over time: no where the [source, destination] pairs are changing over time. No
additional control traffic need be generated in this situation since additional control traffic need be generated in this case since
routes are maintained for all known destinations at all times. Also, routes are maintained for all known destinations at all times. Also,
since routes are maintained continuously, traffic is subject to no since routes are maintained continuously, traffic is subject to no
delays due to buffering or to route discovery. delays due to buffering or to route discovery.
OLSRv2 supports nodes which have multiple interfaces which OLSRv2 supports nodes which have multiple interfaces which
participate in the MANET using OLSRv2. As described in [4], each participate in the MANET using OLSRv2. As described in [4], each
OLSRv2 interface may have one or more network addresses (which may OLSRv2 interface may have one or more network addresses (which may
have prefix lengths). OLSRv2, additionally, supports nodes which have prefix lengths). OLSRv2, additionally, supports nodes which
have non-OLSRv2 interfaces which can serve as gateways towards other have non-OLSRv2 interfaces which may be local or can serve as
networks. gateways towards other networks.
OLSRv2 uses the format specified in [3] for all messages and packets. OLSRv2 uses the format specified in [1] for all messages and packets.
OLSRv2 is thereby able to allow for extensions via "external" and OLSRv2 is thereby able to allow for extensions via "external" and
"internal" extensibility. External extensibility allows a protocol "internal" extensibility. External extensibility allows a protocol
extension to specify and exchange new message types, which can be extension to specify and exchange new message types, which can be
forwarded and delivered correctly even by nodes which do not support forwarded and delivered correctly even by nodes which do not support
that extension. Internal extensibility allows a protocol extension that extension. Internal extensibility allows a protocol extension
to define additional attributes to be carried embedded in the to define additional attributes to be carried embedded in the
standard OLSRv2 control messages detailed in this specification (or standard OLSRv2 control messages detailed in this specification (or
any new message types defined by other protocol extensions) using the any new message types defined by other protocol extensions) using the
TLV mechanism specified in [3], while still allowing nodes not TLV mechanism specified in [1], while still allowing nodes not
supporting that extension to forward messages including the extension supporting that extension to forward messages including the extension
and process messages ignoring the extension. and to process messages ignoring the extension.
The OLSRv2 neighborhood discovery protocol using HELLO messages is The OLSRv2 neighborhood discovery protocol using HELLO messages is
specified in [4]; note that all references to MANET interfaces in [4] specified in [4]; note that all references to MANET interfaces in [4]
refer to OLSRv2 interfaces when using [4] as part of OLSRv2. This refer to OLSRv2 interfaces when using [4] as part of OLSRv2. This
neighborhood discovery protocol serves to ensure that each OLSRv2 neighborhood discovery protocol serves to ensure that each OLSRv2
node has available continuously updated information repositories node has available continuously updated information repositories
describing the node's 1-hop and symmetric 2-hop neighbors. This describing the node's 1-hop and symmetric 2-hop neighbors. This
neighborhood discovery protocol, which also uses [3], is extended in neighborhood discovery protocol, which also uses [1], is extended in
this document by the addition of MPR information. this document by the addition of MPR information.
4. Protocol Overview and Functioning 4. Protocol Overview and Functioning
OLSRv2 is a proactive routing protocol for mobile ad hoc networks. OLSRv2 is a proactive routing protocol for mobile ad hoc networks.
The protocol inherits the stability of a link state algorithm and has The protocol inherits the stability of a link state algorithm and has
the advantage of having routes immediately available when needed due the advantage of having routes immediately available when needed due
to its proactive nature. OLSRv2 is an optimization of the classical to its proactive nature. OLSRv2 is an optimization of the classical
link state protocol, tailored for mobile ad hoc networks. The main link state protocol, tailored for mobile ad hoc networks. The main
tailoring and optimizations of OLSRv2 are: tailoring and optimizations of OLSRv2 are:
skipping to change at page 9, line 27 skipping to change at page 9, line 27
o partial topology maintenance - each node knows only a subset of o partial topology maintenance - each node knows only a subset of
the links in the network, sufficient for a minimum hop route to the links in the network, sufficient for a minimum hop route to
all destinations. all destinations.
The optimized flooding and partial topology maintenance are based on The optimized flooding and partial topology maintenance are based on
the concept on MultiPoint Relays (MPRs), selected independently by the concept on MultiPoint Relays (MPRs), selected independently by
nodes based on the symmetric 1-hop and 2-hop neighbor information nodes based on the symmetric 1-hop and 2-hop neighbor information
maintained using [4]. maintained using [4].
Using the message exchange format [3] and the neighborhood discovery Using the message exchange format [1] and the neighborhood discovery
protocol [4], OLSRv2 also contains the following main components: protocol [4], OLSRv2 also contains the following main components:
o A TLV, to be included within the HELLO messages of [4], allowing a o A TLV, to be included within the HELLO messages of [4], allowing a
node to signal MPR selection. node to signal MPR selection.
o An optimized flooding mechanism for global information exchange, o An optimized flooding mechanism for global information exchange,
denoted "MPR flooding". denoted "MPR flooding".
o A specification of global signaling, denoted TC (Topology Control) o A specification of global signaling, denoted TC (Topology Control)
messages. TC messages in OLSRv2 serve to: messages. TC messages in OLSRv2 serve to:
skipping to change at page 9, line 51 skipping to change at page 9, line 51
* inject addresses of hosts and networks for which they may serve * inject addresses of hosts and networks for which they may serve
as a gateway into the entire network. as a gateway into the entire network.
TC messages are emitted periodically, thereby allowing nodes to TC messages are emitted periodically, thereby allowing nodes to
continuously track global changes in the network. Incomplete TC continuously track global changes in the network. Incomplete TC
messages may be used to report additions to advertised information messages may be used to report additions to advertised information
without repeating unchanged information. Some TC messages may be without repeating unchanged information. Some TC messages may be
flooded over only part of the network, allowing a node to ensure flooded over only part of the network, allowing a node to ensure
that nearer nodes are kept more up to date than distant nodes. that nearer nodes are kept more up to date than distant nodes.
Each node in the network selects an MPR Set. The MPR Set of a node X Each node in the network selects a set of MPRs. The MPRs of a node X
may be any subset of its symmetric 1-hop neighborhood such that every may be any subset of the willing nodes in node X's symmetric 1-hop
node in the symmetric strict 2-hop neighborhood of node X has a neighborhood such that every node in the symmetric strict 2-hop
symmetric link to a node in the MPR Set of node X. The MPR Set of a neighborhood of node X has a symmetric link to at least one of node
node may thus be said to "cover" the node's symmetric strict 2-hop X's MPRs. The MPRs of a node may thus be said to "cover" the node's
neighborhood. Each node also maintains information about the set of symmetric strict 2-hop neighborhood. Each node also maintains
symmetric 1-hop neighbors that have selected it as MPR. This set is information about the set of symmetric 1-hop neighbors that have
called the MPR Selector Set of the node. selected it as MPR, its MPR selectors.
Note that as long as the condition above is satisfied, any algorithm Note that as long as the condition above is satisfied, any algorithm
selecting MPR Sets is acceptable in terms of implementation selecting MPRs is acceptable in terms of implementation
interoperability. However if smaller MPR Sets are selected then the interoperability. However if smaller sets of MPRs are selected then
greater the efficiency gains that are possible. Note that [8] gives the greater the efficiency gains that are possible. Note that [12]
an analysis and example of MPR selection algorithms. gives an analysis and example of MPR selection algorithms.
In OLSRv2, actual efficiency gains are based on the sizes of each In OLSRv2, actual efficiency gains are based on the sizes of each
node's Relay Set, the set of symmetric 1-hop neighbors for which it node's Relay Set, the set of symmetric 1-hop neighbors for which it
is to relay broadcast traffic, and its Advertised Neighbor Set, the is to relay broadcast traffic, and its Advertised Neighbor Set, the
set of symmetric 1-hop neighbors for which it is to advertise link set of symmetric 1-hop neighbors for which it is to advertise link
state information into the network in TC messages. Each of these state information into the network in TC messages. Each of these
sets MUST contain all the nodes in the MPR Selector Set and MAY sets MUST contain all MPR selectors, and MAY contain additional
contain additional nodes. If the Advertised Neighbor Set is empty, nodes. If the Advertised Neighbor Set is empty, TC messages are not
TC messages are not generated by that node, unless needed for gateway generated by that node, unless needed for gateway reporting, or for a
reporting, or for a short period to accelerate the removal of short period to accelerate the removal of unwanted links.
unwanted links.
OLSRv2 is designed to work in a completely distributed manner and OLSRv2 is designed to work in a completely distributed manner and
does not depend on any central entity. The protocol does not require does not depend on any central entity. The protocol does not require
reliable transmission of control messages: each node sends control reliable transmission of control messages: each node sends control
messages periodically, and can therefore sustain a reasonable loss of messages periodically, and can therefore sustain a reasonable loss of
some such messages. Such losses may occur frequently in radio some such messages. Such losses may occur frequently in radio
networks due to collisions or other transmission problems. OLSRv2 networks due to collisions or other transmission problems. OLSRv2
may use "jitter", randomized adjustments to message transmission may use "jitter", randomized adjustments to message transmission
times, to reduce the incidence of collisions. times, to reduce the incidence of collisions [3].
OLSRv2 does not require sequenced delivery of messages. Each control OLSRv2 does not require sequenced delivery of messages. Each control
message contains a sequence number which is incremented for each message contains a sequence number which is incremented for each
message. Thus the recipient of a control message can, if required, message. Thus the recipient of a control message can, if required,
easily identify which information is more recent - even if messages easily identify which information is more recent - even if messages
have been re-ordered while in transmission. have been re-ordered while in transmission.
OLSRv2 does not require any changes to the format of IP packets, any OLSRv2 does not require any changes to the format of IP packets, any
existing IP stack can be used as is: OLSRv2 only interacts with existing IP stack can be used as is: OLSRv2 only interacts with
routing table management. OLSR sends its control messages using UDP. routing table management. OLSR sends its control messages using UDP.
5. Local Information Base 5. Protocol Parameters and Constants
A node maintains a Local Information Base that records information The parameters and constants used in this specification are those
about its OLSRv2 interfaces, and its non-OLSRv2 interfaces that can defined in [4] plus those defined in this section. The separation in
serve as gateways to other networks. The former is maintained using [4] into interface parameters, node parameters and constants is also
a Local Interface Set, as described in [4]. The latter is maintained used in OLSRv2, however all but one (RX_HOLD_TIME) of the parameters
using a Local Attached Network Set. All addresses in the Local added in this section are node parameters. They may be classified
Information Base have an associated prefix length; if an address into the following categories:
otherwise does not have a prefix length then it is set equal to the
address length. Two addresses are considered equal if and only if
their associated prefix lengths are also equal.
The Local Information Base is not modified by this protocol. This o Message intervals
protocol may respond to changes of this Local Information Base which
MUST reflect corresponding changes in the node's status. It is not
the responsibility of OLSRv2 to maintain routes to networks recorded
in the Local Attached Network Set in that node.
5.1. Local Attached Network Set o Advertised information validity times
o Received message validity times
o Jitter times
o Hop limits
o Willingness
In addition constants for particular cases of a node's willingness to
be an MPR are defined. These parameters and constants are detailed
in the following sections. As for the parameters in [4], parameters
defined in this document may be changed dynamically by a node, and
need not be the same on different nodes.
5.1. Message Intervals
The following interface parameters regulate TC message transmissions
by a node. TC messages are usually sent periodically, but MAY also
be sent in response to changes in the node's Advertised Neighbor Set
and Local Attached Network Set. With a larger value of parameter
TC_INTERVAL, and a smaller value of parameter TC_MIN_INTERVAL, TC
messages may often be transmitted in response to changes in a highly
dynamic network. However because a node has no knowledge of, for
example, nodes remote to it joining the network, TC messages MUST NOT
be sent purely responsively.
TC_INTERVAL - is the maximum time between the transmission of two
successive TC messages by this node. When no TC messages are sent
in response to local network changes (by design, or because the
local network is not changing) then TC messages SHOULD be sent at
a regular interval TC_INTERVAL, possibly modified by jitter as
specified in [3].
TC_MIN_INTERVAL - is the minimum interval between transmission of
two successive TC messages by this node. (This minimum interval
MAY be modified by jitter, as defined in [3].)
The following constraints apply to these parameters:
o TC_INTERVAL > 0
o TC_MIN_INTERVAL >= 0
o TC_INTERVAL >= TC_MIN_INTERVAL
o If INTERVAL_TIME TLVs as defined in [2] are included in TC
messages, then TC_INTERVAL MUST be representable as described in
[2].
5.2. Advertised Information Validity Times
The following parameters manage the validity time of information
advertised in TC messages:
T_HOLD_TIME - is used to define the minimum value in the
VALIDITY_TIME TLV included in all TC messages sent by this node.
If a single value of parameter TC_HOP_LIMIT is used then this will
be the only value in that TLV.
A_HOLD_TIME - is the period during which TC messages are sent after
they no longer have any advertised information to report, but are
sent in order to accelerate outdated information removal by other
nodes.
The following constraints apply to these parameters:
o T_HOLD_TIME > 0
o A_HOLD_TIME >= 0
o If TC messages can be lost then both SHOULD be significantly
greater than TC_INTERVAL.
o T_HOLD_TIME MUST be representable as described in [2].
5.3. Received Message Validity Times
The following parameters manage the validity time of recorded
received message information:
RX_HOLD_TIME - is an interface parameter, and is the period after
receipt of a message by the appropriate OLSRv2 interface of this
node for which that information is recorded in order that the
message is recognized as having been previously received on this
OLSRv2 interface.
P_HOLD_TIME - is the period after receipt of a message which is
processed by this node for which that information is recorded in
order that the message is not processed again if received again.
F_HOLD_TIME - is the period after receipt of a message which is
forwarded by this node for which that information is recorded in
order that the message is not forwarded again if received again.
The following constraints apply to these parameters:
o RX_HOLD_TIME > 0
o P_HOLD_TIME > 0
o F_HOLD_TIME > 0
o All of these parameters SHOULD be greater than the maximum
variation in time that a message may take to traverse the MANET,
taking into account any message forwarding jitter as well as
propagation, queuing, and processing delays.
5.4. Jitter
If jitter, as defined in [3], is used then these parameters are as
follows:
TP_MAXJITTER - represents the value of MAXJITTER used in [3] for
periodically generated TC messages sent by this node.
TT_MAXJITTER - represents the value of MAXJITTER used in [3] for
externally triggered TC messages sent by this node.
F_MAXJITTER - represents the default value of MAXJITTER used in [3]
for messages forwarded by this node. However before using
F_MAXJITTER a node MAY attempt to deduce a more appropriate value
of MAXJITTER, for example based on any INTERVAL_TIME or
VALIDITY_TIME TLVs contained in the message to be forwarded.
For constraints on these parameters see [3].
5.5. Hop Limit Parameter
The parameter TC_HOP_LIMIT is the hop limit set in each TC message.
TC_HOP_LIMIT MAY be a single fixed value, or MAY be different in TC
messages sent by the same node. However each other node SHOULD see a
regular pattern of TC messages, in order that meaningful values of
INTERVAL_TIME and VALIDITY_TIME TLVs at each hop count distance can
be included as defined in [2]. Thus the pattern of TC_HOP_LIMIT
SHOULD be defined to have this property. For example the repeating
pattern (255 4 4) satisfies this property (having period TC_INTERVAL
at hop counts up to 4, inclusive, and 3 x TC_INTERVAL at hop counts
greater than 4), but the repeating pattern (255 255 4 4) does not
satisfy this property.
The maximum value of TC_HOP_LIMIT used MUST least equal the network
diameter in hops, a value of 255 is RECOMMENDED. All values of
TC_HOP_LIMIT MUST satisfy TC_HOP_LIMIT >= 2.
5.6. Willingness
Each node has a WILLINGNESS parameter, which MUST be in the range
WILL_NEVER to WILL_ALWAYS, inclusive, and represents its willingness
to be an MPR, and hence its willingness to forward messages and be an
intermediate node on routes. If a node has WILLINGNESS == WILL_NEVER
it does not perform these tasks. A MANET using OLSRv2 with too many
nodes with WILLINGNESS == WILL_NEVER will not function; it MUST be
ensured, by administrative or other means, that this does not happen.
Nodes MAY have different WILLINGNESS values; however the three
constants WILL_NEVER, WILL_DEFAULT and WILL_ALWAYS MUST have the
values defined in Section 16.6. (Use of WILLINGNESS == WILL_DEFAULT
allows a node to avoid including a WILLINGNESS TLV in its TC
messages, use of WILLINGNESS == WILL_ALWAYS means that a node will
always be selected as an MPR by all symmetric 1-hop neighbors.)
5.7. Parameter Change Constraints
This section presents guidelines, applicable if protocol parameters
are changed dynamically.
TC_INTERVAL
* If the TC_INTERVAL for a node increases, then the next TC
message generated by this node MUST be generated according to
the previous, shorter, TC_INTERVAL. Additional subsequent TC
messages MAY be generated according to the previous, shorter,
TC_INTERVAL.
* If the TC_INTERVAL for a node decreases, then the following TC
messages from this node SHOULD be generated according to the
current, shorter, TC_INTERVAL.
T_HOLD_TIME
* If T_HOLD_TIME changes, then T_time for all Topology Tuples,
AN_time for all Attached Network Tuples and AR_time for all
Advertising Remote Node Tuples SHOULD be changed.
RX_HOLD_TIME
* If RX_HOLD_TIME for an OLSRv2 interface changes, then RX_time
for all Received Tuples for that OLSRv2 interface MAY be
changed.
P_HOLD_TIME
* If P_HOLD_TIME changes, then P_time for all Processed Tuples
MAY be changed.
F_HOLD_TIME
* If F_HOLD_TIME changes, then F_time for all Forwarded Tuples
MAY be changed.
TP_MAXJITTER
* If TP_MAXJITTER changes, then the periodic TC message schedule
on this node MAY be changed immediately.
TT_MAXJITTER
* If TT_MAXJITTER changes, then externally triggered TC messages
on this node MAY be rescheduled.
F_MAXJITTER
* If F_MAXJITTER changes, then TC messages waiting to be
forwarded with a delay based on this parameter MAY be
rescheduled.
TC_HOP_LIMIT
* If TC_HOP_LIMIT changes, and the node uses multiple values
after the change, then message intervals and validity times
included in TC messages MUST be respected. The simplest way to
do this is to start any new repeating pattern of TC_HOP_LIMIT
values with its largest value.
6. Information Repositories
The purpose of OLSRv2 is to determine the Routing Set, which may be
used to update IP's Routing Table, providing "next hop" routing
information for IP datagrams. OLSRv2 maintains four information
repositories:
Local Information Base - as defined in [4], extended by the addition
of a Local Attached Network Set, defined in Section 6.1.1.
Neighborhood Information Base - as defined in [4], extended by the
addition of 3 elements to each Neighbor Tuple, as defined in
Section 6.2.
Topology Information Base - this information base is specific to
OLSRv2, defined in Section 6.3.
Processing and Forwarding Information Base - this information base
is specific to OLSRv2, defined in Section 6.4.
All addresses, other than originator addresses, recorded in the
information repositories MUST all be recorded with prefix lengths, in
order to allow comparison with addresses received in HELLO and TC
messages.
The ordering of sequence numbers, when considering which is the
greatest, is as defined in Section 17.
6.1. Local Information Base
The Local Information Base as defined in [4] is extended by the
addition of a Local Attached Network Set, defined in Section 6.1.1.
6.1.1. Local Attached Network Set
A node's Local Attached Network Set records its local non-OLSRv2 A node's Local Attached Network Set records its local non-OLSRv2
interfaces. that can act as gateways to other networks. It consists interfaces that can act as gateways to other networks. The Local
of Local Attached Network Tuples: Attached Network Set is not modified by this protocol. This protocol
MAY respond to changes to the Local Attached Network Set, which MUST
reflect corresponding changes in the node's status. It consists of
Local Attached Network Tuples:
(AL_net_addr, AL_dist) (AL_net_addr, AL_dist)
where: where:
AL_net_addr is the network address of an attached network which can AL_net_addr is the network address of an attached network which can
be reached via this node. be reached via this node.
AL_dist is the number of hops to the network with address AL_dist is the number of hops to the network with address
AL_net_addr from this node. AL_net_addr from this node.
Attached networks with AL_dist == 0 MUST be local to this node and Attached networks with AL_dist == 0 MUST be local to this node and
MUST NOT be attached to any other node. Attached networks with MUST NOT be attached to any other node. Attached networks with
AL_dist > 0 MAY be attached to other nodes. AL_dist > 0 MAY also be attached to other nodes.
Attached networks with AL_dist > 0 MUST be advertised in TC messages Attached networks with AL_dist > 0 MUST be advertised in TC messages
generated by this node, this may result in the node originating TC generated by this node, this may result in the node originating TC
messages when it has no other reason to do so. Attached networks messages when it has no other reason to do so. Attached networks
with AL_dist == 0 MAY be advertised in HELLO messages (which causes with AL_dist == 0 MAY be advertised in HELLO messages (which causes
the MPRs of this node to advertise them in their TC messages) or MAY the MPRs of this node to advertise them in their TC messages) or MAY
be advertised in TC messages; they MUST be advertised in one type of be advertised in TC messages; they MUST be advertised in one type of
message and SHOULD NOT be advertised in both. If a node is sending message and SHOULD NOT be advertised in both. If a node is sending
TC messages for any other reason, then advertising attached networks TC messages for any other reason, then advertising attached networks
in TC messages is more efficient. A node MAY decide which form of in TC messages is more efficient. A node MAY decide which form of
advertisement to use depending on its circumstances. advertisement to use depending on its circumstances.
6. Processing and Forwarding Repositories It is not the responsibility of OLSRv2 to maintain routes to networks
recorded in the Local Attached Network Set.
The following data structures are employed in order to ensure that a 6.2. Neighborhood Information Base
message is processed at most once and is forwarded at most once per
interface of a node.
6.1. Received Set Each Neighbor Tuple in the Neighbor Set has these additional
elements:
A node's Received Sets, one per OLSRv2 interface, each record the N_willingness is the node's willingness to be selected as an MPR, in
signatures of messages which have been received over that interface. the range from WILL_NEVER to WILL_ALWAYS, both inclusive;
Each consists of Received Tuples:
N_mpr is a boolean flag, describing if the neighbor is selected as
an MPR by this node;
N_mpr_selector is a boolean flag, describing if this neighbor has
selected this node as an MPR, i.e. is an MPR selector of this
node.
6.3. Topology Information Base
The Topology Information Base stores information required for the
generation and processing of TC messages, and received in TC
messages. The Advertised Neighbor Set contains interface addresses
of symmetric 1-hop neighbors which are to be reported in TC messages.
The Topology Set and Attached Network Set both record information
received in TC messages. The Advertising Remote Node Set is both
used and populated when processing TC messages.
Additionally, a Routing Set is maintained, derived from the
information recorded in the Neighborhood Information Base, Topology
Set, Attached Network Set and Advertising Remote Node Set.
6.3.1. Advertised Neighbor Set
A node's Advertised Neighbor Set contains interface addresses of
symmetric 1-hop neighbors which are to be advertised through TC
messages:
{A_neighbor_iface_addr}
In addition, an Advertised Neighbor Set Sequence Number (ANSN) is
maintained. Each time the Advertised Neighbor Set is updated, the
ANSN MUST be incremented. The ANSN MUST also be incremented if there
is a change to the set of Local Attached Network Tuples that are to
be advertised in the node's TC messages.
6.3.2. Advertising Remote Node Set
A node's Advertising Remote Node Set records information describing
each remote node in the network that transmits TC messages. It
consists of Advertising Remote Node Tuples:
(AR_orig_addr, AR_seq_number, AR_iface_addr_list, AR_time)
where:
AR_orig_addr is the originator address of a received TC message,
note that this does not include a prefix length;
AR_seq_number is the greatest ANSN in any TC message received which
originated from the node with originator address AR_orig_addr;
AR_iface_addr_list is the list of the interface addresses of the
node with originator address AR_orig_addr;
AR_time is the time at which this Tuple expires and MUST be removed.
6.3.3. Topology Set
A node's Topology Set records topology information about the network.
It consists of Topology Tuples:
(T_dest_iface_addr, T_orig_addr, T_seq_number, T_time)
where:
T_dest_iface_addr is an interface address of a destination node,
which may be reached in one hop from the node with originator
address T_orig_addr;
T_orig_addr is the originator address of a node which is the last
hop on a path towards the node with interface address
T_dest_iface_addr, note that this does not include a prefix
length;
T_seq_number is the greatest received ANSN associated with the
information contained in this Tuple;
T_time specifies the time at which this Tuple expires and MUST be
removed.
6.3.4. Attached Network Set
A node's Attached Network Set records information about networks
attached to other nodes. It consists of Attached Network Tuples:
(AN_net_addr, AN_orig_addr, AN_dist, AN_seq_number, AN_time)
where:
AN_net_addr is the network address of an attached network, which may
be reached via the node with originator address AN_orig_addr;
AN_orig_addr is the originator address of a node which can act as
gateway to the network with address AN_net_addr, note that this
does not include a prefix length;
AN_dist is the number of hops to the network with address
AN_net_addr from the node with originator address AN_orig_addr;
AN_seq_number is the greatest received ANSN associated with the
information contained in this Tuple;
AN_time specifies the time at which this Tuple expires and MUST be
removed.
6.3.5. Routing Set
A node's Routing Set records the selected path to each destination
for which a route is known. It consists of Routing Tuples:
(R_dest_addr, R_next_iface_addr, R_dist, R_local_iface_addr)
where:
R_dest_addr is the address of the destination, either the address of
an interface of a destination node, or the network address of an
attached network;
R_next_iface_addr is the OLSRv2 interface address of the "next hop"
on the selected path to the destination;
R_dist is the number of hops on the selected path to the
destination;
R_local_iface_addr is the address of the local interface over which
a packet MUST be sent to reach the destination by the selected
path.
6.4. Processing and Forwarding Information Base
The Processing and Forwarding Information Base records information
required to ensure that a message is processed at most once and is
forwarded at most once per interface of a node.
6.4.1. Received Set
A node's Received Sets, one per local OLSRv2 interface, each record
the signatures of messages which have been received over that
interface. Each consists of Received Tuples:
(RX_type, RX_orig_addr, RX_seq_number, RX_time) (RX_type, RX_orig_addr, RX_seq_number, RX_time)
where: where:
RX_type is the received message type, or zero if the received RX_type is the received message type, or zero if the received
message sequence number is not type-specific; message sequence number is not type-specific;
RX_orig_addr is the originator address of the received message; RX_orig_addr is the originator address of the received message;
RX_seq_number is the message sequence number of the received RX_seq_number is the message sequence number of the received
message; message;
RX_time specifies the time at which this Tuple expires and MUST be RX_time specifies the time at which this Tuple expires and MUST be
removed. removed.
6.2. Processed Set 6.4.2. Processed Set
A node's Processed Set records signatures of messages which have been A node's Processed Set records signatures of messages which have been
processed by the node. It consists of Processed Tuples: processed by the node. It consists of Processed Tuples:
(P_type, P_orig_addr, P_seq_number, P_time) (P_type, P_orig_addr, P_seq_number, P_time)
where: where:
P_type is the processed message type, or zero if the processed P_type is the processed message type, or zero if the processed
message sequence number is not type-specific; message sequence number is not type-specific;
P_orig_addr is the originator address of the processed message; P_orig_addr is the originator address of the processed message;
P_seq_number is the message sequence number of the processed P_seq_number is the message sequence number of the processed
message; message;
P_time specifies the time at which this Tuple expires and MUST be P_time specifies the time at which this Tuple expires and MUST be
removed. removed.
6.3. Forwarded Set 6.4.3. Forwarded Set
A node's Forwarded Set records signatures of messages which have been A node's Forwarded Set records signatures of messages which have been
processed by the node. It consists of Forwarded Tuples: processed by the node. It consists of Forwarded Tuples:
(F_type, F_orig_addr, F_seq_number, F_time) (F_type, F_orig_addr, F_seq_number, F_time)
where: where:
F_type is the forwarded message type, or zero if the forwarded F_type is the forwarded message type, or zero if the forwarded
message sequence number is not type-specific; message sequence number is not type-specific;
skipping to change at page 13, line 18 skipping to change at page 23, line 4
processed by the node. It consists of Forwarded Tuples: processed by the node. It consists of Forwarded Tuples:
(F_type, F_orig_addr, F_seq_number, F_time) (F_type, F_orig_addr, F_seq_number, F_time)
where: where:
F_type is the forwarded message type, or zero if the forwarded F_type is the forwarded message type, or zero if the forwarded
message sequence number is not type-specific; message sequence number is not type-specific;
F_orig_addr is the originator address of the forwarded message; F_orig_addr is the originator address of the forwarded message;
F_seq_number is the message sequence number of the forwarded F_seq_number is the message sequence number of the forwarded
message; message;
F_time specifies the time at which this Tuple expires and MUST be F_time specifies the time at which this Tuple expires and MUST be
removed. removed.
6.4. Relay Set 6.4.4. Relay Set
A node's Relay Set records the neighbor interface addresses for which
it is to relay flooded messages. It consists of Relay Tuples:
(RY_iface_addr)
where: A node's Relay Sets, one per local OLSRv2 interface, each records the
OLSRv2 interface addresses of symmetric 1-hop neighbors, such that
the node is to forward messages received from those neighbor's OLSRv2
interfaces, on that local OLSRv2 interface, if not otherwise excluded
from forwarding that message (e.g. by it having been previously
forwarded):
RY_iface_addr is the address of a neighbor interface for which the {RY_neighbor_iface_addr}
node SHOULD relay flooded messages. This MUST include a prefix
length.
7. Packet Processing and Message Forwarding 7. Packet Processing and Message Forwarding
On receiving a packet, as defined in [3], a node examines the packet On receiving a packet, as defined in [1], a node examines the packet
header and each of the message headers. If the message type is known header and each of the message headers. If the message type is known
to the node, the message is processed locally according to the to the node, the message is processed locally according to the
specifications for that message type. The message is also specifications for that message type. The message is also
independently evaluated for forwarding. independently evaluated for forwarding.
7.1. Actions when Receiving an OLSRv2 Packet 7.1. Actions when Receiving an OLSRv2 Packet
On receiving a packet, a node MUST perform the following tasks: On receiving a packet, a node MUST perform the following tasks:
1. The packet MAY be fully parsed on reception, or the packet and 1. The packet MAY be fully parsed on reception, or the packet and
skipping to change at page 14, line 40 skipping to change at page 24, line 40
3. Otherwise if the packet header is present and it contains a 3. Otherwise if the packet header is present and it contains a
packet TLV block, then each TLV in it is processed according to packet TLV block, then each TLV in it is processed according to
its type if recognized, otherwise the TLV is ignored; its type if recognized, otherwise the TLV is ignored;
4. Otherwise each message in the packet, if any, is treated 4. Otherwise each message in the packet, if any, is treated
according to Section 7.2. according to Section 7.2.
7.2. Actions when Receiving an OLSRv2 Message 7.2. Actions when Receiving an OLSRv2 Message
A node MUST perform the following tasks for each received OLSRv2 A node MUST perform the following tasks for each received message:
message:
1. If the received OLSRv2 message header cannot be correctly parsed 1. If the message header cannot be correctly parsed according to the
according to the specification in [3], or if the node recognizes specification in [1], or if the node recognizes from the
from the originator address of the message that the message is originator address of the message that the message is one which
one which the receiving node itself originated, then the message the receiving node itself originated, then the message MUST be
MUST be silently discarded; silently discarded.
2. Otherwise: 2. Otherwise:
1. If the received message is of a known type then the message 1. If the message is a HELLO message, then the message is
is considered for processing according to Section 7.3, AND; processed according to Section 10.
2. If for the received message (<hop-limit> + <hop-count>) > 1,
then the message is considered for forwarding according to 2. Otherwise:
1. If the message is of a known type, then the message is
considered for processing according to Section 7.3, AND;
2. If for the message (<hop-limit> + <hop-count>) > 1, then
the message is considered for forwarding according to
Section 7.4. Section 7.4.
7.3. Message Considered for Processing 7.3. Message Considered for Processing
If a message (the "current message") is considered for processing, If a message (the "current message") is considered for processing,
the following tasks MUST be performed: then the following tasks MUST be performed:
1. If an entry exists in the Processed Set where: 1. If a Processed Tuple exists with:
* P_type == the message type of the current message, or 0 if the * P_type == the message type of the current message, or 0 if the
typedep bit in the message semantics octet (in the message typedep bit in the message semantics octet in the message
header) of the current message is cleared ('0'), AND; header of the current message is cleared ('0'), AND;
* P_orig_addr == the originator address of the current message, * P_orig_addr == the originator address of the current message,
AND; AND;
* P_seq_number == the message sequence number of the current * P_seq_number == the message sequence number of the current
message. message;
then the current message MUST NOT be processed. then the current message MUST NOT be processed.
2. Otherwise: 2. Otherwise:
1. Create an entry in the Processed Set with: 1. Create a Processed Tuple with:
+ P_type = the message type of the current message, or 0 if + P_type = the message type of the current message, or 0 if
the typedep bit in the message semantics octet (in the the typedep bit in the message semantics octet in the
message header) of the current message is cleared ('0'); message header of the current message is cleared ('0');
+ P_orig_addr = originator address of the current message; + P_orig_addr = the originator address of the current
message;
+ P_seq_number = sequence number of the current message; + P_seq_number = the sequence number of the current message;
+ P_time = current time + P_HOLD_TIME. + P_time = current time + P_HOLD_TIME.
2. Process the message according to its type. 2. Process the current message according to its type.
7.4. Message Considered for Forwarding 7.4. Message Considered for Forwarding
If a message is considered for forwarding, and it is either of a If a message is considered for forwarding, and it is either of a
message type defined in this document or of an unknown message type, message type defined in this document (i.e. is a TC message) or of an
then it MUST use the following algorithm. A message type not defined unknown message type, then it MUST use the following algorithm. A
in this document MAY specify the use of this, or another algorithm. message of a message type not defined in this document MAY, in an
(Such an other algorithm MAY use the Received Set for the receiving extension to this protocol, specify the use of this, or another
interface, it SHOULD use the Forwarded Set similarly to the following algorithm. (Such an other algorithm MAY use the Received Set for the
algorithm.) receiving interface, it SHOULD use the Forwarded Set similarly to the
If a message is considered for forwarding according to this following algorithm.)
algorithm, the following tasks MUST be performed:
1. If the sending interface (as indicated by the source interface of If a message (the "current message") is considered for forwarding
the IP datagram containing the message) does not match (taking according to this algorithm, the following tasks MUST be performed:
into account any address prefix of) any N_neighbor_iface_addr in
any Symmetric Neighbor Tuple, then the message MUST be silently 1. If the sending interface address (the source address of the IP
discarded. datagram containing the current message) does not match (taking
into account any address prefix of) an OLSRv2 interface address
in an L_neighbor_iface_addr_list of a Link Tuple, with L_status
== SYMMETRIC, in the Link Set for the OLSRv2 interface on which
the current message was received (the "receiving interface") then
the current message MUST be silently discarded.
2. Otherwise: 2. Otherwise:
1. If an entry exists in the Received Set for the receiving 1. If a Received Tuple exists in the Received Set for the
interface, where: receiving interface, with:
+ RX_type == the message type, or 0 if the typedep bit in + RX_type == the message type of the current message, or 0
the message semantics octet (in the message header) is if the typedep bit in the message semantics octet in the
cleared ('0'), AND; message header of the current message is cleared ('0'),
AND;
+ RX_orig_addr == the originator address of the received + RX_orig_addr == the originator address of the current
message, AND; message, AND;
+ RX_seq_number == the sequence number of the received + RX_seq_number == the sequence number of the current
message. message;
then the message MUST be silently discarded. then the current message MUST be silently discarded.
2. Otherwise: 2. Otherwise:
1. Create an entry in the Received Set for the receiving 1. Create a Received Tuple in the Received Set for the
interface with: receiving interface with:
- RX_type = the message type, or 0 if the typedep bit in - RX_type = the message type of the current message, or
the message semantics octet (in the message header) is 0 if the typedep bit in the message semantics octet in
cleared ('0'); the message header of the current message is cleared
('0');
- RX_orig_addr = originator address of the message; - RX_orig_addr = originator address of the current
message;
- RX_seq_number = sequence number of the message; - RX_seq_number = sequence number of the current
message;
- RX_time = current time + RX_HOLD_TIME. - RX_time = current time + RX_HOLD_TIME.
2. If an entry exists in the Forwarded Set where: 2. If a Forwarded Tuple exists with:
- F_type == the message type, or 0 if the typedep bit in - F_type == the message type of the current message, or
the message semantics octet (in the message header) is 0 if the typedep bit in the message semantics octet in
cleared ('0'); the message header of the current message is cleared
- F_orig_addr == the originator address of the received ('0');
- F_orig_addr == the originator address of the current
message, AND; message, AND;
- F_seq_number == the sequence number of the received - F_seq_number == the sequence number of the current
message. message.
then the message MUST be silently discarded. then the current message MUST be silently discarded.
3. Otherwise if a Relay Tuple exists whose RY_iface_addr 3. Otherwise if the sending interface address matches
matches (taking into account any address prefix) the (taking account of any address prefix of) an
sending interface (as indicated by the source interface RY_neighbor_iface_addr in the Relay Set for the receiving
of the IP datagram containing the message): interface, then:
1. Create an entry in the Forwarded Set with: 1. Create a Forwarded Tuple with:
o F_type = the message type, or 0 if the typedep bit o F_type = the message type of the current message,
in the message semantics octet (in the message or 0 if the typedep bit in the message semantics
header) is cleared ('0'); octet in the message header of the current message
is cleared ('0');
o F_orig_addr = originator address of the message; o F_orig_addr = originator address of the current
message;
o F_seq_number = sequence number of the message; o F_seq_number = sequence number of the current
message;
o F_time = current time + F_HOLD_TIME. o F_time = current time + F_HOLD_TIME.
2. The message header is modified as follows: 2. The message header of the current message is modified
by:
o Decrement <hop-limit> in the message header by 1;
o Increment <hop-count> in the message header by 1;
3. Transmit the message on all OLSRv2 interfaces of the
node.
Messages are retransmitted in the format specified by [3] with the
ALL-MANET-NEIGHBORS address (see [4]) as destination IP address.
8. Information Repositories
The purpose of OLSRv2 is to determine the Routing Set, which may be
used to update IP's Routing Table, providing "next hop" routing
information for IP datagrams. In order to accomplish this, OLSRv2
uses a number of protocol sets: the Neighborhood Information Base,
provided by [4], is in OLSRv2 augmented by information allowing MPR
selection and signaling. Additionally, OLSRv2 specifies a Topology
Information Base, which describes the information used for and
acquired through TC message exchange - in other words: the Topology
Information Base represents the network topology graph as seen from
each node.
Addresses (other than originator addresses) recorded in the
Neighborhood Information Base and the Topology Information Base MUST
all be recorded with prefix lengths, in order to allow comparison
with addresses received in HELLO and TC messages.
8.1. Neighborhood Information Base
The Neighborhood Information Base stores information about links
between local interfaces and interfaces on adjacent nodes. In
addition to the sets described in [4], OLSRv2 adds an element to each
Link Tuple to allow a node to record the willingness of a 1-hop
neighbor node to be selected as an MPR. Also, OLSRv2 adds an MPR Set
and an MPR Selector Set to the Neighborhood Information Base. The
MPR Set is used by a node to record which of its symmetric 1-hop
neighbors are selected as MPRs, and the MPR Selector Set is used by a
node to record which of its symmetric 1-hop neighbors have selected
it as MPR. Thus, in addition to what is specified in [4], the MPR
Set is used when generating HELLO messages, and the MPR Selector Set
is populated when processing HELLO messages.
8.1.1. Link Set
Link Tuples are as specified in [4], augmented with:
L_willingness is the node's willingness to be selected as an MPR;
8.1.2. MPR Set
A node's MPR Set contains OLSRv2 interface addresses with which the
node has a symmetric link and which are of 1-hop symmetric neighbors
which the node has selected as MPRs:
(MP_neighbor_iface_addr)
8.1.3. MPR Selector Set
A node's MPR Selector Set records the nodes which have selected this
node as an MPR. It consists of MPR Selector Tuples:
(MS_neighbor_iface_addr, MS_time)
where:
MS_neighbor_iface_addr is an OLSRv2 interface address with which
this node has a symmetric link and which is of a 1-hop symmetric
neighbor which has selected this node as an MPR;
MS_time specifies the time at which this Tuple expires and MUST be
removed.
8.2. Topology Information Base
The Topology Information Base stores information, required for the
generation and processing of TC messages. The Advertised Neighbor
Set contains OLSRv2 interface addresses of symmetric 1-hop neighbors
which are to be reported in TC messages. The Topology Set and
Attached Network Set both record information received through TC
messages. Thus the Advertised Neighbor Set is used for generating TC
messages, while the Topology Set and Attached Network Set are
populated when processing TC messages.
Additionally, a Routing Set is maintained, derived from the
information recorded in the Neighborhood Information Base, Topology
Set and Attached Network Set.
8.2.1. Advertised Neighbor Set
A node's Advertised Neighbor Set contains OLSRv2 interface addresses
of symmetric 1-hop neighbors which are to be advertised through TC
messages:
(A_neighbor_iface_addr)
In addition, an Advertised Neighbor Set Sequence Number (ANSN) is
maintained. Each time the Advertised Neighbor Set is updated, the
ANSN MUST be incremented. The ANSN MUST also be incremented if there
is a change to the set of Local Attached Network Tuples that are to
be advertised in the node's TC messages.
8.2.2. ANSN History Set
A node's ANSN History Set records information about the freshness of
the topology information received from each other node. It consists
of ANSN History Tuples:
(AH_orig_addr, AH_seq_number, AH_time)
where:
AH_orig_addr is the originator address of a received TC message,
note that this does not include a prefix length;
AH_seq_number is the highest ANSN in any TC message received which
originated from AH_orig_addr;
AH_time is the time at which this Tuple expires and MUST be removed.
8.2.3. Topology Set
A node's Topology Set records topology information about the network.
It consists of Topology Tuples:
(T_dest_iface_addr, T_last_iface_addr, T_seq_number, T_time)
where:
T_dest_iface_addr is an OLSRv2 interface address of a destination
node, which may be reached in one hop from the node with the
OLSRv2 interface address T_last_iface_addr;
T_last_iface_addr is, conversely, an OLSRv2 interface address of a
node which is the last hop on a path towards the node with OLSRv2
interface address T_dest_iface_addr.
T_seq_number is the highest received ANSN associated with the
information contained in this Topology Tuple;
T_time specifies the time at which this Tuple expires and MUST be
removed.
8.2.4. Attached Network Set
A node's Attached Network Set records information about networks
attached to other nodes. It consists of Attached Network Tuples:
(AN_net_addr, AN_gw_iface_addr, AN_dist, AN_seq_number, AN_time)
where:
AN_net_addr is the network address of an attached network, which may
be reached via the node with the OLSRv2 interface address
AN_gw_iface_addr;
AN_gw_iface_addr is the address of an OLSRv2 interface of a node
which can act as gateway to the network with address AN_net_addr;
AN_dist is the number of hops to the network with address
AL_net_addr from the node with address AN_gw_iface_addr.
AN_seq_number is the highest received ANSN associated with the
information contained in this Attached Network Tuple;
AN_time specifies the time at which this Tuple expires and MUST be
removed.
8.2.5. Routing Set
A node's Routing Set records the selected path to each destination
for which a route is known. It consists of Routing Tuples:
(R_dest_addr, R_next_iface_addr, R_dist, R_local_iface_addr)
where:
R_dest_addr is the address of the destination, either the address of
an OLSRv2 interface of a destination node, or the network address
of an attached network;
R_next_iface_addr is the OLSRv2 interface address of the "next hop" o decrement <hop-limit> in the message header by 1;
on the selected path to the destination;
R_dist is the number of hops on the selected path to the o increment <hop-count> in the message header by 1.
destination;
R_local_iface_addr is the address of the local interface over which 3. For each OLSRv2 interface of the node, include the
a packet MUST be sent to reach the destination. message in a packet to be transmitted on that OLSRv2
interface, as described in Section 8. This packet
may contain other forwarded messages and/or messages
generated by this node. Forwarded messages may be
jittered as described in [3]. The value of MAXJITTER
used in jittering a forwarded message MAY be based on
information in that message (in particular any
INTERVAL_TIME or VALIDITY_TIME TLVs in that message)
or otherwise SHOULD be with maximum delay of
F_MAXJITTER. A node MAY reduce the jitter applied to
a message in order to more efficiently combine
messages in packets.
9. Control Message Structures 8. Packets and Messages
Nodes using OLSRv2 exchange information through messages. One or Nodes using OLSRv2 exchange information through messages. One or
more messages sent by a node at the same time SHOULD be combined into more messages sent by a node at the same time SHOULD be combined into
a single packet. These messages may have originated at the sending a single packet. These messages may have originated at the sending
node, or have originated at another node and are forwarded by the node, or have originated at another node and are forwarded by the
sending node. Messages with different originators may be combined in sending node. Messages with different originators MAY be combined in
the same packet. the same packet. Messages from other protocols defined using [1] MAY
be combined in the same packet.
The packet and message format used by OLSRv2 is defined in [3]. OLSRv2 packets are sent using UDP, on the port "manet" defined in
[6]. Their IP datagrams are transmitted using the well-known
multicast address "LL MANET Routers" defined in [6].
The packet and message format used by OLSRv2 is defined in [1].
However this specification contains some options which are not used However this specification contains some options which are not used
by OLSRv2. In particular (using the syntactical entities defined in by OLSRv2. In particular (using the syntactical entities defined in
[3]): [1]):
o All OLSRv2 packets, not limited to those defined in this document, o All OLSRv2 packets, not limited to those defined in this document,
include a <packet-header>. include a <packet-header>.
o All OLSRv2 packets, not limited to those defined in this document, o All OLSRv2 packets, not limited to those defined in this document,
have the pseqnum bit of <packet-semantics> cleared ('0'), i.e. have the pseqnum bit of <packet-semantics> cleared ('0'), i.e.
they include a packet sequence number. they include a packet sequence number.
o OLSRv2 packets MAY include packet TLVs, however OLSRv2 itself does o OLSRv2 packets MAY include packet TLVs, however OLSRv2 itself does
not specify any packet TLVs. not specify any packet TLVs.
o All OLSRv2 messages, not limited to those defined in this o All OLSRv2 messages, not limited to those defined in this
document, include a full <msg-header> and hence have the noorig document, include a full <msg-header> and hence have the noorig,
and nohops bits of <msg-semantics> cleared ('0'). nohoplimit, nohopcount and noseqnum bits of <msg-semantics>
cleared ('0').
o All OLSRv2 message defined in this document have the typedep bit o All OLSRv2 messages defined in this document have the typedep bit
of <msg-semantics> cleared ('0'). of <msg-semantics> cleared ('0').
Other options defined in [3] may be freely used, in particular any o All references in this document to specific TLVs in generating and
processing HELLO and TC messages refer to TLVs with Subtype == 0.
TLVs with nonzero subtype are treated as of unknown type when
processing messages, i.e. they are ignored.
Other options defined in [1] may be freely used, in particular any
other values of <packet-semantics>, <addr-semantics> or <tlv- other values of <packet-semantics>, <addr-semantics> or <tlv-
semantics> consistent with its specification. semantics> consistent with their specifications.
The remainder of this section defines, within the framework of [3], The remainder of this section defines, within the framework of [1],
message types and TLVs specific to OLSRv2. message types and TLVs specific to OLSRv2.
9.1. HELLO Messages 8.1. HELLO Messages
A HELLO message in OLSRv2 is generated as specified in [4]. A HELLO message in OLSRv2 is generated as specified in [4].
Additionally, an OLSRv2 node: Additionally, an OLSRv2 node:
o MUST include TLV(s) with Type == MPR associated with all OLSRv2 o MUST include TLV(s) with Type == MPR associated with all OLSRv2
interface addresses included in the HELLO message with a TLV with interface addresses included in the HELLO message with a TLV with
Type == LINK_STATUS and Value == SYMMETRIC if that address is also Type == LINK_STATUS and Value == SYMMETRIC if that address is also
included in the node's MPR Set (if there is more than one copy of included in Neighbor Tuple with N_mpr == true. (If there is more
the address, this applies to the specific copy of the address to than one copy of such an address in the HELLO message, then this
which the LINK_STATUS TLV is associated); applies to the specific copy of the address with which the
LINK_STATUS TLV is associated.)
o MUST NOT include any TLVs with Type == MPR associated with any o MUST NOT include any TLVs with Type == MPR associated with any
other addresses; other addresses.
o MAY include a message TLV with Type == WILLINGNESS, indicating the o MAY include a message TLV with Type == WILLINGNESS, indicating the
node's willingness to be selected as an MPR. node's willingness to be selected as an MPR.
9.1.1. HELLO Message TLVs 8.1.1. HELLO Message TLVs
In a HELLO message, a node MAY include a WILLINGNESS message TLV as In a HELLO message, a node MAY include a WILLINGNESS message TLV as
specified in Table 1. specified in Table 1. A node MUST NOT include more than one
WILLINGNESS message TLV.
+----------------+------+-------------------+-----------------------+ +-------------+------+---------+--------+---------------------------+
| Name | Type | Length | Value | | Name | Type | Subtype | Length | Value |
+----------------+------+-------------------+-----------------------+ +-------------+------+---------+--------+---------------------------+
| WILLINGNESS | TBD | 8 bits | The node's | | WILLINGNESS | TBD | 0 | 8 bits | The node's willingness to |
| | | | willingness to be | | | | | | be selected as MPR; |
| | | | selected as MPR; | | | | | | unused bits (based on the |
| | | | unused bits (based on | | | | | | maximum willingness value |
| | | | the maximum | | | | | | WILL_ALWAYS) are RESERVED |
| | | | willingness value | | | | | | and SHOULD be set to zero |
| | | | WILL_ALWAYS) are | +-------------+------+---------+--------+---------------------------+
| | | | RESERVED and SHOULD |
| | | | be set to zero |
+----------------+------+-------------------+-----------------------+
Table 1 Table 1
A node's willingness to be selected as MPR ranges from WILL_NEVER A node's willingness to be selected as MPR ranges from WILL_NEVER
(indicating that a node MUST NOT be selected as MPR by any node) to (indicating that a node MUST NOT be selected as MPR by any node) to
WILL_ALWAYS (indicating that a node MUST always be selected as MPR). WILL_ALWAYS (indicating that a node MUST always be selected as MPR).
If a node does not advertise a Willingness TLV in HELLO messages, the If a node does not advertise a Willingness TLV in HELLO messages,
node MUST be assumed to have a willingness of WILL_DEFAULT. then the node MUST be assumed to have a willingness of WILL_DEFAULT.
9.1.2. HELLO Message Address Block TLVs 8.1.2. HELLO Message Address Block TLVs
In a HELLO message, a node MAY include MPR address block TLV(s) as In a HELLO message, a node MAY include MPR address block TLV(s) as
specified in Table 2. specified in Table 2.
+----------------+------+-------------------+----------------------+ +------+------+---------+--------+-------+
| Name | Type | Length | Value | | Name | Type | Subtype | Length | Value |
+----------------+------+-------------------+----------------------+ +------+------+---------+--------+-------+
| MPR | TBD | 0 bits | None | | MPR | TBD | 0 | 0 bits | None |
+----------------+------+-------------------+----------------------+ +------+------+---------+--------+-------+
Table 2 Table 2
9.2. TC Messages 8.2. TC Messages
A TC message MUST contain: A TC message MUST contain:
o A message TLV with Type == CONT_SEQ_NUM, as specified in o A message TLV with Type == CONT_SEQ_NUM, as specified in
Section 9.2.1. Section 8.2.1.
o A message TLV with Type == VALIDITY_TIME, as specified in o A message TLV with Type == VALIDITY_TIME, as specified in [2].
Appendix E.
o A first address block containing all of the node's OLSRv2 o A first address block (the Local Address Block) containing all of
interface addresses. This is similar to the Local Interface Block the node's interface addresses. This is similar to the Local
included in HELLO messages as specified in [4], however in a TC Interface Block included in HELLO messages as specified in [4],
message these addresses MUST be included in the same order in all however in a TC message these addresses MUST be included in the
copies of a given TC message, regardless of which OLSRv2 interface same order in all copies of a given TC message, regardless of
it is transmitted on, and no OTHER_IF address block TLVs are which OLSRv2 interface it is transmitted on, and no OTHER_IF
required. address block TLVs are required.
o Additional address block(s) containing all addresses in the o Except when they would be empty, or when including a message TLV
Advertised Address Set and selected addresses in the Local with Type == INCOMPLETE (in which case the TC message does not
Attached Network Set, the latter (only) with associated GATEWAY satisfy the necessary transmission constraints defined by
address block TLV(s), as specified in Section 9.2.2. TC_INTERVAL and T_HOLD_TIME), address block(s) (Advertised Address
Blocks) containing addresses in the Advertised Address Set and
selected addresses in the Local Attached Network Set, the latter
(only) with associated GATEWAY address block TLV(s), as specified
in Section 8.2.2.
A TC message MAY contain: A TC message MAY contain:
o A message TLV with Type == INTERVAL_TIME, as specified in o A message TLV with Type == INTERVAL_TIME, as specified in [2].
Appendix E.
o A message TLV with Type == INCOMPLETE, as specified in o A message TLV with Type == INCOMPLETE, as specified in
Section 9.2.1. Section 8.2.1.
9.2.1. TC Message TLVs 8.2.1. TC Message TLVs
In a TC message, a node MUST include a CONT_SEQ_NUM message TLV, and In a TC message, a node MUST include a CONT_SEQ_NUM message TLV, and
MAY contain an INCOMPLETE message TLV, as specified in Table 3. MAY contain an INCOMPLETE message TLV, as specified in Table 3. A
node MUST NOT include more than one CONT_SEQ_NUM message TLV or
INCOMPLETE message TLV.
+----------------+------+-------------------+-----------------------+ +--------------+------+---------+--------+--------------------------+
| Name | Type | Length | Value | | Name | Type | Subtype | Length | Value |
+----------------+------+-------------------+-----------------------+ +--------------+------+---------+--------+--------------------------+
| CONT_SEQ_NUM | TBD | 8 bits | The ANSN contained in | | CONT_SEQ_NUM | TBD | 0 | 8 bits | The ANSN contained in |
| | | | the Advertised | | | | | | the Advertised Neighbor |
| | | | Neighbor Set | | | | | | Set |
| | | | | | | | | | |
| INCOMPLETE | TBD | 0 bits | None | | INCOMPLETE | TBD | 0 | 0 bits | None |
+----------------+------+-------------------+-----------------------+ +--------------+------+---------+--------+--------------------------+
Table 3 Table 3
9.2.2. TC Message Address Block TLVs 8.2.2. TC Message Address Block TLVs
In a TC message, a node MAY include GATEWAY address block TLV(s) as In a TC message, a node MAY include GATEWAY address block TLV(s) as
specified in Table 4. specified in Table 4.
+----------------+------+-------------------+-----------------------+ +---------+------+---------+--------+-------------------------------+
| Name | Type | Length | Value | | Name | Type | Subtype | Length | Value |
+----------------+------+-------------------+-----------------------+ +---------+------+---------+--------+-------------------------------+
| GATEWAY | TBD | 8 bits | Number of hops to | | GATEWAY | TBD | 0 | 8 bits | Number of hops to attached |
| | | | attached network | | | | | | network |
+----------------+------+-------------------+-----------------------+ +---------+------+---------+--------+-------------------------------+
Table 4 Table 4
10. HELLO Message Generation 9. HELLO Message Generation
An OLSRv2 HELLO message is composed as defined in [4], with the An OLSRv2 HELLO message is composed as defined in [4], with the
following additions: following additions:
o A message TLV with Type == WILLINGNESS and Value == the node's o A message TLV with Type == WILLINGNESS and Value == the node's
willingness to act as an MPR, MAY be included. willingness to act as an MPR, MAY be included.
o For each address which is included in the message with an o For each address which is included in the message with an
associated TLV with Type == LINK_STATUS, and is of an MPR (i.e. is associated TLV with Type == LINK_STATUS and Value == SYMMETRIC,
an MP_neighbor_iface_addr), an address TLV with Type == MPR MUST and is of an MPR (i.e. the address is in the
be included; this TLV MUST be associated with the same copy of the N_neighbor_iface_addr_list of a Neighbor Tuple with N_mpr ==
address as is the TLV with Type == LINK_STATUS. true), an address block TLV with Type == MPR MUST be included;
this TLV MUST be associated with the same copy of the address as
is the TLV with Type == LINK_STATUS.
o For address which is included in the message and is not of an MPR o For each address which is included in the message and is not
(i.e. is not an MP_neighbor_iface_addr) or is not associated with associated with a TLV with Type == LINK_STATUS and Value ==
a TLV with Type == LINK_STATUS, an address TLV with Type == MPR SYMMETRIC, or is not of an MPR (i.e. the address is not in the
MUST NOT be included. N_neighbor_iface_addr_list of a Neighbor Tuple with N_mpr ==
true), an address block TLV with Type == MPR MUST NOT be
associated wit this address.
o For each Local Attached Tuple with AL_dist == 0, a node MAY o For each Local Attached Tuple with AL_dist == 0, a node MAY
include AL_net_addr in the Local Interface Block of the message, include AL_net_addr in the Local Interface Block of the message,
with an associated TLV with Type == OTHER_IF. with an associated TLV with Type == OTHER_IF.
10.1. HELLO Message: Transmission 9.1. HELLO Message: Transmission
HELLO messages are included in packets as specified in [3]. These HELLO messages are included in packets as specified in [1]. These
packets may contain other messages, including TC messages. packets may contain other messages, including TC messages.
11. HELLO Message Processing 10. HELLO Message Processing
Subsequent to the processing of HELLO messages, as specified in [4], Subsequent to the processing of HELLO messages, as specified in [4],
the node MUST: the node MUST identify the Neighbor Tuple which was created or
updated by the processing specified in [4] (the "current Neighbor
1. Determine the willingness of the originating node to be an MPR Tuple") and update N_willingness as described in Section 10.1 and
by: N_mpr_selector as described in Section 10.2.
* if the HELLO message contains a message TLV with Type ==
WILLINGNESS then the willingness is the value of that TLV,
ignoring the reserved bits in that field;
* otherwise the willingness is WILL_DEFAULT.
2. Update each Link Tuple for which any address in its
L_neighbor_iface_addr_list is present in the Local Interface
Block of the HELLO message, with:
* L_willingness = the willingness of the originating node.
3. Update its MPR Selector Set, according to Section 11.1.
11.1. Populating the MPR Selector Set
On receiving a HELLO message:
1. If a node finds one of its OLSRv2 interface addresses with an 10.1. Updating Willingness
associated TLV with Type == MPR in the HELLO message (indicating
that the originator node has selected the receiving node as an
MPR), the MPR Selector Set MUST be updated as follows:
1. For each address, henceforth neighbor address, in the Local N_willingness in the current Neighbor Tuple is updated as follows:
Interface Block of the received HELLO message, where the
neighbor address is present as an N_neighbor_iface_addr in a
Symmetric Neighbor Tuple with N_STATUS == SYMMETRIC:
1. If there exists no MPR Selector Tuple with: 1. if the HELLO message contains a message TLV with Type ==
WILLINGNESS then N_willingness is set to the value of that TLV;
- MS_neighbor_iface_addr == neighbor address 2. otherwise, N_willingness is set to WILL_DEFAULT.
then a new MPR Selector Tuple is created with: 10.2. Updating MPR Selectors
- MS_neighbor_iface_addr = neighbor address N_mpr_selector is updated as follows:
2. The MPR Selector Tuple (new or otherwise) with: 1. If a node finds one of its local OLSRv2 interface addresses with
an associated TLV with Type == MPR in the HELLO message
(indicating that the originator node has selected the receiving
node as an MPR), then N_mpr_selector in the current Neighbor
Tuple is set true.
- MS_neighbor_iface_addr == neighbor address 2. Otherwise, if a node finds one of its own interface addresses
is then modified as follows: with an associated TLV with Type == LINK_STATUS and Value ==
SYMMETRIC in the HELLO message, then N_mpr_selector in the
current Neighbor Tuple is set false.
- MS_time = current time + validity time 10.3. Symmetric 1-Hop and 2-Hop Neighborhood Changes
2. Otherwise if a node finds one of its own interface addresses with A node MUST also perform the following:
an associated TLV with Type == LINK_STATUS and Value == SYMMETRIC
in the HELLO message, the MPR Selector Set MUST be updated as
follows:
1. All MPR Selector Tuples whose MS_neighbor_iface_addr is in 1. If N_symmetric of a Neighbor Tuple changes from true to false,
the Local Interface Block of the HELLO message are removed. then N_mpr_selector of that Neighbor Tuple MUST be set false.
MPR Selector Tuples are also removed upon expiration of MS_time, or 2. The set of MPRs of a node MUST be recalculated if:
upon symmetric link breakage as described in Section 11.2.
11.2. Symmetric Neighborhood and 2-Hop Neighborhood Changes * a Link Tuple is added with L_status == SYMMETRIC, OR;
A node MUST also perform the following: * a Link Tuple with L_status == SYMMETRIC is removed, OR;
1. If a Link Tuple with L_STATUS == SYMMETRIC is removed, or its * a Link Tuple with L_status == SYMMETRIC changes to having
L_STATUS changes from SYMMETRIC to HEARD or LOST, and for each L_status == HEARD or L_status == LOST, OR;
address in that Link Tuple's L_neighbor_iface_addr_list, if it is * a Link Tuple with L_status == HEARD or L_status == LOST
an MS_neighbor_iface_addr of an MPR Selector Tuple, then that MPR changes to having L_status == SYMMETRIC, OR;
Selector Tuple MUST be removed.
2. If any of: * a 2-Hop Tuple is added or removed, OR;
* a Link Tuple is added with L_STATUS == SYMMETRIC, OR; * the N_willingness of a Neighbor Tuple with N_symmetric == true
changes from WILL_NEVER to any other value, OR;
* a Link Tuple with L_STATUS == SYMMETRIC is removed, or its * the N_willingness of a Neighbor Tuple with N_symmetric == true
L_STATUS changes from SYMMETRIC to HEARD or LOST, or vice and N_mpr == true changes to WILL_NEVER from any other value,
versa, OR; OR;
* a 2-Hop Neighbor Tuple is added or removed, OR; * the N_willingness of a Neighbor Tuple with N_symmetric == true
and N_mpr == false changes to WILL_ALWAYS from any other
value.
* the Neighbor Address Association Set is changed such that the 3. Otherwise the set of MPRs of a node MAY be recalculated if the
subset of any NA_neighbor_iface_addr_list consisting of those N_willingness of a Neighbor Tuple with N_symmetric == true
addresses which are in the L_neighbor_iface_addr_list of a changes in any other way; it SHOULD be recalculated if N_mpr ==
Link Tuple with L_STATUS == SYMMETRIC is changed, including false and this is an increase in N_willingness or if N_mpr ==
the cases of removal or addition of a Neighbor Address true and this is a decrease in N_willingness.
Association Tuple containing any such addresses;
then the MPR Set MUST be recalculated. If the set of MPRs of a node is recalculated, this MUST be as
described in Section 13. Before that calculation the N_mpr of all
Neighbor Tuples are set false, after that calculation the N_mpr of
all Neighbor Tuples representing symmetric 1-hop neighbors which are
chosen as MPRs, are set true.
An additional HELLO message MAY be sent when the MPR Set changes, in An additional HELLO message MAY be sent when the node's set of MPRs
addition to the cases specified in [4], and subject to the same changes, in addition to the cases specified in [4], and subject to
constraints. the same constraints.
12. TC Message Generation 11. TC Message Generation
A node with one or more OLSRv2 interfaces, and with a non-empty A node with one or more OLSRv2 interfaces, and with a non-empty
Advertised Neighbor Set or which acts as a gateway to an associated Advertised Neighbor Set or which acts as a gateway to an attached
network which is to be advertised in the MANET, MUST generate TC network which is to be advertised in the MANET by this node, MUST
messages. A node with an empty Advertised Neighbor Set and which is generate TC messages. A node with an empty Advertised Neighbor Set
not acting as such a gateway SHOULD also generate "empty" TC messages and which is not acting as such a gateway SHOULD also generate
for a period A_HOLD_TIME after it last generated a non-empty TC "empty" TC messages for a period A_HOLD_TIME after it last generated
message. TC messages (non-empty and empty) are generated according a non-empty TC message. TC messages (non-empty and empty) are
to the following: generated according to the following:
1. The message hop count MUST be set to zero. 1. The message hop count MUST be set to zero.
2. The message hop limit MAY be set to any positive value, this 2. The message hop limit MAY be set to any positive value, this
SHOULD be at least two. A node MAY: SHOULD be at least two. A node MAY:
* use the same hop limit in all TC messages, this MUST be at * use the same hop limit TC_HOP_LIMIT in all TC messages, this
least equal to the network diameter in hops, a value of 255 is MUST be at least equal to the network diameter in hops; OR
RECOMMENDED in this case; OR
* use different hop limits in TC messages, this MUST regularly * use different values of the hop limit TC_HOP_LIMIT in TC
include messages with hop limit at least equal to the network messages, this MUST regularly include messages with hop limit
diameter, a value of 255 is RECOMMENDED for these messages; as defined above, other, lower, hop limits SHOULD use a
other hop limits SHOULD use a regular pattern with a regular regular pattern with a regular message interval at any given
interval at any given number of hops distance. number of hops distance.
3. The message MUST contain a message TLV with Type == CONT_SEQ_NUM 3. The message MUST contain a message TLV with Type == CONT_SEQ_NUM
and Value == ANSN from the Advertised Neighbor Set. and Value == ANSN from the Advertised Neighbor Set.
4. The message MUST contain a message TLV with Type == 4. The message MUST contain a message TLV with Type ==
VALIDITY_TIME, as specified in Appendix E.2. If all TC messages VALIDITY_TIME, as specified in [2]. If all TC messages are sent
are sent with the same hop limit (usually 255) then this TLV MUST with the same hop limit then this TLV MUST have Value ==
have Value == T_HOLD_TIME. If TC messages are sent with T_HOLD_TIME. If TC messages are sent with different hop limits,
different hop limits, then this TLV MUST specify times which vary
with the number of hops distance appropriate to the chosen
pattern of TC message hop limits, these times SHOULD be
appropriate multiples of T_HOLD_TIME.
5. The message MAY contain a message TLV with Type == INTERVAL_TIME,
as specified in Appendix E.2. If all TC messages are sent with
the same hop limit (usually 255) then this TLV MUST have Value ==
TC_INTERVAL. If TC messages are sent with different hop limits,
then this TLV MUST specify times which vary with the number of then this TLV MUST specify times which vary with the number of
hops distance appropriate to the chosen pattern of TC message hop hops distance appropriate to the chosen pattern of TC message hop
limits, these times SHOULD be appropriate multiples of limits, these times SHOULD be appropriate multiples of
TC_INTERVAL. T_HOLD_TIME.
6. The message MUST contain the addresses of all of its OLSRv2 5. The message MAY contain a message TLV with Type == INTERVAL_TIME,
interfaces in its first address block, note that the TC message as specified in [2]. If all TC messages are sent with the same
generated on all OLSRv2 interfaces MUST be identical (including hop limit then this TLV MUST have Value == TC_INTERVAL. If TC
having identical message sequence number) and hence these messages are sent with different hop limits, then this TLV MUST
addresses are not ordered or otherwise identified according to specify times which vary with the number of hops distance
the interface on which the TC message is transmitted. appropriate to the chosen pattern of TC message hop limits, these
times SHOULD be appropriate multiples of TC_INTERVAL.
7. The message MUST contain, in address blocks other than its first: 6. The message MUST contain the addresses of all of its interfaces
in its first address block (the "Local Address Block"). Note
that the TC message generated on all OLSRv2 interfaces MUST be
identical (including having identical message sequence number)
and hence these addresses are not ordered or otherwise identified
according to the interface on which the TC message is
transmitted.
7. The message MUST contain, in address blocks other than its first
("Advertised Address Blocks"):
1. A_neighbor_iface_addr from each Advertised Neighbor Tuple; 1. A_neighbor_iface_addr from each Advertised Neighbor Tuple;
2. AL_net_addr from each Local Attached Neighbor Tuple with 2. AL_net_addr from each Local Attached Neighbor Tuple with
AL_dist > 0, each associated with a TLV with Type == GATEWAY AL_dist > 0, each associated with a TLV with Type == GATEWAY
and Value == AL_dist. and Value == AL_dist.
8. The message MAY contain, in address blocks other than its first: 8. The message MAY contain, in address blocks other than its first:
1. AL_net_addr from each Local Attached Neighbor Tuple with 1. AL_net_addr from each Local Attached Neighbor Tuple with
AL_dist == 0, each associated with a TLV with Type == GATEWAY AL_dist == 0, each associated with a TLV with Type == GATEWAY
and Value == 0. and Value == 0.
12.1. TC Message: Transmission 11.1. TC Message: Transmission
TC messages are generated and transmitted periodically on all OLSRv2 TC messages are generated and transmitted periodically on all OLSRv2
interfaces, with a default interval between two consecutive TC interfaces, with a default interval between two consecutive TC
emissions by the same node of TC_INTERVAL. emissions by the same node of TC_INTERVAL.
TC messages MAY be generated in response to a change of contents, TC messages MAY be generated in response to a change of contents,
indicated by a change in ANSN. In this case a node MAY send a indicated by a change in ANSN. In this case a node MAY send a
complete TC message, and if so MAY re-start its TC message schedule. complete TC message, and if so MAY re-start its TC message schedule.
Alternatively a node MAY send only new content in its address blocks Alternatively a node MAY send only new content in its address blocks
(with appropriate associated TLVs) in which case it MUST include a (with appropriate associated TLVs) in which case it MUST include a
skipping to change at page 30, line 50 skipping to change at page 37, line 48
using an incomplete TC message. using an incomplete TC message.
When sending a TC message in response to a change of contents, a node When sending a TC message in response to a change of contents, a node
must respect a minimum interval of TC_MIN_INTERVAL between generated must respect a minimum interval of TC_MIN_INTERVAL between generated
TC messages. Sending an incomplete TC message MUST NOT cause the TC messages. Sending an incomplete TC message MUST NOT cause the
interval between complete TC messages to be increased, and thus a interval between complete TC messages to be increased, and thus a
node MUST NOT send an incomplete TC message if within TC_MIN_INTERVAL node MUST NOT send an incomplete TC message if within TC_MIN_INTERVAL
of the next scheduled complete TC message. of the next scheduled complete TC message.
The generation of TC messages, whether scheduled or triggered by a The generation of TC messages, whether scheduled or triggered by a
change of contents, and the forwarding of TC messages, MAY be change of contents MAY be jittered as described in [3]. The values
jittered as described in Appendix F. The values of MAXJITTER used of MAXJITTER used SHOULD be:
SHOULD be:
o TP_MAXJITTER for periodic TC message generation; o TP_MAXJITTER for periodic TC message generation;
o TT_MAXJITTER for triggered TC message generation; o TT_MAXJITTER for triggered TC message generation.
o TF_MAXJITTER for TC message forwarding;
TC messages are included in packets as specified in [3]. These TC messages are included in packets as specified in [1]. These
packets may contain other messages, including HELLO messages and TC packets may contain other messages, including HELLO messages and TC
messages with different originator addresses. TC messages are messages with different originator addresses. TC messages are
forwarded according to the specification in Section 7.4. forwarded according to the specification in Section 7.4.
13. TC Message Processing 12. TC Message Processing
When according to Section 7.3 a TC message is to be processed When according to Section 7.3 a TC message is to be processed
according to its type, this means that: according to its type, this means that:
o if the message does not contain a message TLV with Type == o if the message does not contain a message TLV with Type ==
INCOMPLETE, then processing according to Section 13.1 and then INCOMPLETE, then processing according to Section 12.1 and then
according to Section 13.2 is carried out; according to Section 12.2 is carried out;
o if the message contains a message TLV with Type == INCOMPLETE, o if the message contains a message TLV with Type == INCOMPLETE,
then only processing according to Section 13.1 is carried out. then only processing according to Section 12.1 is carried out.
For all processing purposes, "ANSN" is defined as being the value of For all processing purposes, "ANSN" is defined as being the value of
the message TLV with Type == CONT_SEQ_NUM in the TC message. If a TC the message TLV with Type == CONT_SEQ_NUM in the TC message. If a TC
message has no such TLV then it MUST NOT be processed. message has no such TLV then it MUST NOT be processed.
13.1. Initial TC Message Processing 12.1. Initial TC Message Processing
For the purposes of this section, note the following: For the purposes of this section, note the following:
o "validity time" is calculated from the VALIDITY_TIME message TLV o "validity time" is calculated from the VALIDITY_TIME message TLV
in the TC message according to the specification in Appendix E.2; in the TC message according to the specification in [2];
o "originator address" refers to the originator address in the TC o "originator address" refers to the originator address in the TC
message header; message header;
o "Local Address Block" refers to the Local Address Block (i.e. the
first address block) in the TC message;
o "sending address list" refers to the list of addresses in the
Local Address Block.
o "Advertised Address Block" refers to an Advertised Address Block
(i.e. the an address block other than the first address block) in
the TC message;
o comparisons of sequence numbers are carried out as specified in o comparisons of sequence numbers are carried out as specified in
Section 18. Section 17.
The TC message is processed as follows: The TC message is processed as follows:
1. the ANSN History Set is updated according to Section 13.1.1; if 1. the Advertising Remote Node Set is updated according to
the TC message is indicated as discarded in that processing then Section 12.1.1; if the TC message is indicated as discarded in
the following steps are not carried out; that processing then the following steps are not carried out;
2. the Topology Set is updated according to Section 13.1.2; 2. the Topology Set is updated according to Section 12.1.2;
3. the Attached Network Set is updated according to Section 12.1.3.
3. the Attached Network Set is updated according to Section 13.1.3. 12.1.1. Populating the Advertising Remote Node Set
13.1.1. Populating the ANSN History Set The node MUST update its Advertising Remote Node Set as follows:
The node MUST update its ANSN History Set as follows: 1. If there is an Advertising Remote Node Tuple with:
1. If there is an ANSN History Tuple with: * AR_orig_addr == originator address; AND
* AH_orig_addr == originator address; AND * AR_seq_number > ANSN
* AH_seq_number > ANSN
then the TC message MUST be discarded. then the TC message MUST be discarded.
2. Otherwise 2. Otherwise:
1. If there is no ANSN History Tuple such that: 1. If there is no Advertising Remote Node Tuple such that:
+ AH_orig_addr == originator address; + AR_orig_addr == originator address;
then create a new ANSN History Tuple with: then create an Advertising Remote Node Tuple with:
+ AH_orig_addr = originator address. + AR_orig_addr = originator address.
2. This ANSN History Tuple (existing or new) is then modified as 2. This Advertising Remote Node Tuple (existing or new, the
follows: "current tuple") is then modified as follows:
+ AH_seq_number = ANSN; + AR_seq_number = ANSN;
+ AH_time = current time + validity time. + AR_time = current time + validity time.
13.1.2. Populating the Topology Set + AR_iface_addr_list = sending address list
The node MUST update its Topology Set as follows: 3. for each other Advertising Remote Node Tuple (with a
different AR_orig_addr, the "other tuple") whose
AR_iface_addr_list contains any address in the
AR_iface_addr_list of the current tuple:
1. For each address, henceforth local address, in the first address 1. remove all Topology Tuples with T_orig_addr ==
block in the TC message: AR_orig_addr of the other tuple;
1. For each address, henceforth advertised address, in an 2. remove all Attached Network Tuples with AN_orig_addr ==
address block other than the first in the TC message, and AR_orig_addr of the other tuple;
which does not have an associated TLV with Type == GATEWAY:
3. remove the other tuple.
12.1.2. Populating the Topology Set
The node MUST update its Topology Set as follows:
1. For each address (henceforth advertised address) in an Advertised
Address Block which does not have an associated TLV with Type ==
GATEWAY:
1. If there is no Topology Tuple such that: 1. If there is no Topology Tuple such that:
- T_dest_iface_addr == advertised address; AND + T_dest_iface_addr == advertised address; AND
- T_last_iface_addr == local address + T_orig_addr == originator address
then create a new Topology Tuple with: then create a new Topology Tuple with:
- T_dest_iface_addr = advertised address; + T_dest_iface_addr = advertised address;
- T_last_iface_addr = local address. + T_orig_addr = originator address.
2. This Topology Tuple (existing or new) is then modified as 2. This Topology Tuple (existing or new) is then modified as
follows: follows:
- T_seq_number = ANSN; + T_seq_number = ANSN;
- T_time = current time + validity time. + T_time = current time + validity time.
13.1.3. Populating the Attached Network Set 12.1.3. Populating the Attached Network Set
The node MUST update its Attached Network Set as follows: The node MUST update its Attached Network Set as follows:
1. For each address, henceforth gateway address, in the first 1. For each address (henceforth network address) in an Advertised
address block in the TC message: Address Block which has an associated TLV with Type == GATEWAY:
1. For each address, henceforth network address, in an address
block other than the first in the TC message, and which has
an associated TLV with Type == GATEWAY:
1. If there is no Attached Network Tuple such that: 1. If there is no Attached Network Tuple such that:
- AN_net_addr == network address; AND + AN_net_addr == network address; AND
- AN_gw_iface_addr == gateway address + AN_orig_addr == originator address
then create a new Attached Network Tuple with: then create a new Attached Network Tuple with:
- AN_net_addr = network address; + AN_net_addr = network address;
- AN_gw_iface_addr = gateway address.
+ AN_orig_addr = originator address
2. This Attached Network Tuple (existing or new) is then 2. This Attached Network Tuple (existing or new) is then
modified as follows: modified as follows:
- AN_dist = the value of the associated GATEWAY TLV; + AN_dist = the value of the associated GATEWAY TLV;
- AN_seq_number = ANSN; + AN_seq_number = ANSN;
- AN_time = current time + validity time. + AN_time = current time + validity time.
13.2. Completing TC Message Processing 12.2. Completing TC Message Processing
The TC message is processed as follows: The TC message is processed as follows:
1. the Topology Set is updated according to Section 13.2.1; 1. the Topology Set is updated according to Section 12.2.1;
2. the Attached Network Set is updated according to Section 13.2.2. 2. the Attached Network Set is updated according to Section 12.2.2.
13.2.1. Purging the Topology Set 12.2.1. Purging the Topology Set
The Topology Set MUST be updated as follows: The Topology Set MUST be updated as follows:
1. for each address, henceforth local address, in the first address Any Topology Tuples with:
block of the TC message, all Topology Tuples with:
* T_last_iface_addr == local address; AND o T_orig_addr == originator address; AND
* T_seq_number < ANSN o T_seq_number < ANSN
MUST be removed. MUST be removed.
13.2.2. Purging the Attached Network Set 12.2.2. Purging the Attached Network Set
The Attached Network Set MUST be updated as follows: The Attached Network Set MUST be updated as follows:
1. for each address, henceforth local address, in the first address 1. Any Attached Network Tuples with:
block of the TC message, all Attached Network Tuples with:
* AN_gw_iface_addr == local address; AND * AN_orig_addr == originator address; AND
* AN_seq_number < ANSN * AN_seq_number < ANSN
MUST be removed. MUST be removed.
14. Populating the MPR Set 13. Selecting MPRs
Each node MUST select, from among its symmetric 1-hop neighbors, a Each node MUST select, from among its symmetric 1-hop neighbors, a
subset of nodes as MPRs. This subset MUST be selected such that a subset of nodes as MPRs. MPRs are used to flood control messages
message transmitted by the node, and retransmitted by all its MPRs, from a node into the network while reducing the number of
will be received by all of its symmetric strict 2-hop neighbors. retransmissions that will occur in a region. Thus, the concept of
MPR is an optimization of a classical flooding mechanism. MPRs MAY
Each node selects its MPR Set individually, utilizing the information also be used to reduce the shared topology information in the
in the Symmetric Neighbor Set, the 2-Hop Neighbor Set and the network. Consequently, while it is not essential that the set of
Neighborhood Address Association Set. Initially these sets will be MPRs is minimal, keeping the number of MPRs small ensures that the
empty, as will be the MPR Set. A node SHOULD recalculate its MPR Set overhead of OLSRv2 is kept at a minimum.
when a relevant change is made to the Symmetric Neighbor Set, the
2-Hop Neighbor Set or the Neighborhood Address Association Set.
More specifically, a node MUST calculate MPRs per interface, the
union of the MPR Sets of each interface make up the MPR Set for the
node. All OLSRv2 interfaces of nodes selected as MPRs with which the
node has a symmetric link MUST be added to the MPR Set. Also
symmetric 1-hop neighbor nodes with willingness WILL_NEVER (as
recorded in the Link Set) MUST NOT be considered as MPRs.
MPRs are used to flood control messages from a node into the network
while reducing the number of retransmissions that will occur in a
region. Thus, the concept of MPR is an optimization of a classical
flooding mechanism. While it is not essential that the MPR Set is
minimal, it is essential that all symmetric strict 2-hop neighbors
can be reached through the selected MPR nodes. A node MUST select an
MPR Set such that any strict 2-hop neighbor is "covered" by at least
one MPR node. A node MAY select additional MPRs beyond the minimum
set. Keeping the MPR Set small ensures that the overhead of OLSRv2
is kept at a minimum.
Appendix C contains an example heuristic for selecting MPRs.
15. Populating Derived Sets
The Relay Set and the Advertised Neighbor Set of OLSRv2 are denoted
derived sets, since updates to these sets are not directly a function
of message exchanges, but rather are derived from updates to other
sets, in particular the MPR Selector Set.
15.1. Populating the Relay Set
The Relay Set contains the set of OLSRv2 interface addresses of those
symmetric 1-hop neighbors for which a node is supposed to relay
broadcast traffic. This set MUST at least contain all addresses in
the MPR Selector Set (i.e. all MS_neighbor_iface_addr). This set MAY
contain additional symmetric 1-hop neighbor OLSRv2 interface
addresses.
15.2. Populating the Advertised Neighbor Set
The Advertised Neighbor Set contains the set of OLSRv2 interface
addresses of those 1-hop neighbors to which a node advertises a
symmetric link in TC messages. This set MUST at least contain all
addresses in the MPR Selector Set (i.e. all MS_neighbor_iface_addr).
This set MAY contain additional symmetric 1-hop neighbor OLSRv2
interface addresses.
Whenever an address is added to or removed from the Advertised
Neighbor Set, the ANSN MUST be incremented.
16. Routing Table Calculation
The Routing Set is updated when a change (an entry appearing or
disappearing, or changing between SYMMETRIC and LOST) is detected in:
o the Link Set, OR;
o the Neighbor Address Association Set, OR;
o the 2-Hop Neighbor Set, OR;
o the Topology Set, OR;
o the Attached Network Set. A node MUST select MPRs for each of its OLSRv2 interfaces, but then
forms the union of those sets as its single set of MPRs. This union
MUST include all symmetric 1-hop neighbors with willingness
WILL_ALWAYS. Only this overall set of MPRs is relevant and recorded,
the MPR relationship is one of nodes, not interfaces. Nodes MAY
select their MPRs by any process which satisfies the conditions which
follow. Nodes can freely interoperate whether they use the same or
different MPR selection algorithms.
Note that some changes to these sets do not necessitate a change to For each OLSRv2 interface a node MUST select a set of MPRs which have
the Routing Set, in particular changes to the Link Set which do not the property that none of them have willingness WILL_NEVER, and that
involve Link Tuples with L_STATUS == SYMMETRIC (either before or if the node successfully sends a message on that OLSRv2 interface,
after the change), and similar changes to the Neighbor Address and that message is then successfully forwarded by all of the
Association Set. A node MAY avoid updating the Routing Set in such selected MPRs, that all symmetric strict 2-hop neighbors of the node
cases. by that OLSRv2 interface will receive that message on a symmetric
link.
Updates to the Routing Set do not generate or trigger any messages to Note that it is always possible to select a valid set of MPRs, the
be transmitted. The state of the Routing Set SHOULD, however, be set of all symmetric 1-hop neighbors of a node which do not have
reflected in the IP routing table by adding and removing entries from willingness WILL_NEVER is a (maximal) valid set of MPRs. A node
the routing table as appropriate. SHOULD NOT select a symmetric 1-hop neighbor with willingness not
equal to WILL_ALWAYS as an MPR if there are no symmetric strict 2-hop
neighbors with a symmetric link to that symmetric 1-hop neighbor.
Thus a node with no symmetric 1-hop neighbors with willingness
WILL_ALWAYS and no symmetric strict 2-hop neighbors SHOULD NOT select
any MPRs.
To construct the Routing Set of node X, a shortest path algorithm is A node MAY select its MPRs for each OLSRv2 interface independently,
run on the directed graph containing or it MAY coordinate its MPR selections across its OLSRv2 interfaces,
as long as the required condition is satisfied for each OLSRv2
interface. Each node MAY select its MPRs independently from the MPR
selection by other nodes, or it MAY, for example, give preference to
nodes that either are, or are not, already selected as MPRs by other
nodes.
o the arcs X -> Y where there exists a Link Tuple with Y in the The set of MPRs for each OLSRv2 interface can be selected using
L_neighbor_iface_addr_list and L_STATUS == SYMMETRIC (i.e. Y is a information from the Link Set and 2-Hop Set of that OLSRv2 interface,
symmetric 1-hop neighbor of X), AND; and the Neighbor Set of the node (specifically the N_willingness
elements). The selection of MPRs (overall, not per OLSRv2 interface)
is recorded in the Neighbor Set of the node (using the N_mpr
elements). A selected MPR MUST be in the node's symmetric 1-hop
neighborhood (i.e. the corresponding N_symmetric == true) and MUST
not have the corresponding N_willingness == WILL_NEVER.
o the arcs Y -> Z where Y is added as above and the Link Tuple with A node MUST recalculate its MPRs whenever the currently selected set
Y in its L_neighbor_iface_addr_list has L_willingness not equal to of MPRs does not still satisfy the required conditions. It MAY
WILL_NEVER, and there exists a 2-Hop Neighbor Tuple with Y as recalculate its MPRs if the current set of MPRs is still valid, but
N2_neighbor_iface_addr and Z as N2_2hop_iface_addr (i.e. Z is a could be more efficient. It is sufficient to recalculate a node's
symmetric 2-hop neighbor of Z through Y, which does not have MPRs when there is a change to any of the node's Link Sets affecting
willingness WILL_NEVER), AND; the symmetry of any link (addition or removal of a Link Tuple with
L_status == SYMMETRIC, or change of any L_status to or from
SYMMETRIC), any change to any of the node's 2-Hop Sets, or a change
of the N_willingness (to or from WILL_NEVER or to WILL_ALWAYS is
sufficient) of any Neighbor Tuple with N_symmetric == true.
o the arcs U -> V, where there exists a Topology Tuple with U as An algorithm that creates a set of MPRs that satisfies the required
T_last_iface_addr and V as T_dest_iface_addr (i.e. this is an conditions is given in Appendix B.
advertised link in the network).
The graph is complemented with: 14. Populating Derived Sets
o arcs Y -> W where there exists a Link Tuple with Y in its The Relay Sets and the Advertised Neighbor Set of a node are denoted
L_neighbor_iface_addr_list and L_STATUS == SYMMETRIC and a derived sets, since updates to these sets are not directly a function
Neighborhood Address Association Tuple with Y and W both contained of message exchanges, but rather are derived from updates to other
in its NA_neighbor_iface_addr_list (i.e. Y and W are both sets, in particular to the MPR selector status of other nodes
addresses of the same symmetric 1-hop neighbor), AND; recorded in the Neighbor Set.
o arcs U -> T where there exists an Attached Network Tuple with U as 14.1. Populating the Relay Set
AN_net_addr and T as AN_gw_iface_addr (i.e. U is a gateway to
network T).
The following procedure is given as an example for calculating the The Relay Set for an OLSRv2 interface contains the set of OLSRv2
Routing Set using a variation of Dijkstra's algorithm. Thus: interface addresses of those symmetric 1-hop neighbors for which this
OLSRv2 interface is to relay broadcast traffic. It MUST contain only
addresses of OLSRv2 interfaces with which this OLSRv2 interface has a
symmetric link. It MUST include all such addresses of all such
OLSRv2 interfaces of nodes which are MPR selectors of this node. The
Relay Set for an OLSRv2 interface of this node is thus created by:
1. All Routing Tuples are removed. 1. For each Link Tuple in the Link Set for this OLSRv2 interface
with L_status == SYMMETRIC, and the corresponding Neighbor Tuple
with N_neighbor_iface_addr_list containing
L_neighbor_iface_addr_list:
2. For each Link Tuple with L_STATUS == SYMMETRIC, and for each 1. All addresses from L_neighbor_iface_addr_list MUST be
address (henceforth neighbor address) in that Link Tuple's included in the Relay Set of this OLSRv2 interface if
L_neighbor_iface_addr_list, a new Routing Tuple is added with: N_mpr_selector == true, and otherwise MAY be so included.
* R_dest_addr = neighbor address; 14.2. Populating the Advertised Neighbor Set
* R_next_iface_addr = neighbor address; The Advertised Neighbor Set of a node contains all interface
addresses of those symmetric 1-hop neighbors to which the node
advertises a link in its TC messages. This set MUST at least contain
all addresses in all MPR selector of this node. The Advertised
Neighbor Set for this node is thus created by:
* R_dist = 1; 1. For each Neighbor Tuple with N_symmetric == true:
* R_local_iface_addr = neighbor address. 1. All addresses from N_neighbor_iface_addr_list MUST be
included in the Advertised Neighbor Set if N_mpr_selector ==
true, and otherwise MAY be so included.
3. For each Neighbor Address Association Tuple, for which two Whenever address(es) are added to or removed from the Advertised
addresses A1 and A2 are in NA_neighbor_iface_addr_list where: Neighbor Set, its ANSN MUST be incremented.
* there is a Routing Tuple with: 15. Routing Set Calculation
+ R_dest_addr == A1 The Routing Set of a node is populated with Routing Tuples that
represent paths from that node to all destinations in the network.
These paths are calculated based on the Network Topology Graph, which
is constructed from information in the information repositories,
obtained via HELLO and TC message exchange.
* and there is no Routing Tuple with: 15.1. Network Topology Graph
+ R_dest_addr == A2 The Network Topology Graph is formed from information taken from the
node's Link Sets, Neighbor Set, Topology Set and Attached Network
Set. The Network Topology Graph SHOULD also use information taken
from the node's 2-Hop Sets. The Network Topology Graph forms that
node's topological view of the network in form of a directed graph,
containing the following arcs:
then a Routing Tuple is added with: o Local symmetric links - all arcs X -> Y such that:
* R_dest_addr = A2; * X is an address in the I_local_iface_addr_list of a Local
Interface Tuple of this node, AND;
* R_next_iface_addr = R_next_iface_addr of the Routing Tuple in * Y is an address in the L_neighbor_iface_addr_list of a Link
which R_dest_addr == A1; Tuple in the corresponding (to the OLSRv2 interface of that
I_local_iface_addr_list) Link Set which has L_status ==
SYMMETRIC.
* R_dist = 1; o 2-hop symmetric links - all arcs Y -> Z such that:
* R_local_iface_addr = R_local_iface_addr of the Routing Tuple
in which R_dest_addr == A1.
4. The following procedure, which adds Routing Tuples for * Y is an address in the L_neighbor_iface_addr_list of a Link
destination nodes h+1 hops away, MUST be executed for each value Tuple, in any of the node's Link Sets, which has L_status ==
of h, starting with h=2 and incrementing by 1 for each iteration. SYMMETRIC, AND;
The execution MUST stop if no new Routing Tuples are added in an
iteration.
1. For each Topology Tuple, if * the Neighbor Tuple with Y in its N_neighbor_iface_addr_list has
N_willingness not equal to WILL_NEVER, AND;
+ T_dest_iface_addr is not equal to R_dest_addr of any * Z is the N2_2hop_iface_addr of a 2-Hop Tuple in the 2-Hop Set
Routing Tuple, AND; corresponding to the OLSRv2 interface of the chosen Link Set.
+ T_last_iface_addr is equal to R_dest_addr of a Routing o Advertised symmetric links - all arcs U -> V such that there
Tuple whose R_dist == h; exists a Topology Tuple and a corresponding Advertising Remote
Node Tuple (i.e. with AR_orig_addr == T_orig_addr) with:
then a new Routing Tuple MUST be added, with: * U is in the AR_iface_addr_list of the Advertising Remote Node
Tuple, AND;
+ R_dest_addr = T_dest_iface_addr; * V is the T_dest_iface_addr of the Topology Tuple.
+ R_next_iface_addr = R_next_iface_addr of the Routing Tuple o Symmetric 1-hop neighbor addresses - all arcs Y -> W such that:
whose R_dest_addr == T_last_iface_addr;
+ R_dist = h+1; * Y is, and W is not, an address in the
L_neighbor_iface_addr_list of a Link Tuple, in any of the
node's Link Sets, which has L_status == SYMMETRIC, AND;
+ R_local_iface_addr = R_local_iface_addr of the Routing * W and Y are included in the same N_neighbor_iface_addr_list
Tuple whose R_dest_addr == T_last_iface_addr. (i.e. the one in the Neighbor Tuple whose
N_neighbor_iface_addr_list contains the
L_neighbor_iface_addr_list that includes Y).
Several Topology Tuples may be used to select a next hop o Attached network addresses - all arcs U -> T such that there
R_next_iface_addr for reaching the address R_dest_addr. When exists an Attached Network Tuple and a corresponding Advertising
h == 1, ties should be broken such that nodes with highest Remote Node Tuple (i.e. with AR_orig_addr == AN_orig_addr) with:
willingness are preferred, and between nodes of equal
willingness, MPR selectors are preferred over non-MPR
selectors.
2. After the above iteration has completed, if h == 1, for each * U is in the AR_iface_addr_list of the Advertising Remote Node
2-Hop Neighbor Tuple where: Tuple, AND;
+ N2_2hop_iface_addr is not equal to R_dest_addr of any * T is the AN_net_addr of the Attached Network Tuple.
Routing Tuple, AND;
+ N2_neighbor_iface_addr has a willingness (i.e. the All links in the first three cases above have a hop count of one, the
L_willingness of the Link Tuple whose symmetric 1-hop neighbor addresses have a hop count of zero, and the
L_neighbor_iface_addr_list contains attached network addresses have a hop count given by the appropriate
N2_neighbor_iface_addr) which is not equal to WILL_NEVER; value of AN_dist.
a Routing Tuple is added with: 15.2. Populating the Routing Set
+ R_dest_addr = N2_2hop_iface_addr of the 2-Hop Neighbor The Routing Set MUST contain the shortest paths for all destinations
Tuple; from all local OLSRv2 interfaces using the Network Topology Graph.
This calculation MAY use any algorithm, including any means of
choosing between paths of equal length.
+ R_next_iface_addr = R_next_iface_addr of the Routing Tuple Using the notation of Section 15.1, each path will have as its first
in which R_dest_addr == N2_neighbor_iface_addr; arc a local symmetric link X -> Y. There will be a path for each
terminating Y, Z, V, W and T which can be connected to local OLSRv2
interface address X using the indicated arcs. The corresponding
Routing Tuple for this path will have:
+ R_dist = 2; o R_dest_addr = the terminating Y, Z, V, W or T;
+ R_local_iface_addr = R_local_iface_addr of the Routing o R_next_iface_addr = the first arc's Y;
Tuple in which R_dest_addr == N2_neighbor_iface_addr.
5. For each Attached Network Tuple, if o R_dist = the total hop count of the path;
* AN_net_addr is not equal to R_dest_addr of any Routing Tuple, o R_local_iface_addr = the first arc's X.
AND;
* AN_gw_iface_addr is equal to R_dest_addr of a Routing Tuple; An example algorithm for calculating the Routing Set of a node is
given in Appendix C.
then a new Routing Tuple MUST be added, with: 15.3. Routing Set Updates
* R_dest_addr = AN_net_addr; The Routing Set MUST be updated when changes in the Neighborhood
Information Base or the Topology Information Base indicate a change
of the known symmetric links and/or attached networks in the MANET.
It is sufficient to consider only changes which affect at least one
of:
* R_next_iface_addr = R_next_iface_addr of the Routing Tuple o The Link Set of any OLSRv2 interface, and to consider only Link
whose R_dest_addr == AN_gw_iface_addr; Tuples which have, or just had, L_status == SYMMETRIC (including
removal of such Link Tuples).
* R_dist = (R_dist of the Routing Tuple whose R_dest_addr == o The Neighbor Set of the node, and to consider only Neighbor Tuples
AN_gw_iface_addr) + AN_dist; that have, or just had, N_symmetric == true.
* R_local_iface_addr = R_local_iface_addr of the Routing Tuple o The 2-Hop Set of any OLSRv2 interface.
whose R_dest_addr == AN_gw_iface_addr.
If more than one Attached Network Tuple has the same AN_net_addr, o The Topology Set of the node.
then more than one Routing Tuple MUST NOT be added, and the added
Routing Tuple MUST have minimum R_dist.
17. Proposed Values for Constants o The Attached Network Set of the node.
This section list the values for the constants used in the Updates to the Routing Set do not generate or trigger any messages to
description of the protocol. These proposed values are appropriate be transmitted. The state of the Routing Set SHOULD, however, be
to the case where all TC messages are sent with the same hop limit reflected in the IP routing table by adding and removing entries from
(usually 255). the IP routing table as appropriate.
17.1. Neighborhood Discovery Constants 16. Proposed Values for Parameters and Constants
The constants HELLO_INTERVAL, REFRESH_INTERVAL, HELLO_MIN_INTERVAL, OLSRv2 uses all parameters and constants defined in [4] and
H_HOLD_TIME, L_HOLD_TIME, N_HOLD_TIME, HP_MAXJITTER, HT_MAXJITTER and additional parameters and constants defined in this document. All
C are used as in [4]. but one (RX_HOLD_TIME) of these additional parameters are node
parameters as defined in [4]. These proposed values of the
additional parameters are appropriate to the case where all
parameters (including those defined in [4]) have a single value.
Proposed values for parameters defined in [4] are given in that
document.
17.2. Message Intervals 16.1. Message Interval Parameters
o TC_INTERVAL = 5 seconds o TC_INTERVAL = 5 seconds
o TC_MIN_INTERVAL = TC_INTERVAL/4 o TC_MIN_INTERVAL = TC_INTERVAL/4
17.3. Holding Times 16.2. Advertised Information Validity Time Parameters
o T_HOLD_TIME = 3 x TC_INTERVAL o T_HOLD_TIME = 3 x TC_INTERVAL
o A_HOLD_TIME = T_HOLD_TIME o A_HOLD_TIME = T_HOLD_TIME
o P_HOLD_TIME = 30 seconds 16.3. Received Message Validity Time Parameters
o RX_HOLD_TIME = 30 seconds o RX_HOLD_TIME = 30 seconds
o P_HOLD_TIME = 30 seconds
o F_HOLD_TIME = 30 seconds o F_HOLD_TIME = 30 seconds
17.4. Jitter Times 16.4. Jitter Time Parameters
o TP_MAXJITTER = HP_MAXJITTER o TP_MAXJITTER = HP_MAXJITTER
o TT_MAXJITTER = HT_MAXJITTER o TT_MAXJITTER = HT_MAXJITTER
o TF_MAXJITTER = TT_MAXJITTER o F_MAXJITTER = TT_MAXJITTER
17.5. Willingness 16.5. Hop Limit Parameter
o TC_HOP_LIMIT = 255
16.6. Willingness Parameter and Constants
o WILLINGNESS = WILL_DEFAULT
o WILL_NEVER = 0 o WILL_NEVER = 0
o WILL_DEFAULT = 3 o WILL_DEFAULT = 3
o WILL_ALWAYS = 7 o WILL_ALWAYS = 7
18. Sequence Numbers 17. Sequence Numbers
Sequence numbers are used in OLSRv2 with the purpose of discarding Sequence numbers are used in OLSRv2 with the purpose of discarding
"old" information, i.e. messages received out of order. However with "old" information, i.e. messages received out of order. However with
a limited number of bits for representing sequence numbers, wrap- a limited number of bits for representing sequence numbers, wrap-
around (that the sequence number is incremented from the maximum around (that the sequence number is incremented from the maximum
possible value to zero) will occur. To prevent this from interfering possible value to zero) will occur. To prevent this from interfering
with the operation of OLSRv2, the following MUST be observed when with the operation of OLSRv2, the following MUST be observed when
determining the ordering of sequence numbers. determining the ordering of sequence numbers.
The term MAXVALUE designates in the following one more than the The term MAXVALUE designates in the following one more than the
skipping to change at page 44, line 5 skipping to change at page 52, line 5
o S2 > S1 AND S2 - S1 > MAXVALUE/2 o S2 > S1 AND S2 - S1 > MAXVALUE/2
When sequence numbers S1 and S2 differ by MAXVALUE/2 their ordering When sequence numbers S1 and S2 differ by MAXVALUE/2 their ordering
cannot be determined. In this case, which should not occur, either cannot be determined. In this case, which should not occur, either
ordering may be assumed. ordering may be assumed.
Thus when comparing two messages, it is possible - even in the Thus when comparing two messages, it is possible - even in the
presence of wrap-around - to determine which message contains the presence of wrap-around - to determine which message contains the
most recent information. most recent information.
19. IANA Considerations 18. IANA Considerations
19.1. Message Types 18.1. Message Types
OLSRv2 defines one message type, which must be allocated from the OLSRv2 defines one message type, which must be allocated from the
"Assigned Message Types" repository of [3]. "Assigned Message Types" repository of [1].
+--------------------+-------+--------------------------------------+ +------+-------+--------------------------------------+
| Mnemonic | Value | Description | | Name | Value | Description |
+--------------------+-------+--------------------------------------+ +------+-------+--------------------------------------+
| TC | TBD | Topology Control (global signaling) | | TC | TBD | Topology Control (global signaling) |
+--------------------+-------+--------------------------------------+ +------+-------+--------------------------------------+
Table 5 Table 5
19.2. TLV Types 18.2. TLV Types
OLSRv2 defines three message TLV types, which must be allocated from OLSRv2 defines three message TLV types, which must be allocated from
the "Assigned message TLV Types" repository of [3]. the "Assigned message TLV Types" repository of [1].
+--------------------+-------+--------------------------------------+ +--------------+------+---------+-----------------------------------+
| Mnemonic | Value | Description | | Name | Type | Subtype | Description |
+--------------------+-------+--------------------------------------+ +--------------+------+---------+-----------------------------------+
| WILLINGNESS | TBD | Specifies the originating node's | | WILLINGNESS | TBD | 0 | Specifies the originating node's |
| | | willingness to act as a relay and to | | | | | willingness to act as a relay and |
| | | partake in network formation | | | | | to partake in network formation |
| | | | | | | | |
| CONT_SEQ_NUM | TBD | Specifies a content sequence number | | | TBD | 1-255 | RESERVED |
| | | for this message | | | | | |
| | | | | CONT_SEQ_NUM | TBD | 0 | Specifies a content sequence |
| INCOMPLETE | TBD | Specifies that this message is | | | | | number for this message |
| | | incomplete | | | | | |
+--------------------+-------+--------------------------------------+ | | TBD | 1-255 | RESERVED |
| | | | |
| INCOMPLETE | TBD | 0 | Specifies that this message is |
| | | | incomplete |
| | | | |
| | TBD | 1-255 | RESERVED |
+--------------+------+---------+-----------------------------------+
Table 6 Table 6
Subtypes indicated as RESERVED may be allocated by standards action,
as specified in [7].
OLSRv2 defines two Address Block TLV types, which must be allocated OLSRv2 defines two Address Block TLV types, which must be allocated
from the "Assigned address block TLV Types" repository of [3]. from the "Assigned address block TLV Types" repository of [1].
+--------------------+-------+--------------------------------------+ +---------+------+---------+----------------------------------------+
| Mnemonic | Value | Description | | Name | Type | Subtype | Description |
+--------------------+-------+--------------------------------------+ +---------+------+---------+----------------------------------------+
| MPR | TBD | Specifies that a given address is | | MPR | TBD | 0 | Specifies that a given address is of a |
| | | selected as MPR | | | | | node selected as an MPR |
| | | | | | | | |
| GATEWAY | TBD | Specifies that a given address is | | | TBD | 1-255 | RESERVED |
| | | reached via a gateway on the | | | | | |
| | | originating node | | GATEWAY | TBD | 0 | Specifies that a given address is |
+--------------------+-------+--------------------------------------+ | | | | reached via a gateway on the |
| | | | originating node |
| | | | |
| | TBD | 1-255 | RESERVED |
+---------+------+---------+----------------------------------------+
Table 7 Table 7
20. References Subtypes indicated as RESERVED may be allocated by standards action,
as specified in [7].
20.1. Normative References
[1] Clausen, T. and P. Jacquet, "The Optimized Link State Routing 19. References
Protocol", RFC 3626, October 2003.
[2] Bradner, S., "Key words for use in RFCs to Indicate Requirement 19.1. Normative References
Levels", RFC 2119, BCP 14, March 1997.
[3] Clausen, T., Dean, J., Dearlove, C., and C. Adjih, "Generalized [1] Clausen, T., Dean, J., Dearlove, C., and C. Adjih, "Generalized
MANET Packet/Message Format", work in MANET Packet/Message Format", work in
progress draft-ietf-manet-packetbb-03.txt, January 2007. progress draft-ietf-manet-packetbb-08.txt, July 2007.
[2] Clausen, T. and C. Dearlove, "Representing multi-value time in
MANETs", Work In Progress draft-ietf-manet-timetlv-01.txt,
June 2007.
[3] Clausen, T., Dearlove, C., and B. Adamson, "Jitter
considerations in MANETs", Work In
Progress draft-ietf-manet-jitter-01.txt, June 2007.
[4] Clausen, T., Dean, J., and C. Dearlove, "MANET Neighborhood [4] Clausen, T., Dean, J., and C. Dearlove, "MANET Neighborhood
Discovery Protocol (NHDP)", work in Discovery Protocol (NHDP)", work in
progress draft-ietf-manet-nhdp-01.txt, February 2007. progress draft-ietf-manet-nhdp-04.txt, June 2007.
20.2. Informative References [5] Bradner, S., "Key words for use in RFCs to Indicate Requirement
Levels", RFC 2119, BCP 14, March 1997.
[5] Atkins, D., Stallings, W., and P. Zimmermann, "PGP Message [6] Chakeres, I., "Internet Assigned Numbers Authority (IANA)
Allocations for the Mobile Ad hoc Networks (MANET) Working
Group", Work In Progress draft-ietf-manet-iana-05.txt,
June 2007.
[7] Narten, T. and H. Alvestrand, "Guidelines for Writing an IANA
Considerations Section in RFCs", October 1998.
19.2. Informative References
[8] Clausen, T. and P. Jacquet, "The Optimized Link State Routing
Protocol", RFC 3626, October 2003.
[9] Atkins, D., Stallings, W., and P. Zimmermann, "PGP Message
Exchange Formats", RFC 1991, August 1996. Exchange Formats", RFC 1991, August 1996.
[6] ETSI, "ETSI STC-RES10 Committee. Radio equipment and systems: [10] ETSI, "ETSI STC-RES10 Committee. Radio equipment and systems:
HIPERLAN type 1, functional specifications ETS 300-652", HIPERLAN type 1, functional specifications ETS 300-652",
June 1996. June 1996.
[7] Jacquet, P., Minet, P., Muhlethaler, P., and N. Rivierre, [11] Jacquet, P., Minet, P., Muhlethaler, P., and N. Rivierre,
"Increasing reliability in cable free radio LANs: Low level "Increasing reliability in cable free radio LANs: Low level
forwarding in HIPERLAN.", 1996. forwarding in HIPERLAN.", 1996.
[8] Qayyum, A., Viennot, L., and A. Laouiti, "Multipoint relaying: [12] Qayyum, A., Viennot, L., and A. Laouiti, "Multipoint relaying:
An efficient technique for flooding in mobile wireless An efficient technique for flooding in mobile wireless
networks.", 2001. networks.", 2001.
[13] Macker, J. and S. Corson, "Mobile Ad hoc Networking (MANET):
Routing Protocol Performance Issues and Evaluation
Considerations", RFC 2501, January 1999.
Appendix A. Node Configuration Appendix A. Node Configuration
OLSRv2 does not make any assumption about node addresses, other than OLSRv2 does not make any assumption about node addresses, other than
that each node is assumed to have at least one unique and routable IP that each node is assumed to have at least one unique and routable IP
address for each interface that it has which participates in the address for each interface that it has which participates in the
MANET. MANET.
When applicable, a recommended way of connecting an OLSRv2 network to When applicable, a recommended way of connecting an OLSRv2 network to
an existing IP routing domain is to assign an IP prefix (under the an existing IP routing domain is to assign an IP prefix (under the
authority of the nodes/gateways connecting the MANET with the routing authority of the nodes/gateways connecting the MANET with the routing
domain) exclusively to the OLSRv2 area, and to configure the gateways domain) exclusively to the OLSRv2 area, and to configure the gateways
statically to advertise routes to that IP sequence to nodes in the statically to advertise routes to that IP sequence to nodes in the
existing routing domain. existing routing domain.
Appendix B. Protocol and Port Number Appendix B. Example Algorithm for Calculating MPRs
Packets in OLSRv2 are communicated using UDP. Port 698 has been The following specifies an algorithm which MAY be used to select
assigned by IANA for exclusive usage by the OLSR (v1 and v2) MPRs. MPRs are calculated per OLSRv2 interface, but then a single
protocol. set of MPRs is formed from the union of the MPRs for all OLSRv2
interfaces. A node's MPRs are recorded using the element N_mpr in
Neighbor Tuples.
Appendix C. Example Heuristic for Calculating MPRs If using this algorithm then the following steps MUST be executed in
order for a node to select its MPRs:
The following specifies a proposed heuristic for selection of MPRs. 1. Set N_mpr = false in all Neighbor Tuples;
In graph theory terms, MPR computation is a "set cover" problem, 2. For each Neighbor Tuple with N_symmetric == true and
which is a difficult optimization problem, but for which an easy and N_willingness == WILL_ALWAYS, set N_mpr = true;
efficient heuristics exist: the so-called "Greedy Heuristic", a
variant of which is described here. In simple terms, MPR computation
constructs an MPR Set that enables a node to reach any symmetric
2-hop neighbors by relaying through an MPR node.
There are several peripheral issues that the algorithm needs to 3. For each OLSRv2 interface of the node, use the algorithm in
address. The first one is that some nodes have some willingness Appendix B.2. Note that this sets N_mpr = true for some Neighbor
WILL_NEVER. The second one is that some nodes may have several Tuples, these nodes are already selected as MPRs when using the
interfaces. algorithm for following OLSRv2 interfaces.
The algorithm hence can be summarized by: 4. OPTIONALLY, consider each selected MPR in turn, and if the set of
selected MPRs without that node still satisfies the necessary
conditions, for all OLSRv2 interfaces, then that node MAY be
removed from the set of MPRs. This process MAY be repeated until
no MPRs are removed. Nodes MAY be considered in order of
increasing N_willingness.
o All 1-hop neighbor nodes with willingness equal to WILL_NEVER MUST Symmetric 1-hop neighbor nodes with N_willingness == WILL_NEVER MUST
ignored in the following algorithm: they are not considered as NOT be selected as MPRs, and MUST be ignored in the following
1-hop neighbors (hence not used as MPRs). algorithm, as MUST be symmetric 2-hop neighbor nodes which are also
symmetric 1-hop neighbor nodes (i.e. when considering 2-Hop Tuples,
ignore any 2-Hop Tuples whose N2_2hop_iface_addr is in the
N_neighbor_iface_addr_list of any Neighbor Tuple, or whose
N2_neighbor_iface_addr_list is included in the
N_neighbor_iface_addr_list of any Neighbor Tuple with N_willingness
== WILL_NEVER).
o Because link sensing is performed by interface, the local network B.1. Terminology
topology is best described in terms of links: hence the algorithm
is considering 1-hop neighbor OLSRv2 interfaces, and 2-hop
neighbor OLSRv2 interfaces (and their addresses). Additionally,
asymmetric links are ignored. This is reflected in the
definitions below.
o MPR computation is performed on each interface of the node: on The following terminology will be used when selecting MPRs for the
each interface I, the node MUST select some neighbor interfaces, OLSRv2 interface I:
so that all 2-hop neighbor interfaces are reached.
From now on, MPR calculation will be described for one interface I on N(I) - The set of symmetric 1-hop neighbors which have a symmetric
the node, and the following terminology will be used in describing link to I.
the heuristics:
neighbor interface (of I) - An OLSRv2 interface of a 1-hop neighbor N2(I) - The set of addresses of interfaces of a node with a
to which there exist a symmetric link using interface I. symmetric link to a node in N(I) (i.e. the set of
N2_2hop_iface_addr in 2-Hop Tuples in the 2-Hop Set for OLSRv2
interface I).
N - the set of such neighbor interfaces Connected to I via Y - An address A in D2(I) is connected to I via a
node Y in N(I) if A is an address of an interface of a symmetric
1-hop neighbor of Y (i.e. A is the N2_2hop_iface_addr in a 2-Hop
Tuple in the 2-Hop Set for OLSRv2 interface I, and whose
N2_neighbor_iface_addr_list is contained in the set of interface
addresses of Y).
2-hop neighbor interface (of I) An interface of a symmetric strict D(Y, I) - For a node Y in N(I), the number of addresses in D2(I)
2-hop neighbor which can be reached from a neighbor interface for which are connected to I via Y.
I.
N2 - the set of such 2-hop neighbor interfaces R(Y, I): - For a node Y in N(I), the number of addresses in D2(I)
which are connected to I via Y, but are not connected to I via any
node which has already been selected as an MPR.
D(y): - the degree of a 1-hop neighbor interface y (where y is a B.2. MPR Selection Algorithm for each OLSRv2 Interface
member of N), is defined as the number of symmetric neighbor
interfaces of node y which are in N2
MPR Set - the set of the neighbor interfaces selected as MPRs. When selecting MPRs for the OLSRv2 interface I:
The proposed heuristic selects iteratively some interfaces from N as 1. For each address A in N2(I) for which there is only one node Y in
MPRs in order to cover 2-hop neighbor interfaces from N2, as follows: N(I) such that A is connected to I via Y, select that node Y as
an MPR (i.e. set N_mpr = true in the Neighbor Tuple corresponding
to Y).
1. Start with an MPR Set made of all members of N with L_willingness 2. While there exists any node Y in N(I) with R(Y, I) > 0:
equal to WILL_ALWAYS
2. Calculate D(y), where y is a member of N, for all interfaces in 1. Select a node Y in N(I) with R(Y, I) > 0 in the following
N. order of priority:
3. Add to the MPR Set those interfaces in N, which are the *only* + greatest N_willingness in the Neighbor Tuple corresponding
nodes to provide reachability to an interface in N2. For to Y, THEN;
example, if interface B in N2 can be reached only through a
symmetric link to interface A in N, then add interface B to the
MPR Set. Remove the interfaces from N2 which are now covered by a
interface in the MPR Set.
4. While there exist interfaces in N2 which are not covered by at + greatest R(Y, I), THEN;
least one interface in the MPR Set:
1. For each interface in N, calculate the reachability, i.e., + greatest D(Y, I), THEN;
the number of interfaces in N2 which are not yet covered by
at least one node in the MPR Set, and which are reachable
through this neighbor interface;
2. Select as an MPR the interface with highest L_willingness + any choice.
among the interfaces in N with non-zero reachability. In
case of multiple choice select the interface which provides
reachability to the maximum number of interfaces in N2. In
case of multiple interfaces providing the same amount of
reachability, select the interface as MPR whose D(y) is
greater. Remove the interfaces from N2 which are now covered
by an interface in the MPR Set.
Other algorithms, as well as improvements over this algorithm, are 2. Select Y as an MPR (i.e. set N_mpr = true in the Neighbor
possible. For example: Tuple corresponding to Y).
o Assume that in a multiple interface scenario there exists more Appendix C. Example Algorithm for Calculating the Routing Set
than one link between nodes 'a' and 'b'. If node 'a' has selected
node 'b' as MPR for one of its interfaces, then node 'b' can be
selected as MPR with minimal performance loss by any other
interfaces on node 'a'.
o In a multiple interface scenario MPRs are selected for each The following procedure is given as an example for calculating the
interface of the selecting node, providing full coverage of all Routing Set using a variation of Dijkstra's algorithm. First all
2-hop nodes accessible through that interface. The overall MPR Routing Tuples are removed, and then the procedures in the following
Set is then the union of these sets. These sets do not however sections are applied in turn.
have to be selected independently, if a node is selected as an MPR
for one interface it may be automatically added to the MPR C.1. Add Local Symmetric Links
selection for other interfaces.
1. For each Local Interface Tuple in the Local Interface Set:
1. For each address A in I_local_iface_addr_list:
1. For each Link Tuple in the Link Set for this local
interface, with L_status == SYMMETRIC:
1. For each address, B, in that Link Tuple's
L_neighbor_iface_addr_list, add a new Routing Tuple
with:
o R_dest_addr = B;
o R_next_iface_addr = B;
o R_dist = 1;
o R_local_iface_addr = A.
2. For each Neighbor Tuple, for which there is an address B in
N_neighbor_iface_addr_list, for which there is a Routing Tuple
(the "previous Routing Tuple") with R_dest_addr == B:
1. For each address C in N_neighbor_iface_addr_list for which
there is no Routing Tuple with R_dest_addr == C, add a
Routing Tuple with:
+ R_dest_addr = C;
+ R_next_iface_addr = B;
+ R_dist = 1;
+ R_local_iface_addr = R_local_iface_addr of the previous
Routing Tuple.
C.2. Add Remote Symmetric Links
The following procedure, which adds Routing Tuples for destination
nodes h+1 hops away, MUST be executed for each value of h, starting
with h = 1 and incrementing by 1 for each iteration. The execution
MUST stop if no new Routing Tuples are added in an iteration.
1. For each Topology Tuple, if:
* T_dest_iface_addr is not equal to R_dest_addr of any Routing
Tuple, AND;
* for the Advertising Remote Node Tuple with AR_orig_addr ==
T_orig_addr, there is an address in the AR_iface_addr_list
which is equal to the R_dest_addr of a Routing Tuple (the
"previous Routing Tuple") whose R_dist == h
then add a new Routing Tuple, with:
* R_dest_addr = T_dest_iface_addr;
* R_next_iface_addr = R_next_iface_addr of the previous Routing
Tuple;
* R_dist = h+1;
* R_local_iface_addr = R_local_iface_addr of the previous
Routing Tuple.
More than one Topology Tuple may be usable to select the next hop
R_next_iface_addr for reaching the address R_dest_addr. When h
== 1, ties should be broken such that nodes with greater
willingness are preferred, and between nodes of equal
willingness, MPR selectors are preferred over non-MPR selectors.
2. After the above iteration has completed, if h == 1, for each
2-Hop Neighbor Tuple where:
* N2_2hop_iface_addr is not equal to R_dest_addr of any Routing
Tuple, AND;
* The Neighbor Tuple whose N_neighbor_iface_addr_list contains
N2_neighbor_iface_addr_list has N_willingness not equal to
WILL_NEVER
select a Routing Tuple (the "previous Routing Tuple") whose
R_dest_addr is contained in N2_neighbor_iface_addr_list, and add
a new Routing Tuple with:
* R_dest_addr = N2_2hop_iface_addr;
* R_next_iface_addr = R_next_iface_addr of the previous Routing
Tuple;
* R_dist = 2;
* R_local_iface_addr = R_local_iface_addr of the previous
Routing Tuple.
C.3. Add Attached Networks
1. For each Attached Network Tuple, if for the Advertising Remote
Node Tuple with AR_orig_addr == AN_orig_addr, there is an address
in the AR_iface_addr_list which is equal to the R_dest_addr of a
Routing Tuple (the "previous Routing Tuple"), then:
1. If there is no Routing Tuple with R_dest_addr == AN_net_addr,
then add a new Routing Tuple with:
+ R_dest_addr = AN_net_addr;
+ R_next_iface_addr = R_next_iface_addr of the previous
Routing Tuple;
+ R_dist = (R_dist of the previous Routing Tuple) + AN_dist;
+ R_local_iface_addr = R_local_iface_addr of the previous
Routing Tuple.
2. Otherwise if the Routing Tuple with R_dest_addr ==
AN_net_addr (the "current Routing Tuple") has R_dist >
(R_dist of the previous Routing Tuple) + AN_dist, then modify
the current Routing Tuple by:
+ R_next_iface_addr = R_next_iface_addr of the previous
Routing Tuple;
+ R_dist = (R_dist of the previous Routing Tuple) + AN_dist;
+ R_local_iface_addr = R_local_iface_addr of the previous
Routing Tuple.
Appendix D. Packet and Message Layout Appendix D. Packet and Message Layout
This appendix illustrates the translation from the abstract This appendix illustrates the translation from the abstract
descriptions of packets employed in the protocol specification, and descriptions of packets employed in the protocol specification, and
the bit-layout packets actually exchanged between the nodes. the bit-layout packets actually exchanged between the nodes.
Appendix D.1. Packet and Message Options Appendix D.1. Packet and Message Options
The basic layout of an OLSRv2 packet is as described in [3]. However The basic layout of an OLSRv2 packet is as described in [1]. However
the following points should be noted. the following points should be noted.
In the following figures, reserved bits marked Reserved or Resv MUST In the following figures, reserved bits marked Reserved or Resv MUST
be cleared ('0'). Octets indicated as Padding are optional and MAY be cleared ('0'). Octets indicated as Padding are optional and MAY
be omitted; if not omitted they SHOULD be used to pad to a 32 bit be omitted; if not omitted they SHOULD be used to pad to a 32 bit
boundary and MUST all be zero. boundary and MUST all be zero.
OLSRv2 uses only packets with a packet header including a packet OLSRv2 uses only packets with a packet header including a packet
sequence number, either with or without a packet TLV block. Thus all sequence number, either with or without a packet TLV block. Thus all
OLSRv2 packets have the layout of either OLSRv2 packets have the layout of either
skipping to change at page 53, line 34 skipping to change at page 63, line 34
| Message + Padding | | Message + Padding |
| | | |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
OLSRv2 uses only messages with a complete message header. Thus all OLSRv2 uses only messages with a complete message header. Thus all
OLSRv2 messages, plus padding if any, have the following layout. OLSRv2 messages, plus padding if any, have the following layout.
0 1 2 3 0 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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Message Type | Resv |N|0|0| Message Size | | Message Type | Rsv |N|0|0|0|0| Message Size |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Originator Address | | Originator Address |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Hop Limit | Hop Count | Message Sequence Number | | Hop Limit | Hop Count | Message Sequence Number |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| | | |
| Message Body | | Message Body |
| | | |
| +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| | Padding | | | Padding |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
In standard OLSRv2 messages (HELLO and TC) the type dependent In standard OLSRv2 messages (HELLO and TC) the type dependent
sequence number bit marked N MUST be cleared ('0'). sequence number bit marked N MUST be cleared ('0').
The layouts of the message body, address block, TLV block and TLV are The layouts of the message body, address block, TLV block and TLV are
as in [3], allowing all options. Standard (HELLO and TC) messages as in [1], allowing all options. Standard (HELLO and TC) messages
contain a first address block which contains local interface address contain a first address block which contains local interface address
information, all other address blocks contain neighbor interface information, all other address blocks contain neighbor interface
address information (or for a TC message address information for address information (or, for a TC message, address information for
which it is a gateway) specific to the message type. which it is a gateway) specific to the message type.
Appendix D.2. Example HELLO Message Appendix D.2. Example HELLO Message
An example HELLO message, using IPv4 (four octet) addresses is as An example HELLO message, using IPv4 (four octet) addresses is as
follows. The overall message length is 58 octets. The message has a follows. The overall message length is 58 octets. The message has a
hop limit of 1 and a hop count of 0, as sent by its originator. hop limit of 1 and a hop count of 0, as sent by its originator.
The message has a message TLV block with content length 12 octets The message has a message TLV block with content length 12 octets
containing three message TLVs. These TLVs represent message validity containing three message TLVs. These TLVs represent message validity
time, message interval time and willingness. Each uses a TLV with time, message interval time and willingness. Each uses a TLV with
semantics value 4, indicating no start and stop indexes are included, semantics value 8, indicating no start and stop indexes are included,
and each has a value length of 1 octet. and each has a value length of 1 octet.
The first address block contains 1 local interface address. The The first address block contains 1 local interface address. The
semantics octet 2 indicates it has no tail section. It has head semantics octet 2 indicates it has no tail section. It has head
length 4, this is equal to the address length, it thus has no mid length 4, this is equal to the address length, it thus has no mid
section. This address block has no TLVs (TLV block content length is section. This address block has no TLVs (TLV block content length is
0 octets). 0 octets).
The second, and last, address block includes 4 neighbor interface The second, and last, address block includes 4 neighbor interface
addresses. The semantics octet 2 indicates they have no tail addresses. The semantics octet 2 indicates they have no tail
section. The addresses have head length 3 octets, thus each mid section. The addresses have head length 3 octets, thus each mid
section is of length one octet. The following address TLV block section is of length one octet. The following address TLV block
(content length 11 octets) includes two TLVs. (content length 11 octets) includes two TLVs.
The first of these TLVs reports the link status of all four neighbors The first of these TLVs reports the link status of all four neighbors
in a single multivalue TLV, the first two addresses are HEARD, the in a single multivalue TLV, the first two addresses are HEARD, the
last two addresses are SYMMETRIC. The TLV semantics octet value of last two addresses are SYMMETRIC. The TLV semantics octet value of
20 indicates, in addition to that this is a multivalue TLV, that no 40 indicates, in addition to that this is a multivalue TLV, that no
start index and stop index are included, hence values for all start index and stop index are included, hence values for all
addresses are included. The TLV value length of 4 octets indicates addresses are included. The TLV value length of 4 octets indicates
one octet per value per address. one octet per value per address.
The second of these TLVs indicates that the last address (start index The second of these TLVs indicates that the last address (start index
3, stop index 3) is an MPR. This TLV has no value, or value length, 3, stop index 3) is an MPR. This TLV has no value, or value length,
fields, as indicated by its semantics octet being equal to 2. fields, as indicated by its semantics octet being equal to 4.
0 1 2 3 0 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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| HELLO |0 0 0 0 0 0 0 0|0 0 0 0 0 0 0 0 0 0 1 1 1 0 1 0| | HELLO |0 0 0 0 0 0 0 0|0 0 0 0 0 0 0 0 0 0 1 1 1 0 1 0|
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Originator Address | | Originator Address |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|0 0 0 0 0 0 0 1|0 0 0 0 0 0 0 0| Message Sequence Number | |0 0 0 0 0 0 0 1|0 0 0 0 0 0 0 0| Message Sequence Number |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0| VALIDITY_TIME |0 0 0 0 0 1 0 0| |0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0| VALIDITY_TIME |0 0 0 0 1 0 0 0|
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|0 0 0 0 0 0 0 1| Value | INTERVAL_TIME |0 0 0 0 0 1 0 0| |0 0 0 0 0 0 0 1| Value | INTERVAL_TIME |0 0 0 0 1 0 0 0|
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|0 0 0 0 0 0 0 1| Value | WILLINGNESS |0 0 0 0 0 1 0 0| |0 0 0 0 0 0 0 1| Value | WILLINGNESS |0 0 0 0 1 0 0 0|
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|0 0 0 0 0 0 0 1| Value |0 0 0 0 0 0 0 1|0 0 0 0 0 0 1 0| |0 0 0 0 0 0 0 1| Value |0 0 0 0 0 0 0 1|0 0 0 0 0 0 1 0|
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|0 0 0 0 0 1 0 0| Head | |0 0 0 0 0 1 0 0| Head |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Head (cont) |0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0|0 0 0 0 0 1 0 0| | Head (cont) |0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0|0 0 0 0 0 1 0 0|
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|0 0 0 0 0 0 1 0|0 0 0 0 0 0 1 1| Head | |0 0 0 0 0 0 1 0|0 0 0 0 0 0 1 1| Head |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Head (cont) | Mid | Mid | Mid | | Head (cont) | Mid | Mid | Mid |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Mid |0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 1| LINK_STATUS | | Mid |0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 1| LINK_STATUS |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|0 0 0 1 0 1 0 0|0 0 0 0 0 1 0 0| HEARD | HEARD | |0 0 1 0 1 0 0 0|0 0 0 0 0 1 0 0| HEARD | HEARD |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| SYMMETRIC | SYMMETRIC | MPR |0 0 0 0 0 0 1 0| | SYMMETRIC | SYMMETRIC | MPR |0 0 0 0 0 1 0 0|
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|0 0 0 0 0 0 1 1|0 0 0 0 0 0 1 1| |0 0 0 0 0 0 1 1|0 0 0 0 0 0 1 1|
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
Appendix D.3. Example TC Message Appendix D.3. Example TC Message
An example TC message, using IPv4 (four octet) addresses, is as An example TC message, using IPv4 (four octet) addresses, is as
follows. The overall message length is 67 octets. follows. The overall message length is 67 octets.
The message has a message TLV block with content length 13 octets The message has a message TLV block with content length 13 octets
containing three TLVs. The first two TLVs are validity and interval containing three TLVs. The first two TLVs are validity and interval
times as for the HELLO message above. The third TLV is a content times as for the HELLO message above. The third TLV is a content
sequence number TLV used to carry the 2 octet ANSN. The semantics sequence number TLV used to carry the 2 octet ANSN. The semantics
value is also 4. value is also 8.
The message has three address blocks. The first address block The message has three address blocks. The first address block
contains 3 local interface addresses (with semantics octet 2, hence contains 3 local interface addresses (with semantics octet 2, hence
no tail section, head length 2 octets, and hence mid sections with no tail section, head length 2 octets, and hence mid sections with
length two octets) and has no TLVs (TLV block content length 0 length two octets) and has no TLVs (TLV block content length 0
octets). octets).
The other two address blocks contain neighbor interface addresses. The other two address blocks contain neighbor interface addresses.
The first contains 3 addresses (semantics octet 2, no tail section, The first contains 3 addresses (semantics octet 2, no tail section,
head length 2 octets, hence mid sections length two octets) and has head length 2 octets, hence mid sections length two octets) and has
no TLVs (TLV block content length 0 octets). The second contains 1 no TLVs (TLV block content length 0 octets). The second contains 1
address, with semantics octet 4 indicating that the tail section, address, with semantics octet 4 indicating that the tail section,
length 2 octets, consists of zero valued octets (not included). The length 2 octets, consists of zero valued octets (not included). The
following TLV block (content length 6 octets) includes two TLVs, the following TLV block (content length 6 octets) includes two TLVs, the
first (semantics value 4 indicating no indexes are needed) indicates first (semantics value 8 indicating no indexes are needed) indicates
that the address has a netmask, with length given by the value (of that the address has a netmask, with length given by the value (of
length 1 octet) of 16. Thus this address is Head.0.0/16. The second length 1 octet) of 16. Thus this address is Head.0.0/16. The second
TLV indicates that the originating node is a gateway to this network, TLV indicates that the originating node is a gateway to this network,
at a given number of hops distance. The TLV semantics value of 4 at a given number of hops distance. The TLV semantics value of 8
indicates that no indexes are needed. indicates that no indexes are needed.
0 1 2 3 0 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
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| TC |0 0 0 0 0 0 0 0|0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 0| | TC |0 0 0 0 0 0 0 0|0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 0|
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Originator Address | | Originator Address |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Hop Limit | Hop Count | Message Sequence Number | | Hop Limit | Hop Count | Message Sequence Number |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 1| VALIDITY_TIME |0 0 0 0 0 1 0 0| |0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 1| VALIDITY_TIME |0 0 0 0 1 0 0 0|
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|0 0 0 0 0 0 0 1| Value | INTERVAL_TIME |0 0 0 0 0 1 0 0| |0 0 0 0 0 0 0 1| Value | INTERVAL_TIME |0 0 0 0 1 0 0 0|
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|0 0 0 0 0 0 0 1| Value | CONT_SEQ_NUM |0 0 0 0 0 1 0 0| |0 0 0 0 0 0 0 1| Value | CONT_SEQ_NUM |0 0 0 0 1 0 0 0|
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|0 0 0 0 0 0 1 0| Value (ANSN) |0 0 0 0 0 0 1 1| |0 0 0 0 0 0 1 0| Value (ANSN) |0 0 0 0 0 0 1 1|
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| 0x02 |0 0 0 0 0 0 1 0| Head | | 0x02 |0 0 0 0 0 0 1 0| Head |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Mid | Mid | | Mid | Mid |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Mid |0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0| | Mid |0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0|
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|0 0 0 0 0 0 1 1| 0x02 |0 0 0 0 0 0 1 0| Head | |0 0 0 0 0 0 1 1| 0x02 |0 0 0 0 0 0 1 0| Head |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Head (cont) | Mid | Mid | | Head (cont) | Mid | Mid |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Mid (cont) | Mid |0 0 0 0 0 0 0 0| | Mid (cont) | Mid |0 0 0 0 0 0 0 0|
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|0 0 0 0 0 0 0 0|0 0 0 0 0 0 0 1|0 0 0 0 0 1 0 0|0 0 0 0 0 0 1 0| |0 0 0 0 0 0 0 0|0 0 0 0 0 0 0 1|0 0 0 0 0 1 0 0|0 0 0 0 0 0 1 0|
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Head |0 0 0 0 0 0 1 0|0 0 0 0 0 0 0 0| | Head |0 0 0 0 0 0 1 0|0 0 0 0 0 0 0 0|
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|0 0 0 0 0 1 1 1| PREFIX_LENGTH |0 0 0 0 0 1 0 0|0 0 0 0 0 0 0 1| |0 0 0 0 0 1 1 1| PREFIX_LENGTH |0 0 0 0 1 0 0 0|0 0 0 0 0 0 0 1|
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
|0 0 0 1 0 0 0 0| GATEWAY |0 0 0 0 0 1 0 0| Number Hops | |0 0 0 1 0 0 0 0| GATEWAY |0 0 0 0 1 0 0 0| Number Hops |
+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
Appendix E. Time TLVs Appendix E. Constraints
This appendix specifies a general time TLV structure for expressing
either single time values or a set of time values with each value
associated with a range of distances. Furthermore, using this
general time TLV structure, this document specifies the INTERVAL_TIME
and VALIDITY_TIME TLVs, which are used by OLSRv2.
E.1. Representing Time
This document specifies a TLV structure in which time values are each
represented in an 8 bit time code, one or more of which may be used
in a TLV's value field. Of these 8 bits, the least significant four
bits represent the mantissa (a), and the most significant four bits
represent the exponent (b), so that:
o time value = (1 + a/16) * 2^b * C
o time code = 16 * b + a
All nodes in the network MUST use the same value of C, which will be
specified in seconds, hence so will be all time values. Note that
ascending values of the time code represent ascending time values,
time values may thus be compared by comparison of time codes.
An algorithm for computing the time code representing the smallest
representable time value not less than the time value t is:
1. find the largest integer b such that t/C >= 2^b;
2. set a = 16 * (t / (C * 2^b) - 1), rounded up to the nearest
integer;
3. if a == 16 then set b = b + 1 and set a = 0;
4. if a and b are in the range 0 and 15 then the required time value
can be represented by the time code 16 * b + a, otherwise it can
not.
The minimum time value that can be represented in this manner is C.
The maximum time value that can be represented in this manner is
63488 * C.
E.2. General Time TLV Structure
A Time TLV may be a packet, message or address block TLV. If it is a
packet or message TLV then it must be a single value TLV as defined
in [3]; if it is an address block TLV then it may be single value or
multivalue TLV. The specific Time TLVs specified in this document,
in Appendix E.3 are message, and hence single value, TLVs. Note that
even a single value Time TLV may contain a multiple octet <value>
field.
The purpose of a single value Time TLV is to allow a single time
value to be determined by a node receiving an entity containing the
Time TLV, based on its distance from the entity's originator. The
Time TLV may contain information that allows that time value to be a
function of distance, and thus different receiving nodes may
determine different time values. If a receiving node will not be
able to determine its distance from the originating node, then the
form of this Time TLV with a single time code in a <value> field (or
single value subfield) SHOULD be used.
The <value> field of a single value Time TLV is specified, using the Any process which updates the Local Information Base, the
regular expression syntax of [3], by: Neighborhood Information Base or the Topology Information Base MUST
ensure that all constraints specified in this appendix are
maintained, as well as those specified in [4].
<value> = {<time><distance>}*<time> In each Local Attached Network Tuple:
where: o AL_net_addr MUST NOT be in the I_local_iface_addr_list of any
Local Interface Tuple.
<time> is an 8 bit field containing a time code as defined in o AL_dist MUST NOT be less than zero.
Appendix E.1.
<distance> is an 8 bit field specifying a distance from the message In each Link Tuple:
originator, in hops.
A single value <value> field thus consists of an odd number of o L_neighbor_iface_addr_list MUST NOT contain the AL_net_addr of any
octets; with a repetition factor of n in the regular expression Local Attached Network Tuple.
syntax it contains 2n+1 octets, thus the <length> field of a single
value Time TLV, which MUST always be present, is given by:
o <length> = 2n+1 o If L_status == SYMMETRIC and the Neighbor Tuple whose
N_neighbor_iface_addr_list contains L_neighbor_iface_addr_list has
N_mpr_selector == true, then, for each address in this
L_neighbor_iface_addr_list, there MUST be an equal
RY_neighbor_iface_addr in the Relay Set associated with the same
OLSRv2 interface.
A single value <value> field may be thus represented by: In each Neighbor Tuple:
<t_1><d_1><t_2><d_2> ... <t_i><d_i> ... <t_n><d_n><t_default> o N_neighbor_iface_addr_list MUST NOT contain the AL_net_addr of any
Local Attached Network Tuple.
<d_1>, ... <d_n>, if present, MUST be a strictly increasing sequence. o If N_willingness MUST be in the range from WILL_NEVER to
Then, at the receiving node's distance from the originator node, the WILL_ALWAYS, inclusive.
time value indicated is that represented by the time code:
o <t_1>, if n > 0 and distance <= <d_1>; o If N_mpr == true, then N_symmetric MUST be true and N_willingness
MUST NOT equal WILL_NEVER.
o <t_i+1>, if n > 1 and <d_i> < distance <= <d_i+1> for some i such o If N_symmetric == true and N_mpr == false, then N_willingness MUST
that 1 <= i < n; NOT equal WILL_ALWAYS.
o <t_default> otherwise, i.e. if n == 0 or distance > <d_n>.
In a multivalue Time TLV, each single value subfield of the o If N_mpr_selector == true, then N_symmetric MUST be true.
multivalue Time TLV is defined as above. Note that [3] requires that
each single value subfield has the same length (i.e. the same value
of n) but they need not use the same values of <d_1> to <d_n>.
E.3. Message TLVs o If N_mpr_selector == true, then, for each address in this
N_neighbor_iface_addr_list, there MUST be an equal
A_neighbor_iface_addr in the Advertised Neighbor Set.
Two message TLVs are defined, for signaling message validity time In each Lost Neighbor Tuple:
(VALIDITY_TIME) and message interval (INTERVAL_TIME).
E.3.1. VALIDITY_TIME TLV o NL_neighbor_iface_addr MUST NOT equal the AL_net_addr of any Local
Attached Network Tuple.
A VALIDITY TIME TLV is a message TLV that defines the validity time In each 2-Hop Tuple:
of the information carried in the message in which the TLV is
contained. After this time the receiving node MUST consider the
message content to no longer be valid (unless repeated in a later
message). The validity time of a message MAY be specified to depend
on the distance from its originator. (This is appropriate if
messages are sent with different hop limits, so that receiving nodes
at greater distances receive information less frequently and must
treat is as valid for longer.)
A VALIDITY_TIME TLV is an example of a Time TLV specified as in o N2_2hop_iface_addr MUST NOT equal the AL_net_addr of any Local
Appendix E.1. Attached Network Tuple.
E.3.2. INTERVAL_TIME TLV In each Received Tuple:
An INTERVAL_TIME TLV is a message TLV that defines the maximum time o RX_orig_addr SHOULD NOT be in the I_local_iface_addr_list of any
before another message of the same type as this message from the same Local Interface Tuple.
originator should be received. This interval time MAY be specified
to depend on the distance from the originator. (This is appropriate
if messages are sent with different hop limits, so that receiving
nodes at greater distances have an increased interval time.)
An INTERVAL_TIME TLV is an example of a Time TLV specified as in o RX_orig_addr SHOULD NOT equal the AL_net_addr of any Local
Appendix E.1. Attached Network Tuple.
Appendix F. Message Jitter o Each ordered triple (RX_type, RX_orig_addr, RX_seq_number) SHOULD
NOT equal the corresponding triple in any other Received Tuple in
the same Received Set.
Since NHDP employs periodic message transmission in order to detect In each Processed Tuple:
neighborhoods, and since NHDP is a building block for MANET routing
protocols employing other triggered or periodic message exchanges,
this appendix presents global concerns pertaining to jittering of
MANET control traffic.
F.1. Jitter o P_orig_addr SHOULD NOT be in the I_local_iface_addr_list of any
Local Interface Tuple.
In order to prevent nodes in a MANET from simultaneous transmission, o P_orig_addr SHOULD NOT equal the AL_net_addr of any Local Attached
whilst retaining the MANET characteristic of maximum node autonomy, a Network Tuple.
randomization of the transmission time of packets by nodes, known as
jitter, MAY be employed. Three jitter mechanisms, which target
different aspects of this problem, MAY be employed, with the aim of
reducing the likelihood of simultaneous transmission, and, if it
occurs, preventing it from continuing.
Three cases exist: o Each ordered triple (P_type, P_orig_addr, P_seq_number) SHOULD NOT
equal the corresponding triple in any other Processed Tuple.
o Periodic message generation; In each Forwarded Tuple:
o Externally triggered message generation; o F_orig_addr SHOULD NOT be in the I_local_iface_addr_list of any
Local Interface Tuple.
o Message forwarding. o F_orig_addr SHOULD NOT equal the AL_net_addr of any Local Attached
Network Tuple.
Each of these cases uses a parameter, denoted MAXJITTER, for the o Each ordered triple (F_type, F_orig_addr, F_seq_number) SHOULD NOT
maximum timing variation that it introduces. If more than one of equal the corresponding triple in any other Forwarded Tuple.
these cases is used by a protocol, it MAY use the same or a different
value of MAXJITTER for each case. It also MAY use the same or
different values of MAXJITTER according to message type, and under
different circumstances - in particular if other parameters (such as
message interval) vary.
Issues relating to the value of MAXJITTER are considered in In each Relay Set:
Appendix F.1.4.
F.1.1. Periodic message generation o Each RY_neighbor_iface_addr SHOULD NOT equal any other
RY_neighbor_iface_addr.
When a node generates a message periodically, two successive messages o Each RY_neighbor_iface_addr MUST be in the
will be separated by a well-defined interval, denoted L_neighbor_iface_addr_list of a Link Tuple with L_status ==
MESSAGE_INTERVAL. A node MAY maintain more than one such interval, SYMMETRIC.
e.g. for different message types or in different circumstances (such
as backing off transmissions to avoid congestion). Jitter MAY be
applied by reducing this delay by a random amount, so that the delay
between consecutive transmissions of a messages of the same type is
equal to (MESSAGE_INTERVAL - jitter), where jitter is the random
value.
Subtraction of the random value from the message interval ensures In the Advertised Neighbor Set:
that the message interval never exceeds MESSAGE_INTERVAL, and does
not adversely affect timeouts or other mechanisms which may be based
on message late arrival or failure to arrive. By basing the message
transmission time on the previous transmission time, rather than by
jittering a fixed clock, nodes can become completely desynchronized,
which minimizes their probability of repeated collisions. This is
particularly useful when combined with externally triggered message
generation and rescheduling.
The jitter value SHOULD be taken from a uniform distribution between o Each A_neighbor_iface_addr MUST NOT equal any other
zero and MAXJITTER. A_neighbor_iface_addr.
Note that a node will know its own MESSAGE_INTERVAL value and can o Each A_neighbor_iface_addr MUST be in the
readily ensure that any MAXJITTER value used satisfies the conditions N_neighbor_iface_addr_list of a Neighbor Tuple with N_symmetric ==
in Appendix F.1.4. true.
F.1.2. Externally triggered message generation In each Advertising Remote Node Tuple:
An internal or external condition or event MAY trigger message o AR_orig_addr SHOULD NOT be in the I_local_iface_addr_list of any
generation by a node. Depending upon the protocol, this condition Local Interface Tuple.
MAY trigger generation of a single message, initiation of a new
periodic message schedule, or rescheduling of existing periodic
messaging. Collision between externally triggered messages is made
more likely if more than one node is likely to respond to the same
event. To reduce this likelihood, an externally triggered message
MAY be jittered by delaying it by a random duration; an internally
triggered message MAY also be so jittered if appropriate. This delay
SHOULD be generated uniformly in an interval between zero and
MAXJITTER. If periodically transmitted messages are rescheduled,
then this SHOULD be based on this delayed time, with subsequent
messages treated as described in Appendix F.1.1.
When messages are triggered, whether or not they are also o AR_orig_addr SHOULD NOT equal the AL_net_addr of any Local
periodically transmitted, a protocol MAY impose a minimum interval Attached Network Tuple.
between messages of the same type, denoted MESSAGE_MIN_INTERVAL. It
is however appropriate to also allow this interval to be reduced by
jitter, so that when a message is transmitted the next message is
allowed after a time (MESSAGE_MIN_INTERVAL - jitter), where jitter
SHOULD be generated uniformly in an interval between zero and
MAXJITTER (using a value of MAXJITTER appropriate to periodic message
transmission). This is because otherwise, when external triggers are
more frequent than MESSAGE_MIN_INTERVAL, it takes the role of
MESSAGE_INTERVAL and the arguments applying to jittering of the
latter also apply to the former. This also permits
MESSAGE_MIN_INTERVAL to equal MESSAGE_INTERVAL even when jitter is
used.
F.1.3. Message forwarding o Each AR_orig_addr MUST NOT equal the AR_orig_addr in any other
ANSN History Tuple.
When a node forwards a message, it may be jittered by delaying it by o AR_iface_addr_list MUST NOT contain any address which is in the
a random duration. This delay SHOULD be generated uniformly in an I_local_iface_addr_list of any Local Interface Tuple.
interval between zero and MAXJITTER.
Unlike the cases of periodically generated and externally triggered o AR_iface_addr_list MUST NOT contain any address which is the
messages, a node is not automatically aware of the message AL_net_addr of any Local Attached Network Tuple.
originator's value of MESSAGE_INTERVAL, which is required to select a
value of MAXJITTER which is known to be valid. This may require
prior agreement as to the value (or minimum value) of
MESSAGE_INTERVAL, may be by inclusion in the message of
MESSAGE_INTERVAL (the time until the next relevant message, rather
than the time since the last message) or be by any other protocol
specific mechanism, which may include estimation of the value of
MESSAGE_INTERVAL based on received message times.
For several possible reasons (differing parameters, message In each Topology Tuple:
rescheduling, extreme random values) a node may receive a message
while still waiting to forward an earlier message of the same type
originating from the same node. This is possible without jitter, but
may occur more often with it. The appropriate action to take is
protocol specific (typically to discard the earlier message or to
forward both, possible modifying timing to maintain message order).
In many cases, including [1] and protocols using the full o T_dest_iface_addr MUST NOT be in the I_local_iface_addr_list of
functionality of [3], messages are transmitted hop by hop in any Local Interface Tuple.
potentially multi-message packets, and some or all of those messages
may need to be forwarded. For efficiency this should be in a single
packet, and hence the forwarding jitter of all messages received in a
single packet should be the same. (This also requires that a single
value of MAXJITTER is used in this case.) For this to have the
intended uniform distribution it is necessary to choose a single
random jitter for all messages. It is not appropriate to give each
message a random jitter and then to use the smallest of these jitter
values, as that produces a jitter with a non-uniform distribution and
a reduced mean value.
In addition, the protocol may permit messages received in different o T_dest_iface_addr MUST NOT equal the AL_net_addr of any Local
packets to be combined, possibly also with locally generated messages Attached Network Tuple.
(periodically generated or triggered). However in this case the
purpose of the jitter will be accomplished by choosing any of the
independently scheduled times for these events as the single
forwarding time; this may have to be the earliest time to achieve all
constraints. This is because without combining messages, a
transmission was due at this time anyway.
F.1.4. Maximum Jitter Determination o There MUST be an Advertising Remote Node Tuple with AR_orig_addr
== T_orig_addr.
In considering how the maximum jitter (one or more instances of o T_dest_iface_addr MUST NOT be in the AR_iface_addr_list of the
parameter MAXJITTER) may be determined, the following points may be Advertising Remote Node Tuple with AR_orig_addr == T_orig_addr.
noted:
o While jitter may resolve the problem of simultaneous o T_seq_number MUST NOT be greater than AR_seq_number of the
transmissions, the timing changes (in particular the delays) it Advertising Remote Node Tuple with AR_orig_addr == T_orig_addr.
introduces will otherwise only have a negative impact on a well-
designed protocol. Thus MAXJITTER should always be minimized,
subject to acceptably achieving its intent.
o When messages are periodically generated, all of the following o The ordered pair (T_dest_iface_addr, T_orig_addr) MUST NOT equal
that are relevant apply to each instance of MAXJITTER: the corresponding pair in any other Topology Tuple.
* it MUST NOT be greater than MESSAGE_INTERVAL/2; In each Attached Network Tuple:
* it SHOULD be significantly less than MESSAGE_INTERVAL; o AN_net_addr MUST NOT be in the I_local_iface_addr_list of any
Local Interface Tuple.
* it MUST NOT be greater than MESSAGE_MIN_INTERVAL; o AN_net_addr MUST NOT equal the AL_net_addr of any Local Attached
Network Tuple.
* it SHOULD NOT be greater than MESSAGE_MIN_INTERVAL/2. o There MUST be an Advertising Remote Node Tuple with AR_orig_addr
== AN_orig_addr.
o As well as the decision as to whether to use jitter being o AN_seq_number MUST NOT be greater than AR_seq_number of the
dependent on the medium access control and lower layers, the Advertising Remote Node Tuple with AR_orig_addr == AN_orig_addr.
selection of the MAXJITTER parameter should be appropriate to
those mechanisms.
o As jitter is intended to reduce collisions, greater jitter, i.e. o AN_dist MUST NOT be less than zero.
an increased value of MAXJITTER, is appropriate when the chance of
collisions is greater. This is particularly the case with
increased node density, where node density should be considered
relative to (the square of) the interference range rather than
useful signal range.
o The choice of MAXJITTER used when forwarding messages may also o The ordered pair (AN_net_addr, AN_orig_addr) MUST NOT equal the
take into account the expected number of times that the message corresponding pair in any other Attached Network Tuple.
may be sequentially forwarded, up to the network diameter in hops.
Appendix G. Security Considerations Appendix F. Security Considerations
Currently, OLSRv2 does not specify any special security measures. As Currently, OLSRv2 does not specify any special security measures. As
a proactive routing protocol, OLSRv2 makes a target for various a proactive routing protocol, OLSRv2 makes a target for various
attacks. The various possible vulnerabilities are discussed in this attacks. The various possible vulnerabilities are discussed in this
section. section.
Appendix G.1. Confidentiality Appendix F.1. Confidentiality
Being a proactive protocol, OLSRv2 periodically diffuses topological Being a proactive protocol, OLSRv2 periodically diffuses topological
information. Hence, if used in an unprotected wireless network, the information. Hence, if used in an unprotected wireless network, the
network topology is revealed to anyone who listens to OLSRv2 control network topology is revealed to anyone who listens to OLSRv2 control
messages. messages.
In situations where the confidentiality of the network topology is of In situations where the confidentiality of the network topology is of
importance, regular cryptographic techniques, such as exchange of importance, regular cryptographic techniques, such as exchange of
OLSRv2 control traffic messages encrypted by PGP [5] or encrypted by OLSRv2 control traffic messages encrypted by PGP [9] or encrypted by
some shared secret key, can be applied to ensure that control traffic some shared secret key, can be applied to ensure that control traffic
can be read and interpreted by only those authorized to do so. can be read and interpreted by only those authorized to do so.
Appendix G.2. Integrity Appendix F.2. Integrity
In OLSRv2, each node is injecting topological information into the In OLSRv2, each node is injecting topological information into the
network through transmitting HELLO messages and, for some nodes, TC network through transmitting HELLO messages and, for some nodes, TC
messages. If some nodes for some reason, malicious or malfunction, messages. If some nodes for some reason, malicious or malfunction,
inject invalid control traffic, network integrity may be compromised. inject invalid control traffic, network integrity may be compromised.
Therefore, message authentication is recommended. Therefore, message authentication is recommended.
Different such situations may occur, for instance: Different such situations may occur, for instance:
1. a node generates TC messages, advertising links to non-neighbor 1. a node generates TC messages, advertising links to non-neighbor
skipping to change at page 66, line 16 skipping to change at page 73, line 16
Authentication of the originator node for control messages (for Authentication of the originator node for control messages (for
situations 2, 4 and 5) and on the individual links announced in the situations 2, 4 and 5) and on the individual links announced in the
control messages (for situations 1 and 3) may be used as a control messages (for situations 1 and 3) may be used as a
countermeasure. However to prevent nodes from repeating old (and countermeasure. However to prevent nodes from repeating old (and
correctly authenticated) information (situation 9) temporal correctly authenticated) information (situation 9) temporal
information is required, allowing a node to positively identify such information is required, allowing a node to positively identify such
delayed messages. delayed messages.
In general, digital signatures and other required security In general, digital signatures and other required security
information may be transmitted as a separate OLSRv2 message type, information may be transmitted as a separate OLSRv2 message type, or
thereby allowing that "secured" and "unsecured" nodes can coexist in signatures and security information may be transmitted within the
the same network, if desired, or signatures and security information OLSRv2 HELLO and TC messages, using the TLV mechanism. Either option
may be transmitted within the OLSRv2 HELLO and TC messages, using the permits that "secured" and "unsecured" nodes can coexist in the same
TLV mechanism. network, if desired,
Specifically, the authenticity of entire OLSRv2 control messages can Specifically, the authenticity of entire OLSRv2 control messages can
be established through employing IPsec authentication headers, be established through employing IPsec authentication headers,
whereas authenticity of individual links (situations 1 and 3) require whereas authenticity of individual links (situations 1 and 3) require
additional security information to be distributed. additional security information to be distributed.
An important consideration is, that all control messages in OLSRv2 An important consideration is, that all control messages in OLSRv2
are transmitted either to all nodes in the neighborhood (HELLO are transmitted either to all nodes in the neighborhood (HELLO
messages) or broadcast to all nodes in the network (TC messages). messages) or broadcast to all nodes in the network (TC messages).
For example, a control message in OLSRv2 is always a point-to- For example, a control message in OLSRv2 is always a point-to-
multipoint transmission. It is therefore important that the multipoint transmission. It is therefore important that the
authentication mechanism employed permits that any receiving node can authentication mechanism employed permits that any receiving node can
validate the authenticity of a message. As an analogy, given a block validate the authenticity of a message. As an analogy, given a block
of text, signed by a PGP private key, then anyone with the of text, signed by a PGP private key, then anyone with the
corresponding public key can verify the authenticity of the text. corresponding public key can verify the authenticity of the text.
Appendix G.3. Interaction with External Routing Domains Appendix F.3. Interaction with External Routing Domains
OLSRv2 does, through the use of TC messages, provide a basic OLSRv2 does, through the use of TC messages, provide a basic
mechanism for injecting external routing information to the OLSRv2 mechanism for injecting external routing information to the OLSRv2
domain. Appendix A also specifies that routing information can be domain. Appendix A also specifies that routing information can be
extracted from the topology table or the routing table of OLSRv2 and, extracted from the topology table or the routing table of OLSRv2 and,
potentially, injected into an external domain if the routing protocol potentially, injected into an external domain if the routing protocol
governing that domain permits. governing that domain permits.
Other than as described in Appendix A, when operating nodes, Other than as described in Appendix A, when operating nodes
connecting OLSRv2 to an external routing domain, care MUST be taken connecting OLSRv2 to an external routing domain, care MUST be taken
not to allow potentially insecure and untrustworthy information to be not to allow potentially insecure and untrustworthy information to be
injected from the OLSRv2 domain to external routing domains. Care injected from the OLSRv2 domain to external routing domains. Care
MUST be taken to validate the correctness of information prior to it MUST be taken to validate the correctness of information prior to it
being injected as to avoid polluting routing tables with invalid being injected as to avoid polluting routing tables with invalid
information. information.
A recommended way of extending connectivity from an existing routing A recommended way of extending connectivity from an existing routing
domain to an OLSRv2 routed MANET is to assign an IP prefix (under the domain to an OLSRv2 routed MANET is to assign an IP prefix (under the
authority of the nodes/gateways connecting the MANET with the exiting authority of the nodes/gateways connecting the MANET with the exiting
routing domain) exclusively to the OLSRv2 MANET area, and to routing domain) exclusively to the OLSRv2 MANET area, and to
configure the gateways statically to advertise routes to that IP configure the gateways statically to advertise routes to that IP
sequence to nodes in the existing routing domain. sequence to nodes in the existing routing domain.
Appendix G.4. Node Identity Appendix F.4. Node Identity
OLSRv2 does not make any assumption about node addresses, other than OLSRv2 does not make any assumption about node addresses, other than
that each node is assumed to have at least one a unique and routable that each node is assumed to have at least one a unique and routable
IP address for each interface that it has which participates in the IP address for each interface that it has which participates in the
MANET. MANET.
Appendix H. Flow and Congestion Control Appendix G. Flow and Congestion Control
Due to its proactive nature, the OLSRv2 protocol has a natural Due to its proactive nature, the OLSRv2 protocol has a natural
control over the flow of its control traffic. Nodes transmit control control over the flow of its control traffic. Nodes transmit control
messages at predetermined rates specified and bounded by message messages at predetermined rates specified and bounded by message
intervals. intervals.
OLSRv2 employs [4] for local signalling, embedding MPR selection OLSRv2 employs [4] for local signaling, embedding MPR selection
advertisement through a simple address block TLV, and node advertisement through a simple address block TLV, and node
willingness advertisement (if any) as a single message TLV. OLSRv2 willingness advertisement (if any) as a single message TLV. OLSRv2
local signalling, therefore, shares the characteristics and local signaling, therefore, shares the characteristics and
constraints of [4]. constraints of [4].
Furthermore, the MPR optimization greatly constrains global Furthermore, the MPR optimization greatly constrains global signaling
signalling overhead from link state diffusion in two ways. First, overhead from link state diffusion in two ways. First, the messages
the messages that advertise the topology need only contain MPR that advertise the topology need only contain MPR selectors, reducing
selectors, reducing their size as compared to full link state. their size as compared to full link state. Second, the cost of
Second, the cost of diffusing these messages throughout the network diffusing these messages throughout the network is greatly reduced as
is greatly reduced as compared to when using classic flooding, since compared to when using classic flooding, since only MPRs need to
only MPRs need to forward broadcast messages. In dense networks, the forward broadcast messages. In dense networks, the reduction of
reduction of control traffic can be of several orders of magnitude control traffic can be of several orders of magnitude compared to
compared to routing protocols using classical flooding [8]. This routing protocols using classical flooding [12]. This feature
feature naturally provides more bandwidth for useful data traffic and naturally provides more bandwidth for useful data traffic and pushes
pushes further the frontier of congestion. further the frontier of congestion.
Since the control traffic is continuous and periodic, it keeps the Since the control traffic is continuous and periodic, it keeps the
quality of the links used in routing more stable. However, using quality of the links used in routing more stable. However, using
certain OLSRv2 options, some control messages (HELLO messages or TC certain OLSRv2 options, some control messages (HELLO messages or TC
messages) may be intentionally sent in advance of their deadline in messages) may be intentionally sent in advance of their deadline in
order to increase the responsiveness of the protocol to topology order to increase the responsiveness of the protocol to topology
changes. This may cause a small, temporary and local increase of changes. This may cause a small, temporary, and local increase of
control traffic, however this is at all times bounded by the use of control traffic, however this is at all times bounded by the use of
minimum message intervals. minimum message intervals.
Appendix I. Contributors Appendix H. Contributors
This specification is the result of the joint efforts of the This specification is the result of the joint efforts of the
following contributors -- listed alphabetically. following contributors -- listed alphabetically.
o Cedric Adjih, INRIA, France, <Cedric.Adjih@inria.fr> o Cedric Adjih, INRIA, France, <Cedric.Adjih@inria.fr>
o Emmanuel Baccelli, Hitachi Labs Europe, France, o Emmanuel Baccelli, INRIA , France, <Emmanuel.Baccelli@inria.fr>
<Emmanuel.Baccelli@inria.fr>
o Thomas Heide Clausen, PCRI, France<T.Clausen@computer.org> o Thomas Heide Clausen, LIX, France, <T.Clausen@computer.org>
o Justin Dean, NRL, USA<jdean@itd.nrl.navy.mil> o Justin Dean, NRL, USA<jdean@itd.nrl.navy.mil>
o Christopher Dearlove, BAE Systems, UK, o Christopher Dearlove, BAE Systems, UK,
<Chris.Dearlove@baesystems.com> <Chris.Dearlove@baesystems.com>
o Satoh Hiroki, Hitachi SDL, Japan, <h-satoh@sdl.hitachi.co.jp> o Satoh Hiroki, Hitachi SDL, Japan, <hiroki.satoh.yj@hitachi.com>
o Philippe Jacquet, INRIA, France, <Philippe.Jacquet@inria.fr> o Philippe Jacquet, INRIA, France, <Philippe.Jacquet@inria.fr>
o Monden Kazuya, Hitachi SDL, Japan, <monden@sdl.hitachi.co.jp> o Monden Kazuya, Hitachi SDL, Japan, <kazuya.monden.vw@hitachi.com>
o Kenichi Mase, Niigata University, Japan, <mase@ie.niigata-u.ac.jp> o Kenichi Mase, Niigata University, Japan, <mase@ie.niigata-u.ac.jp>
o Ryuji Wakikawa, KEIO University, Japan, <ryuji@sfc.wide.ad.jp> o Ryuji Wakikawa, KEIO University, Japan, <ryuji@sfc.wide.ad.jp>
Appendix J. Acknowledgements Appendix I. Acknowledgements
The authors would like to acknowledge the team behind OLSRv1, The authors would like to acknowledge the team behind OLSRv1,
specified in RFC3626, including Anis Laouiti, Pascale Minet, Laurent specified in RFC3626, including Anis Laouiti, Pascale Minet, Laurent
Viennot (all at INRIA, France), and Amir Qayyum (Center for Advanced Viennot (all at INRIA, France), and Amir Qayyum (Center for Advanced
Research in Engineering, Pakistan) for their contributions. Research in Engineering, Pakistan) for their contributions.
The authors would like to gratefully acknowledge the following people The authors would like to gratefully acknowledge the following people
for intense technical discussions, early reviews and comments on the for intense technical discussions, early reviews and comments on the
specification and its components: Li Li (CRC), Louise Lamont (CRC), specification and its components: Li Li (CRC), Louise Lamont (CRC),
Joe Macker (NRL), Alan Cullen (BAE Systems), Philippe Jacquet Joe Macker (NRL), Alan Cullen (BAE Systems), Philippe Jacquet
(INRIA), Khaldoun Al Agha (LRI), Richard Ogier (SRI), Song-Yean Cho (INRIA), Khaldoun Al Agha (LRI), Richard Ogier (SRI), Song-Yean Cho
(Samsung Software Center), Shubhranshu Singh (Samsung AIT) and the (Samsung Software Center), Shubhranshu Singh (Samsung AIT), Charles
entire IETF MANET working group. E. Perkins (Nokia) and the entire IETF MANET working group.
Authors' Addresses Authors' Addresses
Thomas Heide Clausen Thomas Heide Clausen
LIX, Ecole Polytechnique, France LIX, Ecole Polytechnique, France
Phone: +33 6 6058 9349 Phone: +33 6 6058 9349
Email: T.Clausen@computer.org Email: T.Clausen@computer.org
URI: http://www.lix.polytechnique.fr/Labo/Thomas.Clausen/ URI: http://www.ThomasClausen.org/
Christopher M. Dearlove Christopher M. Dearlove
BAE Systems Advanced Technology Centre BAE Systems Advanced Technology Centre
Phone: +44 1245 242194 Phone: +44 1245 242194
Email: chris.dearlove@baesystems.com Email: chris.dearlove@baesystems.com
URI: http://www.baesystems.com/ocs/sharedservices/atc/ URI: http://www.baesystems.com/
Philippe Jacquet Philippe Jacquet
Project Hipercom, INRIA Project Hipercom, INRIA
Phone: +33 1 3963 5263 Phone: +33 1 3963 5263
Email: philippe.jacquet@inria.fr Email: philippe.jacquet@inria.fr
URI: http://hipercom.inria.fr/test/Jacquet.htm
The OLSRv2 Design Team The OLSRv2 Design Team
MANET Working Group MANET Working Group
Full Copyright Statement Full Copyright Statement
Copyright (C) The IETF Trust (2007). Copyright (C) The IETF Trust (2007).
This document is subject to the rights, licenses and restrictions This document is subject to the rights, licenses and restrictions
contained in BCP 78, and except as set forth therein, the authors contained in BCP 78, and except as set forth therein, the authors
 End of changes. 417 change blocks. 
1347 lines changed or deleted 1612 lines changed or added

This html diff was produced by rfcdiff 1.33. The latest version is available from http://tools.ietf.org/tools/rfcdiff/