このページは 2024-01 に最終更新され、ルーターバージョン 0.9.61 について正確です。

DEPRECATED - SSU has been replaced by SSU2. SSU support was removed from i2pd in release 2.44.0 (API 0.9.56) 2022-11. SSU support was removed from Java I2P in release 2.4.0 (API 0.9.61) 2023-12.

SSU (also called "UDP" in much of the I2P documentation and user interfaces) is one of two transports currently implemented in I2P. The other is NTCP2. Support for NTCP has been removed.

SSU was introduced in I2P release 0.6. In a standard I2P installation, the router uses both NTCP and SSU for outbound connections. SSU-over-IPv6 is supported as of version 0.9.8.

SSU is called "semireliable" because it will repeatedly retransmit unacknowledged messages, but only up to a maximum number of times. After that, the message is dropped.

SSU Services

Like the NTCP transport, SSU provides reliable, encrypted, connection-oriented, point-to-point data transport. Unique to SSU, it also provides IP detection and NAT traversal services, including:

  • Cooperative NAT/Firewall traversal using introducers
  • Local IP detection by inspection of incoming packets and peer testing
  • Communication of firewall status and local IP, and changes to either to NTCP
  • Communication of firewall status and local IP, and changes to either, to the router and the user interface

Router Address Specification

The following properties are stored in the network database.

  • Transport name: SSU
  • caps: [B,C,4,6] See below.
  • host: IP (IPv4 or IPv6). Shortened IPv6 address (with "::") is allowed. May or may not be present if firewalled. Host names were previously allowed, but are deprecated as of release 0.9.32. See proposal 141.
  • iexp[0-2]: Expiration of this introducer. ASCII digits, in seconds since the epoch. Only present if firewalled, and introducers are required. Optional (even if other properties for this introducer are present). As of release 0.9.30, proposal 133.
  • ihost[0-2]: Introducer's IP (IPv4 or IPv6). Host names were previously allowed, but are deprecated as of release 0.9.32. See proposal 141. Shortened IPv6 address (with "::") is allowed. Only present if firewalled, and introducers are required. See below.
  • ikey[0-2]: Introducer's Base 64 introduction key. See below. Only present if firewalled, and introducers are required. See below.
  • iport[0-2]: Introducer's port 1024 - 65535. Only present if firewalled, and introducers are required. See below.
  • itag[0-2]: Introducer's tag 1 - (2**32 - 1) ASCII digits. Only present if firewalled, and introducers are required. See below.
  • key: Base 64 introduction key. See below.
  • mtu: Optional. Default and max is 1484. Min is 620. Must be present for IPv6, where the min is 1280 and the max is 1488 (max was 1472 prior to version 0.9.28). IPv6 MTU must be a multiple of 16. (IPv4 MTU + 4) must be a multiple of 16. See below.
  • port: 1024 - 65535 May or may not be present if firewalled.

Protocol Details

Congestion control

SSU's need for only semireliable delivery, TCP-friendly operation, and the capacity for high throughput allows a great deal of latitude in congestion control. The congestion control algorithm outlined below is meant to be both efficient in bandwidth as well as simple to implement.

Packets are scheduled according to the router's policy, taking care not to exceed the router's outbound capacity or to exceed the measured capacity of the remote peer. The measured capacity operates along the lines of TCP's slow start and congestion avoidance, with additive increases to the sending capacity and multiplicative decreases in face of congestion. Unlike for TCP, routers may give up on some messages after a given period or number of retransmissions while continuing to transmit other messages.

The congestion detection techniques vary from TCP as well, since each message has its own unique and nonsequential identifier, and each message has a limited size - at most, 32KB. To efficiently transmit this feedback to the sender, the receiver periodically includes a list of fully ACKed message identifiers and may also include bitfields for partially received messages, where each bit represents the reception of a fragment. If duplicate fragments arrive, the message should be ACKed again, or if the message has still not been fully received, the bitfield should be retransmitted with any new updates.

The current implementation does not pad the packets to any particular size, but instead just places a single message fragment into a packet and sends it off (careful not to exceed the MTU).

MTU

As of router version 0.8.12, two MTU values are used for IPv4: 620 and 1484. The MTU value is adjusted based on the percentage of packets that are retransmitted.

For both MTU values, it is desirable that (MTU % 16) == 12, so that the payload portion after the 28-byte IP/UDP header is a multiple of 16 bytes, for encryption purposes.

For the small MTU value, it is desirable to pack a 2646-byte Variable Tunnel Build Message efficiently into multiple packets; with a 620-byte MTU, it fits into 5 packets with nicely.

Based on measurements, 1492 fits nearly all reasonably small I2NP messages (larger I2NP messages may be up to 1900 to 4500 bytes, which isn't going to fit into a live network MTU anyway).

The MTU values were 608 and 1492 for releases 0.8.9 - 0.8.11. The large MTU was 1350 prior to release 0.8.9.

The maximum receive packet size is 1571 bytes as of release 0.8.12. For releases 0.8.9 - 0.8.11 it was 1535 bytes. Prior to release 0.8.9 it was 2048 bytes.

As of release 0.9.2, if a router's network interface MTU is less than 1484, it will publish that in the network database, and other routers should honor that when a connection is established.

For IPv6, the minimum MTU is 1280. The IPv6 IP/UDP header is 48 bytes, so we use an MTU where (MTN % 16 == 0), which is true for 1280. The maximum IPv6 MTU is 1488. (max was 1472 prior to version 0.9.28).

Message Size Limits

While the maximum message size is nominally 32KB, the practical limit differs. The protocol limits the number of fragments to 7 bits, or 128. The current implementation, however, limits each message to a maximum of 64 fragments, which is sufficient for 64 * 534 = 33.3 KB when using the 608 MTU. Due to overhead for bundled LeaseSets and session keys, the practical limit at the application level is about 6KB lower, or about 26KB. Further work is necessary to raise the UDP transport limit above 32KB. For connections using the larger MTU, larger messages are possible.

Idle Timeout

Idle timeout and connection close is at the discretion of each endpoint and may vary. The current implementation lowers the timeout as the number of connections approaches the configured maximum, and raises the timeout when the connection count is low. The recommended minimum timeout is two minutes or more, and the recommended maximum timeout is ten minutes or more.

All encryption used is AES256/CBC with 32 byte keys and 16 byte IVs. When Alice originates a session with Bob, the MAC and session keys are negotiated as part of the DH exchange, and are then used for the HMAC and encryption, respectively. During the DH exchange, Bob's publicly knowable introKey is used for the MAC and encryption.

Both the initial message and the subsequent reply use the introKey of the responder (Bob) - the responder does not need to know the introKey of the requester (Alice). The DSA signing key used by Bob should already be known to Alice when she contacts him, though Alice's DSA key may not already be known by Bob.

Upon receiving a message, the receiver checks the "from" IP address and port with all established sessions - if there are matches, that session's MAC keys are tested in the HMAC. If none of those verify or if there are no matching IP addresses, the receiver tries their introKey in the MAC. If that does not verify, the packet is dropped. If it does verify, it is interpreted according to the message type, though if the receiver is overloaded, it may be dropped anyway.

If Alice and Bob have an established session, but Alice loses the keys for some reason and she wants to contact Bob, she may at any time simply establish a new session through the SessionRequest and related messages. If Bob has lost the key but Alice does not know that, she will first attempt to prod him to reply, by sending a DataMessage with the wantReply flag set, and if Bob continually fails to reply, she will assume the key is lost and reestablish a new one.

For the DH key agreement, RFC3526 2048bit MODP group (#14) is used:

  p = 2^2048 - 2^1984 - 1 + 2^64 * { [2^1918 pi] + 124476 }
  g = 2

These are the same p and g used for I2P's ElGamal encryption.

Replay prevention

Replay prevention at the SSU layer occurs by rejecting packets with exceedingly old timestamps or those which reuse an IV. To detect duplicate IVs, a sequence of Bloom filters are employed to "decay" periodically so that only recently added IVs are detected.

The messageIds used in DataMessages are defined at layers above the SSU transport and are passed through transparently. These IDs are not in any particular order - in fact, they are likely to be entirely random. The SSU layer makes no attempt at messageId replay prevention - higher layers should take that into account.

Addressing

To contact an SSU peer, one of two sets of information is necessary: a direct address, for when the peer is publicly reachable, or an indirect address, for using a third party to introduce the peer. There is no restriction on the number of addresses a peer may have.

    Direct: host, port, introKey, options
  Indirect: tag, relayhost, port, relayIntroKey, targetIntroKey, options

Each of the addresses may also expose a series of options - special capabilities of that particular peer. For a list of available capabilities, see below.

The addresses, options, and capabilities are published in the network database.

Direct Session Establishment

Direct session establishment is used when no third party is required for NAT traversal. The message sequence is as follows:

Connection establishment (direct)

Alice connects directly to Bob. IPv6 is supported as of version 0.9.8.

        Alice                         Bob
    SessionRequest --------------------->
          <--------------------- SessionCreated
    SessionConfirmed ------------------->
          <--------------------- DeliveryStatusMessage
          <--------------------- DatabaseStoreMessage
    DatabaseStoreMessage --------------->
    Data <---------------------------> Data

After the SessionConfirmed message is received, Bob sends a small DeliveryStatus message as a confirmation. In this message, the 4-byte message ID is set to a random number, and the 8-byte "arrival time" is set to the current network-wide ID, which is 2 (i.e. 0x0000000000000002).

After the status message is sent, the peers usually exchange DatabaseStore messages containing their RouterInfos, however, this is not required.

It does not appear that the type of the status message or its contents matters. It was originally added becasue the DatabaseStore message was delayed several seconds; since the store is now sent immediately, perhaps the status message can be eliminated.

序章

Introduction keys are delivered through an external channel (the network database), where they have traditionally been identical to the router Hash through release 0.9.47, but may be random as of release 0.9.48. They must be used when establishing a session key. For the indirect address, the peer must first contact the relayhost and ask them for an introduction to the peer known at that relayhost under the given tag. If possible, the relayhost sends a message to the addressed peer telling them to contact the requesting peer, and also gives the requesting peer the IP and port on which the addressed peer is located. In addition, the peer establishing the connection must already know the public keys of the peer they are connecting to (but not necessary to any intermediary relay peer).

Indirect session establishment by means of a third party introduction is necessary for efficient NAT traversal. Charlie, a router behind a NAT or firewall which does not allow unsolicited inbound UDP packets, first contacts a few peers, choosing some to serve as introducers. Each of these peers (Bob, Bill, Betty, etc) provide Charlie with an introduction tag - a 4 byte random number - which he then makes available to the public as methods of contacting him. Alice, a router who has Charlie's published contact methods, first sends a RelayRequest packet to one or more of the introducers, asking each to introduce her to Charlie (offering the introduction tag to identify Charlie). Bob then forwards a RelayIntro packet to Charlie including Alice's public IP and port number, then sends Alice back a RelayResponse packet containing Charlie's public IP and port number. When Charlie receives the RelayIntro packet, he sends off a small random packet to Alice's IP and port (poking a hole in his NAT/firewall), and when Alice receives Bob's RelayResponse packet, she begins a new full direction session establishment with the specified IP and port.

Connection establishment (indirect using an introducer)

Alice first connects to introducer Bob, who relays the request to Charlie.

        Alice                         Bob                  Charlie
    RelayRequest ---------------------->
         <-------------- RelayResponse    RelayIntro ----------->
         <-------------------------------------------- HolePunch (data ignored)
    SessionRequest -------------------------------------------->
         <-------------------------------------------- SessionCreated
    SessionConfirmed ------------------------------------------>
         <-------------------------------------------- DeliveryStatusMessage
         <-------------------------------------------- DatabaseStoreMessage
    DatabaseStoreMessage -------------------------------------->
    Data <--------------------------------------------------> Data

After the hole punch, the session is established between Alice and Charlie as in a direct establishment.

IPv6 Notes

IPv6 is supported as of version 0.9.8. Published relay addresses may be IPv4 or IPv6, and Alice-Bob communication may be via IPv4 or IPv6. Through release 0.9.49, Bob-Charlie and Alice-Charlie communication is via IPv4 only. Relaying for IPv6 is supported as of release 0.9.50. See the specification for details.

While the specification was changed as of version 0.9.8, Alice-Bob communication via IPv6 was not actually supported until version 0.9.50. Earlier versions of Java routers erroneously published the 'C' capability for IPv6 addresses, even though they did not actually act as an introducer via IPv6. Therefore, routers should only trust the 'C' capability on an IPv6 address if the router version is 0.9.50 or higher.

Peer testing

The automation of collaborative reachability testing for peers is enabled by a sequence of PeerTest messages. With its proper execution, a peer will be able to determine their own reachability and may update its behavior accordingly. The testing process is quite simple:

        Alice                  Bob                  Charlie
    PeerTest ------------------->
                             PeerTest-------------------->
                                <-------------------PeerTest
         <-------------------PeerTest
         <------------------------------------------PeerTest
    PeerTest------------------------------------------>
         <------------------------------------------PeerTest

Each of the PeerTest messages carry a nonce identifying the test series itself, as initialized by Alice. If Alice doesn't get a particular message that she expects, she will retransmit accordingly, and based upon the data received or the messages missing, she will know her reachability. The various end states that may be reached are as follows:

  • If she doesn't receive a response from Bob, she will retransmit up to a certain number of times, but if no response ever arrives, she will know that her firewall or NAT is somehow misconfigured, rejecting all inbound UDP packets even in direct response to an outbound packet. Alternately, Bob may be down or unable to get Charlie to reply.
  • If Alice doesn't receive a PeerTest message with the expected nonce from a third party (Charlie), she will retransmit her initial request to Bob up to a certain number of times, even if she has received Bob's reply already. If Charlie's first message still doesn't get through but Bob's does, she knows that she is behind a NAT or firewall that is rejecting unsolicited connection attempts and that port forwarding is not operating properly (the IP and port that Bob offered up should be forwarded).
  • If Alice receives Bob's PeerTest message and both of Charlie's PeerTest messages but the enclosed IP and port numbers in Bob's and Charlie's second messages don't match, she knows that she is behind a symmetric NAT, rewriting all of her outbound packets with different 'from' ports for each peer contacted. She will need to explicitly forward a port and always have that port exposed for remote connectivity, ignoring further port discovery.
  • If Alice receives Charlie's first message but not his second, she will retransmit her PeerTest message to Charlie up to a certain number of times, but if no response is received she knows that Charlie is either confused or no longer online.

Alice should choose Bob arbitrarily from known peers who seem to be capable of participating in peer tests. Bob in turn should choose Charlie arbitrarily from peers that he knows who seem to be capable of participating in peer tests and who are on a different IP from both Bob and Alice. If the first error condition occurs (Alice doesn't get PeerTest messages from Bob), Alice may decide to designate a new peer as Bob and try again with a different nonce.

Alice's introduction key is included in all of the PeerTest messages so that Charlie can contact her without knowing any additional information. As of release 0.9.15, Alice must have an established session with Bob, to prevent spoofing attacks. Alice must not have an established session with Charlie for the peer test to be valid. Alice may go on to establish a session with Charlie, but it is not required.

IPv6 Notes

Through release 0.9.26, only testing of IPv4 addresses is supported. Only testing of IPv4 addresses is supported. Therefore, all Alice-Bob and Alice-Charlie communication must be via IPv4. Bob-Charlie communication, however, may be via IPv4 or IPv6. Alice's address, when specified in the PeerTest message, must be 4 bytes. As of release 0.9.27, testing of IPv6 addresses is supported, and Alice-Bob and Alice-Charlie communication may be via IPv6, if Bob and Charlie indicate support with a 'B' capability in their published IPv6 address. See Proposal 126 for details.

Prior to release 0.9.50, Alice sends the request to Bob using an existing session over the transport (IPv4 or IPv6) that she wishes to test. When Bob receives a request from Alice via IPv4, Bob must select a Charlie that advertises an IPv4 address. When Bob receives a request from Alice via IPv6, Bob must select a Charlie that advertises an IPv6 address. The actual Bob-Charlie communication may be via IPv4 or IPv6 (i.e., independent of Alice's address type).

As of release 0.9.50, If the message is over IPv6 for an IPv4 peer test, or (as of release 0.9.50) over IPv4 for an IPv6 peer test, Alice must include her introduction address and port. See Proposal 158 for details.

Transmission window, ACKs and Retransmissions

The DATA message may contain ACKs of full messages and partial ACKs of individual fragments of a message. See the data message section of the protocol specification page for details.

The details of windowing, ACK, and retransmission strategies are not specified here. See the Java code for the current implementation. During the establishment phase, and for peer testing, routers should implement exponential backoff for retransmission. For an established connection, routers should implement an adjustable transmission window, RTT estimate and timeout, similar to TCP or streaming. See the code for initial, min and max parameters.

セキュリティ

UDP source addresses may, of course, be spoofed. Additionally, the IPs and ports contained inside specific SSU messages (RelayRequest, RelayResponse, RelayIntro, PeerTest) may not be legitimate. Also, certain actions and responses may need to be rate-limited.

The details of validation are not specified here. Implementers should add defenses where appropriate.

Peer capabilities

One or more capabilities may be published in the "caps" option. Capabilities may be in any order, but "BC46" is the recommended order, for consistency across implementations.

B
If the peer address contains the 'B' capability, that means they are willing and able to participate in peer tests as a 'Bob' or 'Charlie'. Through 0.9.26, peer testing was not supported for IPv6 addresses, and the 'B' capability, if present for an IPv6 address, must be ignored. As of 0.9.27, peer testing is supported for IPv6 addresses, and the presence or absense of the 'B' capability in an IPv6 address indicates actual support (or lack of support).
C
If the peer address contains the 'C' capability, that means they are willing and able to serve as an introducer via that address - serving as an introducer Bob for an otherwise unreachable Charlie. Prior to release 0.9.50, Java routers incorrectly published the 'C' capability for IPv6 addresses, even though IPv6 introducers was not fully implemented. Therefore, routers should assume that versions prior to 0.9.50 cannot act as an introducer over IPv6, even if the 'C' capability is advertised.
4
As of 0.9.50, indicates outbound IPv4 capability. If an IP is published in the host field, this capability is not necessary. If this is an address with introducers for IPv4 introductions, '4' should be included. If the router is hidden, '4' and '6' may be combined in a single address.
6
As of 0.9.50, indicates outbound IPv6 capability. If an IP is published in the host field, this capability is not necessary. If this is an address with introducers for IPv6 introductions, '6' should be included (not currently supported). If the router is hidden, '4' and '6' may be combined in a single address.

Future Work

Note: These issues will be addressed in the development of SSU2.
  • Analysis of current SSU performance, including assessment of window size adjustment and other parameters, and adjustment of the protocol implementation to improve performance, is a topic for future work.
  • The current implementation repeatedly sends acknowledgments for the same packets, which unnecessarily increases overhead.
  • The default small MTU value of 620 should be analyzed and possibly increased. The current MTU adjustment strategy should be evaluated. Does a streaming lib 1730-byte packet fit in 3 small SSU packets? Probably not.
  • The protocol should be extended to exchange MTUs during the setup.
  • Rekeying is currently unimplemented and will never be.
  • The potential use of the 'challenge' fields in RelayIntro and RelayResponse, and use of the padding field in SessionRequest and SessionCreated, is undocumented.
  • A set of fixed packet sizes may be appropriate to further hide the data fragmentation to external adversaries, but the tunnel, garlic, and end to end padding should be sufficient for most needs until then.
  • Signed-on times in SessionCreated and SessionConfirmed appear to be unused or unverified.

Implementation Diagram

This diagram should accurately reflect the current implementation, however there may be small differences.

Specification

Now on the SSU specification page.