By submitting this Internet-Draft, each author represents that any 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 aware will be disclosed, in accordance with Section 6 of BCP 79.
Internet-Drafts are working documents of the Internet Engineering Task Force (IETF), its areas, and its working groups. Note that other groups may also distribute working documents as Internet-Drafts.
Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as “work in progress.”
The list of current Internet-Drafts can be accessed at http://www.ietf.org/ietf/1id-abstracts.txt.
The list of Internet-Draft Shadow Directories can be accessed at http://www.ietf.org/shadow.html.
This Internet-Draft will expire on September 3, 2007.
Copyright © The IETF Trust (2007).
The Dynamic MANET On-demand (DYMO) routing protocol is intended for use by mobile nodes in wireless, multihop networks. It offers adaptation to changing network topology and determines unicast routes between nodes within the network in an on-demand fashion.
2. Applicability Statement
4. Data Structures
4.1. Route Table Entry
4.2. DYMO Messages
4.2.1. Generalized MANET Packet and Message Structure
4.2.2. Routing Messages (RM) - RREQ & RREP
4.2.3. Route Error (RERR)
5. Detailed Operation
5.1. DYMO Sequence Numbers
5.1.1. Maintaining A Node's Own Sequence Number
5.1.2. Numerical Operations on OwnSeqNum
5.1.3. OwnSeqNum Rollover
5.1.4. Actions After OwnSeqNum Loss
5.2. DYMO Routing Table Operations
5.2.1. Judging Routing Information's Usefulness
5.2.2. Creating or Updating a Route Table Entry with New Routing Information
5.2.3. Route Table Entry Timeouts
5.3. Routing Messages
5.3.1. RREQ Creation
5.3.2. RREP Creation
5.3.3. Intermediate Node RREP Creation
5.3.4. RM Processing
5.3.5. Adding Additional Routing Information to a RM
5.4. Route Discovery
5.5. Route Maintenance
5.5.1. Active Link Monitoring
5.5.2. Updating Route Lifetimes during Packet Forwarding
5.5.3. Route Error Generation
5.5.4. Route Error Processing
5.6. Unknown Message & TLV Types
5.7. Advertising Network Addresses
5.8. Simple Internet Attachment and Gatewaying
5.9. Multiple Interfaces
5.10. Packet/Message Generation Limits
6. Configuration Parameters and Other Administrative Options
7. IANA Considerations
7.1. DYMO Message Type Specification
7.2. Packet TLV Type Specification
7.3. Address Block TLV Specification
8. Security Considerations
10.1. Normative References
10.2. Informative References
§ Authors' Addresses
§ Intellectual Property and Copyright Statements
The Dynamic MANET On-demand (DYMO) routing protocol enables reactive, multihop routing between participating nodes that wish to communicate. The basic operations of the DYMO protocol are route discovery and route management. During route discovery, the originating node initiates dissemination of a Route Request (RREQ) throughout the network to find a route to the target node. During this hop-by-hop dissemination process, each intermediate node records a route to the originating node. When the target node receives the RREQ, it responds with a Route Reply (RREP) sent hop-by-hop toward the originating node. Each node that receives the RREP records a route to the target node, and then the RREP is unicast hop-by-hop toward the originating node. When the originating node receives the RREP, routes have then been established between the originating node and the target node in both directions.
In order to react to changes in the network topology nodes maintain their routes and monitor links over which traffic is moving. When a data packet is received for forwarding if a route for the destination is not known or the route is broken, then the source of the packet is notified. A Route Error (RERR) is sent to the packet source to indicate the current route to a particular destination is broken. When the source receives the RERR, it knows that it must perform route discovery if it still has packets to deliver to that destination.
DYMO uses sequence numbers to ensure loop freedom [Perkins99] (Perkins, C. and E. Belding-Royer, “Ad hoc On-Demand Distance Vector (AODV) Routing,” February 1999.). Sequence numbers enable nodes to determine the order of DYMO route discovery messages, thereby avoiding use of stale routing information.
The DYMO routing protocol is designed for stub mobile ad hoc networks. DYMO handles a wide variety of mobility patterns by dynamically determining routes on-demand. DYMO also handles a wide variety of traffic patterns. In large networks DYMO is best suited for traffic scenarios where nodes communicate with only a portion of other the nodes.
DYMO is applicable to memory constrained devices, since little routing state needs to be maintained in each node. Only routing information related to active sources and destinations must be maintained, in contrast to other routing protocols that require routing information to all nodes within the autonomous system be maintained.
The routing algorithm in DYMO may be operated at layers other than the network layer, using layer-appropriate addresses. Only modification of the packet format is required. The routing algorithm need not change. Note that, using the DYMO algorithm with message formats (other than those specified in this document) will not be interoperable.
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in RFC 2119 (Bradner, S., “Key words for use in RFCs to Indicate Requirement Levels,” March 1997.) [RFC2119].
Additionally, this document uses some terminology from [I‑D.ietf‑manet‑packetbb] (Clausen, T., “Generalized MANET Packet/Message Format,” January 2007.).
This document defines the following terminology:
- DYMO Sequence Number (SeqNum)
- A DYMO Sequence Number is maintained by each node. This sequence number is used by other nodes to identify the order of routing information generated by a node and to ensure loop-free routes.
- Forwarding Route
- A route that is used to forward data packets. Forwarding routes are generally maintained in a forwarding information base (FIB) or the kernel forwarding/routing table.
- Hop Count (HopCnt)
- The number of IP hops a message or piece of information has traversed.
- Originating Node (OrigNode)
- The originating node is the node that created a DYMO Message in an effort to disseminate some information. The originating node is also referred to as a particular message's originator.
- Route Error (RERR)
- A node generates and disseminates a RERR to indicate that it does not have forwarding route to a one or more particular addresses.
- Route Reply (RREP)
- A RREP is used to disseminate routing information about the RREP OrigNode to the TargetNode and the nodes between them.
- Route Request (RREQ)
- A node (the RREQ OrigNode) generates a RREQ to discover a valid route to a particular destination address, called the RREQ TargetNode. When a node processes a RREQ, it learns routing information on how to reach the RREQ OrigNode.
- Target Node (TargetNode)
- The TargetNode is the ultimate destination of a message.
- This Node (ThisNode)
- ThisNode corresponds to the node currently performing a calculation or processing a message.
- Type-Length-Value structure (TLV)
- A generic way to represent information, see [I‑D.ietf‑manet‑packetbb] (Clausen, T., “Generalized MANET Packet/Message Format,” January 2007.).
- Unreachable Node (UnreachableNode)
- An UnreachableNode is a node for which ThisNode does not have a forwarding route.
The route table entry is a conceptual data structure. Implementations may use any internal representation that conforms to the semantics of a route as specified in this document.
Conceptually, a route table entry has the following fields:
- The IP destination address of the node associated with the routing table entry.
- The DYMO SeqNum associated with this routing information.
- The IP address of the next node on the path toward the Route.Address.
- The interface used to send packets toward the Route.Address.
- A flag indicating whether this Route is broken. This flag is set if the next hop becomes unreachable or in response to processing a RERR (see Section 5.5.4 (Route Error Processing)).
The following fields are optional:
- The number of intermediate node hops traversed before reaching the Route.Address node. Route.HopCnt assists in determining whether received routing information is better than existing known information.
- Indicates that the associated address is a network address, rather than a host address. The value is the length of the netmask/prefix. If an address block does not have an associated PREFIX_LENGTH TLV [I‑D.ietf‑manet‑packetbb] (Clausen, T., “Generalized MANET Packet/Message Format,” January 2007.), the prefix may be considered to have a prefix length equal to the address length (in bits).
Not including optional information may cause performance degradation, but it will not cause the protocol to operate incorrectly otherwise.
In addition to a route table data structure, each route table entry may have several timers associated with the information. These timers/timeouts are discussed in Section 5.2.3 (Route Table Entry Timeouts).
When describing DYMO protocol messages, it is necessary to refer to fields in several distinct parts of the overall packet. These locations include the IP or IPv6 header, the UDP header, and fields from [I‑D.ietf‑manet‑packetbb] (Clausen, T., “Generalized MANET Packet/Message Format,” January 2007.). This document uses the following notation conventions. Information found in the table.
|Information Location||Notational Prefix|
|packetbb message header||MsgHdr.|
|packetbb message TLV||MsgTLV.|
|packetbb address blocks||AddBlk.|
|packetbb address block TLV||AddTLV.|
| Table 1 |
DYMO messages conform to the generalized packet and message format as described in [I‑D.ietf‑manet‑packetbb] (Clausen, T., “Generalized MANET Packet/Message Format,” January 2007.). Here is a brief description of the format. A packet is made up of messages. A message is made up of a message header, message TLV block, and zero or more address blocks. Each of the address blocks may also have an associated address TLV block.
All DYMO messages specified in this document are sent using UDP to the destination port TBD.
Most DYMO messages are sent with the IP destination address set to the link local multicast address LL_ALL_MANET_ROUTER unless otherwise stated. Unicast DYMO messages specified in this document are sent with the IP destination set to the Route.NextHopAddress of the route to the TargetNode.
The IP TTL (IP Hop Limit) field for DYMO messages is set to one (1) for all messages specified in this document.
The length of an IP address (32 bits for IPv4 and 128 bits for IPv6) inside a DYMO message depends on the IP packet header containing the DYMO message/packet. For example, if the IP header uses IPv6 addresses then all messages and addresses contained in the payload use IPv6 addresses. In the case of mixed IPv6 and IPv4 addresses, IPv4 addresses are carried in IPv6 as specified in [RFC4291] (Hinden, R. and S. Deering, “IP Version 6 Addressing Architecture,” February 2006.).
Routing Messages (RMs) are used to disseminate routing information. There are two DYMO message types that are considered to be routing messages (RMs): RREQ and RREP. They contain very similar information and function, but have slightly different processing rules. The main difference between the two messages is that RREQ messages solicit a RREP, whereas a RREP is the response to RREQ.
RM creation and processing are described in Section 5.3 (Routing Messages).
A RM requires the following information:
- The IP address of the packet destination. For RREQ the IP.DestinationAddress is set to LL_ALL_MANET_ROUTERS. For RREP the IP.DestinationAddress is set to the NextHopAddress toward the TargetNode.
- The UDP destination port is set to TBD.
- The remaining number of hops this message is allowed to traverse.
- The IP address of the message TargetNode. In a RREQ the TargetNode is the destination for which a forwarding route does not exist and route discovery is being performed. In a RREP the target node is the RREQ OrigNode. The TargetNode address is the first address in the routing message.
- The IP address of the OrigNode. This address is in an address block and not in the message header to allow for address compression and additional AddTLVs. This address is the second address in the message for RREQ.
- The DYMO sequence number of the OrigNode.
A RM may optionally include the following information:
- The last known DYMO sequence number of the TargetNode.
- The last known HopCnt to the TargetNode.
- The IP address of an additional node that can be reached via the node adding this information. Each AdditionalNode.Address must have an associated SeqNum in the address TLV block.
- The DYMO sequence number of an additional intermediate node's routing information.
- The number of IP hops to reach the associated Node.Address. This field is incremented at each intermediate hop, for each node except the TargetNode's HopCnt information.
- The Node.Address is a network address with a particular prefix length.
Example IPv4 RREQ
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 IP Header +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | IP.DestinationAddress=LL_ALL_MANET_ROUTERS | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ... UDP Header +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Destination Port=TBD | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ... Message Header +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | RREQ-type | Resv |0|0|1| msg-size=23 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | msg-hoplimit | msg-hopcnt | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ... Message Body - Message TLV Block +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | msg-tlv-block-size=0 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Message Body - Address Block +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ |Number Addrs=2 |0|HeadLength=3 | Head : +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ : (cont) | Target.Tail | Orig.Tail | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ Message Body - Address Block TLV Block +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | tlv-block-size=6 |DYMOSeqNum-type|Resv |0|1|0|0|0| +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Index-start=1 | tlv-length=2 | Orig.SeqNum | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Figure 1 |
A RERR message is used to disseminate the information that a route is not available for one or more particular IP addresses.
RERR creation and processing are described in Section 5.5 (Route Maintenance).
A RERR requires the following information:
- The IP address is set to LL_ALL_MANET_ROUTERS.
- The UDP destination port is set to TBD.
- The remaining number of hops this message is allowed to traverse.
- The IP address of an UnreachableNode. Multiple unreachable addresses may be included in a RERR.
A Route Error may optionally include the following information:
- The last known DYMO sequence number of the unreachable node. If a SeqNum for an address is not included, it is assumed to be unknown. This case occurs when a node receives a message to forward for which it does not have any information in its routing table.
Example IPv4 RERR
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 IP Header +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | IP.DestinationAddress=LL_ALL_MANET_ROUTERS | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ... UDP Header +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Destination Port=TBD | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ... Message Header +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | RERR-type | Resv |0|0|1| msg-size=16 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | msg-hoplimit | msg-hopcnt | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ ... Message Body +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | msg-tlv-block-size=0 |Number Addrs=1 |1|HeadLength=4 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | Unreachable.Address | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+ | TLV-blk-size=0 | +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
| Figure 2 |
DYMO sequence numbers allow nodes to judge the freshness of routing information and ensure loop freedom.
DYMO requires that each node in the network to maintain its own DYMO sequence number (OwnSeqNum), a 16-bit unsigned integer. The circumstances for ThisNode to incrementing its OwnSeqNum are described in Section 5.3 (Routing Messages).
When ThisNode increments its OwnSeqNum (as described in Section 5.3 (Routing Messages)) it MUST do so by treating the sequence number value as an unsigned number.
Note: The sequence number zero (0) is reserved.
If the sequence number has been assigned to be the largest possible number representable as a 16-bit unsigned integer (i.e., 65535), then the sequence number is set to 256 when incremented. Setting the sequence number to 256 allows other nodes to detect that the number has rolled over and the node has not lost its sequence number.
A node should maintain its sequence number in persistent storage, between reboots.
If a node's OwnSeqNum is lost, it must take certain actions to avoid creating routing loops. To prevent this possibility after OwnSeqNum loss a node MUST wait for at least ROUTE_DELETE_TIMEOUT before fully participating in the DYMO routing protocol. If a DYMO control message is received during this waiting period, the node SHOULD process it normally but MUST not transmit or retransmit any DYMO messages. If a data packet is received for forwarding to another destination during this waiting period, the node MUST generate a RERR message indicating that this route is not available and reset its waiting timeout. At the end of the waiting period a node sets its OwnSeqNum to one (1).
The longest a node must wait is ROUTE_AGE_MAX_TIMEOUT. At the end of the maximum waiting period a node sets its OwnSeqNum to one (1) and begins participating.
Given a route table entry (Route.SeqNum, Route.HopCnt, and Route.Broken) and new incoming routing information for a particular node in a RM (Node.SeqNum, Node.HopCnt, and RM message type - RREQ/RREP), the quality of the new routing information is evaluated to determine its usefulness. Incoming routing information is classified as follows:
- 1. Stale
- If Node.SeqNum - Route.SeqNum < 0 (using signed 16-bit arithmetic) the incoming information is stale. Using stale routing information is not allowed, since doing so might result in routing loops.(Node.SeqNum - Route.SeqNum < 0)
- 2. Loop-possible
- If Node.SeqNum == Route.SeqNum the incoming information may cause loops if used; in this case additional information must be examined. If Route.HopCnt or Node.HopCnt is unknown or zero (0), then the routing information is loop-possible. If Node.HopCnt > Route.HopCnt + 1, then the routing information is loop-possible. Using loop-possible routing information is not allowed, otherwise routing loops may be formed.(Node.SeqNum == Route.SeqNum) AND ((Node.HopCnt is unknown) OR (Route.HopCnt is unknown) OR (Node.HopCnt > Route.HopCnt + 1))
- 3. Inferior
- If Node.SeqNum == Route.SeqNum the incoming information may be inferior; additional information must be examined. If Node.HopCnt >= to Route.HopCnt, the current route is not Broken, and the message is a RREQ, then the new information is inferior. This rule will stop RREQ propagation if the HopCnt is not shorter. If Node.HopCnt > Route.HopCnt + 1, the current route is not Broken and the message is RREP, then the new information is inferior. This rule will stop RREP propagation if the information is inferior. Inferior routes will not cause routing loops if introduced, but should not be used since better information is already available.(Node.SeqNum == Route.SeqNum) AND (Route.Broken == false) AND ((Node.HopCnt >= Route.HopCnt) AND (RM is RREQ)) OR ((Node.HopCnt > Route.HopCnt + 1) AND (RM is RREP)))
- 4. Superior
- Incoming routing information that does not match any of the above criteria is loop-free and better than the information existing in the routing table. This type of information is used to update the routing table. For completeness, the following other cases are possible:(Node.SeqNum - Route.SeqNum > 0) OR ((Node.SeqNum == Route.Seqnum) AND (Node.HopCnt <= Route.HopCnt + 1) AND ((Route.Broken == true) OR ((Node.HopCnt < Route.HopCnt) AND (RM is RREQ))))
The route table entry is populated with the following information:
Fields without known values are not populated with any value.
Previous timers for this route table entry are removed. A timer for the minimum delete timeout (ROUTE_AGE_MIN) is set to ROUTE_AGE_MIN_TIMEOUT. A timer to indicate a recently learned route (ROUTE_NEW) is set to ROUTE_NEW_TIMEOUT. A timer for the maximum delete timeout (ROUTE_AGE_MAX). ROUTE_AGE_MAX is set to Node.AddTLV.MaxAge if included; otherwise, ROUTE_AGE_MAX is set to ROUTE_AGE_MAX_TIMEOUT. The usage of these timers and others are described in Section 5.2.3 (Route Table Entry Timeouts).
At this point, a forwarding route should be installed. Afterward, the route can be used to send any queued data packets and forwarding any incoming data packets for Route.Address. This route also fulfills any outstanding route discovery attempts for Node.Address.
When a node transmits a RM, other nodes expect the transmitting node to have a forwarding route to the RM originator. After updating a route table entry, it should be maintained for at least ROUTE_AGE_MIN. Failure to maintain the information might result in lost messages/packets, or in the worst case scenario several duplicate messages.
After the ROUTE_AGE_MIN timeout a route can safely be deleted.
Sequence number information is time sensitive, and must be deleted after a time in order to avoid conflicts due to reboots and rollovers. When a node has lost its sequence number (e.g, due to daemon reboot or node replacement) the node must wait until routing information associated with its IP address and sequence number are no longer maintained by other nodes in the network to ensure loop-free routing.
After the ROUTE_AGE_MAX timeout a route must be deleted. All information about the route is deleted upon ROUTE_AGE_MAX timeout. If a forwarding route exists it is also removed.
As time progresses the likelihood that a route remains intact decreases, if the network nodes are mobile. Maintaining and using old routing information can lead to many DYMO messages and excess route discovery delay.
After the ROUTE_NEW timeout if the route has not been used, a timer for deleting the route (ROUTE_DELETE) is set to ROUTE_DELETE_TIMEOUT.
When a route is used to forward data packets, this timer is set to expire after ROUTE_USED_TIMEOUT. This operation is also discussed in Section 5.5.2 (Updating Route Lifetimes during Packet Forwarding).
If a route has not been used recently, then a timer for ROUTE_DELETE is set to ROUTE_DELETE_TIMEOUT.
As time progresses the likelihood that old routing information is useful decreases, especially if the network nodes are mobile. Therefore, old information should be deleted.
After the ROUTE_DELETE timeout, the routing table entry should be deleted. If a forwarding route exists, it should also be removed.
When a node creates a RREQ it SHOULD increment its OwnSeqNum by one (1) according to the rules specified in Section 5.1.2 (Numerical Operations on OwnSeqNum). Incrementing OwnSeqNum will ensure that all nodes with existing routing information to consider this new information fresh. If the sequence number is not incremented, certain nodes might not consider this information useful if they have superior information already.
First, the node adds the AddBlk.TargetNode.Address to the RREQ.
If a previous value of the TargetNode.SeqNum is known (from a routing table entry), it SHOULD be placed in TargetNode.AddTLV.SeqNum in all but the last RREQ attempt. If a TargetNode.SeqNum is not included, it is assumed to be unknown by processing nodes. This operation ensures that no intermediate nodes reply, and ensures that the TargetNode increments its sequence number.
Similarly, if a previous value of the TargetNode.HopCnt is known, it SHOULD be placed in TargetNode.AddTLV.HopCnt. Otherwise, the TargetNode.AddTLV.HopCnt is not included and assumed unknown by processing nodes.
Next, the node adds AddBlk.OrigNode.Address to the RM and the OrigNode.AddTLV.SeqNum (OwnSeqNum) in an address block TLV. The OrigNode.Address is this node's address, and it must be a routable IP address. This information will be used by nodes to create a route toward the OrigNode and enable delivery of a RREP.
If OrigNode.HopCnt is included it is set to zero (0).
The MsgHdr.HopCnt is set to zero (0). The MsgHdr.HopLimit should be set to MAX_HOPLIMIT, but may be set smaller. For RREQ, the MsgHdr.HopLimit may be set in accordance with an expanding ring search as described in (Perkins, C., Belding-Royer, E., and S. Das, “Ad hoc On-Demand Distance Vector (AODV) Routing,” July 2003.) [RFC3561] to limit the RREQ propagation to a subset of the network and possibly reduce route discovery overhead.
The IP.DestinationAddress for RREQ is set to LL_ALL_MANET_ROUTERS.
When ThisNode creates a RREP, if the ThisNode.SeqNum was not included in the RREQ it SHOULD increment its OwnSeqNum by one (1) according to the rules specified in Section 5.1.2 (Numerical Operations on OwnSeqNum).
If ThisNode.SeqNum is included in the RM and ThisNode.SeqNum from the RM is less than OwnSeqNum, OwnSeqNum SHOULD be incremented by one (1) according to the rules specified in Section 5.1.2 (Numerical Operations on OwnSeqNum).
If OwnSeqNum is not incremented the routing information might be considered stale. In this case, the RREP would not reach the RREP Target.
Since RREP messages are not broadcast throughout the network, changes to the sequence number are unlikely to reach most nodes in the network. Therefore, it is important to avoid incrementing the sequence number when issuing a RREP is an important mechanism to reduce the unnecessary devaluing of good routing information, and the ability to issue intermediate node replies. When intermediate node replies are coupled with expanding ring search, route discovery cost can be reduced.
ThisNode first adds the RREP AddBlk.TargetNode.Address to the RREP. The TargetNode is the ultimate destination of this RREP.
ThisNode then adds the RREP AddBlk.OrigNode.Address (ThisNode.Address) and the RREP OrigNode.AddTLV.SeqNum (OwnSeqNum) to the RREP.
Other AddTLVs in the RREP for the OrigNode and TargetNode SHOULD be included and set accordingly. If OrigNode.HopCnt is included it is set to zero (0).
The MsgHdr.HopCnt is set to zero (0). The MsgHdr.HopLimit is set to MAX_HOPLIMIT.
The IP.DestinationAddress for RREP is set to the IP address of the Route.NextHopAddress for the route to the RREP TargetNode.
Sometimes a node other than the TargetNode (call it an "intermediate node") has routing information that can satisfy an incoming RREQ. When an intermediate node originates a RREP in response to a RREQ, it sends the RREP to the RREQ OrigNode with additional routing information (Address, SeqNum, etc.) about the RREQ TargetNode. Appending additional routing information is described in Section 5.3.5 (Adding Additional Routing Information to a RM).
The Intermediate Node SHOULD also issue a gratuitous RREP to the RREQ TargetNode, so that the RREQ TargetNode receives routing information on how to reach the RREQ OrigNode.
When an intermediate node creates a gratuitous RREP, it sends a RREP to the RREQ TargetNode with additional routing information (Address, SeqNum, etc.) about the RREQ OrigNode.
Before processing a RM, a node checks the IP.Destination to ensure that it is a link local packet.
When a RM is received the MsgHdr.HopLimit is decremented by one (1) and MsgHdr.HopCnt is incremented by one (1).
For each address (except the TargetNode) in the RM that includes AddTLV.HopCnt information, the AddTLV.HopCnt information is incremented by one (1).
Next, this node checks whether its routing table has an entry to the AddBlk.OrigNode.Address using longest-prefix matching (Baker, F., “Requirements for IP Version 4 Routers,” June 1995.) [RFC1812]. If a route does not exist, the new routing information is considered fresh and a new route table entry is created and updated as described in Section 5.2.2 (Creating or Updating a Route Table Entry with New Routing Information). If a route table entry does exists, the new node's information is compared with the route table entry following the procedure described in Section 5.2.1 (Judging Routing Information's Usefulness). If the new node's routing information is considered superior, the route table entry is updated as described in Section 5.2.2 (Creating or Updating a Route Table Entry with New Routing Information).
After processing the OrigNode's routing information, then each address that is not the TargetNode should be considered for creating and updating routes. Creating and updating routes to other nodes can eliminate RREQ for those IP destinations, in the event that data needs to be forwarded to the IP destination(s) in the near future.
For each of the additional addresses considered, if the routing table does not have a matching route using longest-prefix matching, then a route is created and updated as described in Section 5.2.2 (Creating or Updating a Route Table Entry with New Routing Information). If a route table entry exists, the new node's information is compared with the route table entry following the procedure described in Section 5.2.1 (Judging Routing Information's Usefulness). If the new node's routing information is considered superior, the route table entry is updated as described in Section 5.2.2 (Creating or Updating a Route Table Entry with New Routing Information).
If the routing information for an AdditionalNode.Address is not considered superior, then it is removed from the RM. Removing this information ensures that the information is not propagated.
At this point, if the routing information for the OrigNode was not superior then this RM should be discarded and no further processing of this message is performed.
If the ThisNode is the TargetNode and this RM is a RREQ, then ThisNode responds with a RREQ flood (a RREQ addressed to oneself) or a RREP to the RREQ OrigNode (the new RREP's TargetNode). The procedure for issuing a new RREP is described in Section 5.3.2 (RREP Creation). Note: it is important that when creating the RREP, the RREP OrigNode.Address be the same as the RREQ TargetNode.Address, if ThisNode has several addresses. At this point, ThisNode need not perform any more operations for this RM.
If ThisNode is not the TargetNode, this RM is a RREQ, the RREQ contains the TargetNode.AddTLV.SeqNum, and ThisNode has an forwarding route to the TargetNode with a SeqNum (Route.TargetNode.SeqNum) greater than or equal to the RREQ TargetNode.AddTLV.SeqNum; then this node MAY respond with an intermediate node RREP. The procedure for performing intermediate node RREP is described in Section 5.3.3 (Intermediate Node RREP Creation). At this point, ThisNode need not perform any more operations for this RM.
After processing a RM or creating a new RM, a node can append additional routing information to the RM, according to the procedure described in Section 5.3.5 (Adding Additional Routing Information to a RM). The additional routing information can help reduce route discoveries at the expense of increased message size.
If this RM's MsgHdr.HopLimit is greater than or equal to one (1), ThisNode is not the TargetNode, AND this RM is a RREQ, then the current RM (altered by the procedure defined above) is sent to the LL_ALL_MANET_ROUTERS IP.DestinationAddress.
If this RM's MsgHdr.HopLimit is greater than or equal to one (1), ThisNode is not the TargetNode, AND this RM is a RREP, then the current RM is sent to the Route.NextHopAddress for the RREP's TargetNode.Address. If no forwarding route exists to Target.Address, then a RERR is issued to the OrigNode of the RREP.
Appending routing information can alleviate route discovery attempts to the nodes whose information is included, if other nodes use this information to update their routing tables.
Nodes can append routing information to a RM. Appending additional routing information can help alleviate future RREQ. This option should be administratively configurable.
Prior to appending its own address to a RM, ThisNode MAY increment its OwnSeqNum as defined in Section 5.1.2 (Numerical Operations on OwnSeqNum). If OwnSeqNum is not incremented the appended routing information might not be considered fresh, when received by nodes with existing routing information. Incrementation of the sequence number when appending information to an RM in transit should be administratively configurable.
If included the Node.HopCnt for ThisNode is included, it is set to zero (0). Additional information about the address(es) can also be appended, such as a PREFIX_LENGTH AddTLV.
A node creates and sends a RREQ (described in Section 5.3.1 (RREQ Creation)) to discover a route to a particular destination (TargetNode) for which it does not currently have a forwarding route.
After issuing a RREQ, the OrigNode waits for a route to be created to the TargetNode. If a route is not created within RREQ_WAIT_TIME, ThisNode may again try to discover a route by issuing another RREQ.
To reduce congestion in a network, repeated attempts at route discovery for a particular TargetNode should utilize an exponential backoff.
For example, the first time a node issues a RREQ, it waits RREQ_WAIT_TIME for a route to the TargetNode. If a route is not found within that time, the node MAY send another RREQ. If a route is not found within two (2) times the current waiting time, another RREQ may be sent, up to a total of RREQ_TRIES. For each additional attempt, the waiting time for the previous RREQ is multiplied by two (2) so that the waiting time conforms to a binary exponential backoff.
Data packets awaiting a route should be buffered at the source. This buffer should have a fixed limited size (BUFFER_SIZE_PACKETS or BUFFER_SIZE_BYTES) and older data packets SHOULD be discarded first.
If a route discovery has been attempted RREQ_TRIES times without receiving a route to the TargetNode, all data packets destined for the corresponding TargetNode are dropped from the buffer and a Destination Unreachable ICMP message should be delivered to the application.
A RERR MUST be issued if a data packet is received and it cannot be delivered to the next hop when no forwarding route exists; RERR generation is described in Section 5.5.3 (Route Error Generation).
In addition to inability to deliver a data packet, a RERR SHOULD be issued immediately after detecting a broken link of an forwarding route to quickly notify nodes that a link break occurred and that certain routes are no longer available. If the route with the broken link has not been used recently (indicated by ROUTE_USED), the RERR SHOULD NOT be generated.
Nodes MUST monitor next hop links on forwarding routes. This monitoring can be accomplished by one or several mechanisms, including:
Upon detecting a link break (or an unreachable next hop) ThisNode must remove the affected forwarding routes (those with an unreachable next hop). ThisNode also flags these routes as Broken. For each broken route a timer for ROUTE_DELETE is set to ROUTE_DELETE_TIMEOUT.
To avoid removing forwarding routes that are being used, a node SHOULD set a timeout (ROUTE_USED) to ROUTE_USED_TIMEOUT for the route to the IP.SourceAddress upon receiving a data packet. If a timer for ROUTE_DELETE is set, it is removed.
To avoid removing forwarding routes that are being used, a node SHOULD set a timeout (ROUTE_USED) to ROUTE_USED_TIMEOUT for the route to the IP.DestinationAddress upon sending a data packet. If a timer for ROUTE_DELETE is set, it is removed.
A RERR informs the IP.SourceAddress or RREP.OrigNode.Address that the route does not exist, and a route is not available through this node.
When creating a new RERR, the address of first UnreachableNode (IP.DestinationAddress from the data packet or RREP.TargetNode.Address) is inserted into an Address Block AddBlk.UnreachableNode.Address. If a value for the UnreachableNode's SeqNum (UnreachableNode.AddTLV.SeqNum) is known, it SHOULD be placed in the RERR. The MsgHdr.HopLimit is set to MAX_HOPLIMIT. The MsgHdr.HopCnt is set to one (1).
Additional UnreachableNodes that require the same unavailable link (routes with the same Route.NextHopAddress and Route.NextHopInterface) SHOULD be added to the RERR, as additional AddBlk.UnreachableNode.Address. The SeqNum if known SHOULD also be included. Appending UnreachableNode information notifies each processing node of additional routes that are no longer available. This option SHOULD be administratively configurable.
If SeqNum information is not known or not included in the RERR, all nodes processing the RERR will assume their routing information associated with the UnreachableNode is no longer valid.
The RERR is sent to the IP.DestinationAddress LL_ALL_MANET_ROUTERS. Sending the RERR to the LL_ALL_MANET_ROUTERS address notifies nearby nodes that might depend on the now broken link.
The packet or message that forced generation of this RERR is discarded.
Before processing a RERR, a node checks the IP.Destination to ensure that it is a link local packet.
When a node processes a RERR, it processes each UnreachableNode's information. The processing node removes the forwarding route and sets the broken flag for each UnreachableNode.Address found using longest prefix matching that meet all of the following conditions:
Each UnreachableNode that did not result in a broken route is removed from the RERR, since propagation of this information will not result in any benefit. Any other information (AddTLVs) associated with the removed address(es) is also removed.
If no UnreachableNode addresses remain in the RERR, no other processing is required and the RERR is discarded.
If this RERR's MsgHdr.HopLimit is greater than one (1) and at least one unreachable node address remains in the RERR, then the updated RERR is sent to the IP.DestinationAddress LL_ALL_MANET_ROUTERS.
If a message with an unknown type is received, the message is discarded.
If a message contains TLVs of an unknown type, a node ignores these during processing. The processing node can remove these TLVs from any resulting transmitted messages. The behavior for unknown TLV types should be administratively configurable.
Any node can advertise a network address by using a PREFIX_LENGTH TLV [I‑D.ietf‑manet‑packetbb] (Clausen, T., “Generalized MANET Packet/Message Format,” January 2007.). Any nodes (other than the advertising node) within the advertised prefix SHOULD NOT participate in the DYMO protocol directly and these nodes MUST be reachable by forwarding packets to the node advertising connectivity. Nodes other than the advertising node that do participate in DYMO must forward the DYMO control packets to the advertising node. For example, A.B.C.1 with a prefix length of 24 indicates all nodes with the matching A.B.C.X are reachable through the node with address A.B.C.1.
Simple Internet attachment consists of a network of MANET nodes connected to the Internet via a single Internet gateway node. The gateway is responsible for responding to RREQs for TargetNodes outside its configured DYMO prefix, as well as delivering packets to destinations outside the MANET.
/--------------------------\ / Internet \ \ / \------------+-------------/ Gateway's | Advertised | A.B.C.X Prefix | +-----+-----+ | DYMO | /------| Internet |------\ / | Gateway | \ / | A.B.C.1 | \ | +-----------+ | | DYMO Region | | | | +------------+ | | | DYMO Node | | | | A.B.C.2 | | | +------------+ | | +------------+ | | | DYMO Node | | | | A.B.C.3 | | \ +------------+ / \ / \-------------------------/
| Figure 7: Simple Internet Attachament Example |
DYMO nodes wishing to be reachable from nodes in the Internet MUST have IP addresses within the gateway's configured and advertised prefix. Given a node with a globally routeable address or care-of address handled by the gateway, the gateway is responsible for routing and forwarding packets received from the Internet destined for nodes inside its MANET.
When nodes within the MANET want to send messages to nodes in the Internet, they simply issue RREQ for those IP.DestinationAddresses. The gateway is responsible for responding to RREQ on behalf of the Internet destinations and maintaining their associated sequence numbers.
For an Internet gateway and other nodes that maintain the sequence number on behalf of other nodes, these routers must be administratively configurable to know the IP addresses for which they must generate DYMO messages and maintain OwnSeqNum.
DYMO may be used with multiple interfaces; therefore, the particular interface over which packets arrive must be known whenever a packet is received. Whenever a new route is created, the interface through which the Route.Address can be reached is also recorded in the route table entry.
When multiple interfaces are available, a node transmitting a packet with IP.DestinationAddress set to LL_ALL_MANET_ROUTERS SHOULD send the packet on all interfaces that have been configured for DYMO operation.
To avoid congestion, a node's rate of packet/message generation should be limited. The rate and algorithm for limiting messages is left to the implementor and should be administratively configurable. Messages should be discarded in the following order of preferences RREQ, RREP, and finally RERR.
Suggested Parameter Values
|ROUTE_DELETE_TIMEOUT||2 * ROUTE_TIMEOUT|
|ROUTE_RREQ_WAIT_TIME||2 * NET_TRAVERSAL_TIME|
| Table 2 |
These suggested values work well for small and medium well connected networks with infrequent topology changes. These parameters should be administratively configurable for the network where DYMO is used. Ideally, for networks with frequent topology changes the DYMO parameters should be adjusted using either experimentally determined values or dynamic adaptation. For example, in networks with infrequent topology changes ROUTE_USED_TIMEOUT may be set to a much larger value.
In addition to the parameters above several administrative options exist. The following table enumerates several of the options and suggested values.
Suggested Options Settings
|RESPONSIBLE_ADDRESSES||Self or Prefix|
|APPEND_ADDRESS||Yes - RREQ & RREP|
|APPEND_OWN_ADDRESS_INCREMENT_SEQNUM||Yes for RREQ|
|BUFFER_SIZE_BYTES||1500 * BUFFER_SIZE_PACKETS|
| Table 3 |
DYMO requires a UDP port number to carry protocol packets - TBD. DYMO also requires the link-local multicast address LL_ALL_MANET_ROUTERS; IPv4 TBD, IPv6 TBD [I‑D.chakeres‑manet‑iana] (Chakeres, I., “MANET IANA Needs,” October 2006.).
This section specifies several messages types, message tlv-types, and address tlv-types.
Future types will be allocated using standard actions as described in (Narten, T. and H. Alvestrand, “Guidelines for Writing an IANA Considerations Section in RFCs,” October 1998.) [RFC2434].
DYMO Message Types
|Route Request (RREQ)||10 - TBD|
|Route Reply (RREP)||11 - TBD|
|Route Error (RERR)||12 - TBD|
| Table 4 |
Packet TLV Types
|Unicast Response Request||10 - TBD||0||Indicates to the processing node that the previous hop (IP.SourceAddress) expects a unicast message within UNICAST_MESSAGE_SENT_TIMEOUT. Any unicast packet will serve this purpose, and it MAY be an ICMP REPLY message. If a message is not sent, then the previous hop may assume that the link is unidirectional and may blacklist the link to this node.|
| Table 5 |
Address Block TLV Types
|DYMOSeqNum||10 - TBD||16 bits||The DYMO sequence num associated with this address. The sequence number may be the last known sequence number.|
|HopCount||11 - TBD||8 bits||The number of hops traversed by the information associated with this address.|
|MaxAge||12 - TBD||Any length||The maximum number of milliseconds that the associated routing information can be kept before being deleted.|
| Table 6 |
Currently, DYMO does not specify any special security measures. Routing protocols, however, are prime targets for impersonation attacks. In networks where the node membership is not known, it is difficult to determine the occurrence of impersonation attacks, and security prevention techniques are difficult at best. However, when the network membership is known and there is a danger of such attacks, DYMO messages must be protected by the use of authentication techniques, such as those involving generation of unforgeable and cryptographically strong message digests or digital signatures. While DYMO does not place restrictions on the authentication mechanism used for this purpose, IPsec Authentication Message (AH) is an appropriate choice for cases where the nodes share an appropriate security association that enables the use of AH.
In particular, RM messages SHOULD be authenticated to avoid creation of spurious routes to a destination. Otherwise, an attacker could masquerade as that destination and maliciously deny service to the destination and/or maliciously inspect and consume traffic intended for delivery to the destination. RERR messages SHOULD be authenticated in order to prevent malicious nodes from disrupting active routes between communicating nodes.
If the mobile nodes in the ad hoc network have pre-established security associations, the purposes for which the security associations are created should include that of authorizing the processing of DYMO control packets. Given this understanding, the mobile nodes should be able to use the same authentication mechanisms based on their IP addresses as they would have used otherwise.
DYMO is a descendant of the design of previous MANET reactive protocols, especially AODV [RFC3561] (Perkins, C., Belding-Royer, E., and S. Das, “Ad hoc On-Demand Distance Vector (AODV) Routing,” July 2003.) and DSR [Johnson96] (Johnson, D. and D. Maltz, “Dynamic Source Routing (DSR) in Ad hoc Networks,” 1996.). Changes to previous MANET reactive protocols stem from research and implementation experiences. Thanks to Elizabeth Belding-Royer for her long time authorship of DYMO. Additional thanks to Luke Klein-Berndt, Pedro Ruiz, Fransisco Ros, Koojana Kuladinithi, Ramon Caceres, Thomas Clausen, Christopher Dearlove, Seung Yi, and Romain Thouvenin for reviewing of DYMO, as well as several specification suggestions.
|[I-D.ietf-manet-packetbb]||Clausen, T., “Generalized MANET Packet/Message Format,” draft-ietf-manet-packetbb-03 (work in progress), January 2007.|
|[RFC1812]||Baker, F., “Requirements for IP Version 4 Routers,” RFC 1812, June 1995.|
|[RFC2119]||Bradner, S., “Key words for use in RFCs to Indicate Requirement Levels,” BCP 14, RFC 2119, March 1997 (TXT, HTML, XML).|
|[RFC2434]||Narten, T. and H. Alvestrand, “Guidelines for Writing an IANA Considerations Section in RFCs,” BCP 26, RFC 2434, October 1998 (TXT, HTML, XML).|
|[RFC4291]||Hinden, R. and S. Deering, “IP Version 6 Addressing Architecture,” RFC 4291, February 2006.|
|[I-D.chakeres-manet-iana]||Chakeres, I., “MANET IANA Needs,” draft-chakeres-manet-iana-02 (work in progress), October 2006.|
|[I-D.ietf-manet-nhdp]||Clausen, T., “MANET Neighborhood Discovery Protocol (NHDP),” draft-ietf-manet-nhdp-01 (work in progress), February 2007.|
|[Johnson96]||Johnson, D. and D. Maltz, “Dynamic Source Routing (DSR) in Ad hoc Networks,” In Mobile Computing, Chapter 5, pp. 153-181, 1996.|
|[Perkins99]||Perkins, C. and E. Belding-Royer, “Ad hoc On-Demand Distance Vector (AODV) Routing,” Proceedings of the 2nd IEEE Workshop on Mobile Computing Systems and Applications, New Orleans, LA, pp. 90-100, February 1999.|
|[RFC3561]||Perkins, C., Belding-Royer, E., and S. Das, “Ad hoc On-Demand Distance Vector (AODV) Routing,” RFC 3561, July 2003.|
|Boeing Phantom Works|
|The Boeing Company|
|P.O. Box 3707 Mailcode 7L-49|
|Seattle, WA 98124-2207|
|Charles E. Perkins|
|Palo Alto Systems Research Center|
|975 Page Mill Road, Suite 200|
|Palo Alto, CA 94304-1003|
Copyright © The IETF Trust (2007).
This document is subject to the rights, licenses and restrictions contained in BCP 78, and except as set forth therein, the authors retain all their rights.
This document and the information contained herein are provided on an “AS IS” basis and THE CONTRIBUTOR, THE ORGANIZATION HE/SHE REPRESENTS OR IS SPONSORED BY (IF ANY), THE INTERNET SOCIETY, THE IETF TRUST AND THE INTERNET ENGINEERING TASK FORCE DISCLAIM ALL WARRANTIES, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO ANY WARRANTY THAT THE USE OF THE INFORMATION HEREIN WILL NOT INFRINGE ANY RIGHTS OR ANY IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
The IETF takes no position regarding the validity or scope of any Intellectual Property Rights or other rights that might be claimed to pertain to the implementation or use of the technology described in this document or the extent to which any license under such rights might or might not be available; nor does it represent that it has made any independent effort to identify any such rights. Information on the procedures with respect to rights in RFC documents can be found in BCP 78 and BCP 79.
Copies of IPR disclosures made to the IETF Secretariat and any assurances of licenses to be made available, or the result of an attempt made to obtain a general license or permission for the use of such proprietary rights by implementers or users of this specification can be obtained from the IETF on-line IPR repository at http://www.ietf.org/ipr.
The IETF invites any interested party to bring to its attention any copyrights, patents or patent applications, or other proprietary rights that may cover technology that may be required to implement this standard. Please address the information to the IETF at firstname.lastname@example.org.
Funding for the RFC Editor function is provided by the IETF Administrative Support Activity (IASA).