Network Working Group Henry Spencer Internet Draft SP Systems Expires: 21 May 2002 D. Hugh Redelmeier Mimosa Systems 21 Nov 2001 IKE Implementation Issues Status of this Memo This document is an Internet-Draft and is in full conformance with all provisions of Section 10 of RFC2026. If approved as an Informational RFC... This memo provides information for the Internet community. This memo does not specify an Internet standard of any kind. Distribution of this memo is unlimited. 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 21 May 2002. Copyright Notice Copyright (C) The Internet Society 2001. All Rights Reserved. Spencer & Redelmeier [Page 1] Internet Draft IKE Implementation Issues 21 Nov 2001 Table of Contents 1. Introduction ................................................... 3 2. Lower-level Background and Notes ............................... 4 3. IKE Infrastructural Issues ..................................... 5 3.1. Continuous Channel ........................................... 5 3.2. Retransmission ............................................... 5 3.3. Replay Prevention ............................................ 6 4. Basic Keying and Rekeying ...................................... 7 4.1. When to Create SAs ........................................... 7 4.2. When to Rekey ................................................ 7 4.3. Choosing an SA ............................................... 8 4.4. Why to Rekey ................................................. 9 4.5. Rekeying ISAKMP SAs .......................................... 9 4.6. Bulk Negotiation ............................................. 9 5. Deletions, Teardowns, Crashes .................................. 9 5.1. Deletions ................................................... 10 5.2. Teardowns and Shutdowns ..................................... 10 5.3. Crashes ..................................................... 11 5.4. Unknown SAs ................................................. 12 6. Misc. IKE Issues .............................................. 13 6.1. Group 1 ..................................................... 13 6.2. To PFS Or Not To PFS ........................................ 13 6.3. Debugging Tools, Lack Thereof ............................... 13 6.4. Terminology, Vagueness Thereof .............................. 14 6.5. A Question of Identity ...................................... 14 6.6. Opportunistic Encryption .................................... 14 7. Security Considerations ....................................... 15 8. References .................................................... 15 Author's Addresses ............................................... 16 Full Copyright Statement ......................................... 17 Spencer & Redelmeier [Page 2] Internet Draft IKE Implementation Issues 21 Nov 2001 Abstract The current IPsec specifications for key exchange and connection management, RFCs 2408 [ISAKMP] and 2409 [IKE], leave many aspects of connection management unspecified, most prominently rekeying practices. We can hope that future revisions of the specifications will clear this up. Meanwhile, it seems useful to set down some successful experiences, to minimize the extent to which new implementors have to rely on unwritten folklore. The Linux FreeS/WAN implementation of IPsec interoperates with almost every other IPsec implementation. This document describes how the FreeS/WAN project has resolved some of the gaps in the IPsec specifications (and plans to resolve some others), and what difficulties have been encountered, in hopes that this generally- successful experience might be informative to new implementors. This is offered as an Informational RFC. This -01 revision mostly just fills in the gaps in the -00 draft. 1. Introduction The current IPsec specifications for key exchange and connection management, RFCs 2408 [ISAKMP] and 2409 [IKE], leave many aspects of connection management unspecified, most prominently rekeying practices. This is a cryptic puzzle which each group of implementors has to struggle with, and differences in how the ambiguities and gaps are resolved are potentially a fruitful source of interoperability problems. We can hope that future revisions of the specifications will clear this up. Meanwhile, it seems useful to set down some successful experiences, to minimize the extent to which new implementors have to rely on unwritten folklore. The Linux FreeS/WAN implementation of IPsec interoperates with almost every other IPsec implementation, and because of its free nature, it also sees some use as a reference implementation by other implementors. The high degree of interoperability is noteworthy given its organizers' strong minimalist bias, which has caused them to implement only a small subset of the full glory of IPsec. This document describes how the FreeS/WAN project has resolved some of the gaps in the IPsec specifications (and plans to resolve some others), and what difficulties have been encountered, in hopes that this generally-successful experience might be informative to new implementors. One small caution about applicability: this experience may not be relevant to severely resource-constrained implementations. Spencer & Redelmeier [Page 3] Internet Draft IKE Implementation Issues 21 Nov 2001 FreeS/WAN's target environment is previous-generation PCs, now available at trivial cost (often, within an organization, at no cost), which have quite impressive CPU power and memory by the standards of only a few years ago. Some of the approaches discussed here may be inapplicable to implementations with severe external constraints which prevent them from taking advantage of modern hardware technology. 2. Lower-level Background and Notes FreeS/WAN implements ESP [ESP] and AH [AH] straightforwardly, although AH sees little use among our users. Our ESP/AH implementation cannot currently handle packets with IP options; somewhat surprisingly, this has caused little difficulty. We insist on encryption and do not support authentication-only connections, and this has not caused significant difficulty either. MTU and fragmentation issues, by contrast, have been a constant headache. We will not describe the details of our current approach to them, because it still needs work. One difficulty we have encountered is that many combinations of packet source and packet destination apparently cannot cope with an "interior minimum" in the path MTU, e.g. where an IPsec tunnel intervenes and its headers reduce the MTU for an intermediate link. This is particularly prevalent when using common PC software to connect to large well- known web sites; we think it is largely due to misconfigured firewalls which do not pass ICMP Fragmentation Required messages. The only solution we have yet found is to lie about the MTU of the tunnel, accepting the (undesirable) fragmentation of the ESP packets for the sake of preserving connectivity. We currently zero out the TOS field of ESP packets, rather than copying it from the inner header, on the grounds that it lends itself too well to traffic analysis and covert channels. We provide an option to restore RFC 2401 [IPSEC] copying behavior, but this appears to see little use. We initially implemented both DES [DES] and 3DES [CIPHERS] for both IKE and ESP, but after the Deep Crack effort [CRACK] demonstrated its inherent insecurity, we dropped support for DES. Somewhat surprisingly, our insistence on 3DES has caused almost no interoperability problems, despite DES being officially mandatory. A very few other systems either do not support 3DES or support it only as an optional upgrade, which inconveniences a few would-be users. There have also been one or two cases of systems which don't quite seem to know the difference! Spencer & Redelmeier [Page 4] Internet Draft IKE Implementation Issues 21 Nov 2001 See also section 6.1 for a consequence of our insistence on 3DES. We currently employ PF_KEY version 2 [PFKEY], plus various non- standard extensions, as our interface between keying and ESP. This has not proven entirely satisfactory. Our feeling now is that keying issues and policy issues do not really lend themselves to the clean separation that PF_KEY envisions. 3. IKE Infrastructural Issues A number of problems in IPsec connection management become easier if some attention is first paid to providing an infrastructure to support solving them. 3.1. Continuous Channel FreeS/WAN uses an approximation to the "continuous channel" model, in which ISAKMP SAs are maintained between IKEs so long as any IPsec SAs are open between the two systems. The resource consumption of this is minor: the only substantial overhead is occasional rekeying. IPsec SA management becomes significantly simpler if there is always a channel for transmission of control messages. We suggest (although we do not yet fully implement this) that inability to maintain (e.g., to rekey) this control path should be grounds for tearing down the IPsec SAs as well. 3.2. Retransmission The unreliable nature of UDP transmission is a nuisance. IKE implementations should always be prepared to retransmit the most recent message they sent on an ISAKMP SA, since there is some possibility that the other end did not get it. This means, in particular, that the system sending the supposedly-last message of an exchange cannot relax and assume that the exchange is complete, at least not until a significant timeout has elapsed. Systems must also retain information about the message most recently received in an exchange, so that a duplicate of it can be detected (and possibly interpreted as a NACK for the response). The retransmission rules FreeS/WAN follows are: (1) if a reply is expected, retransmit only if it does not appear before a timeout; and (2) if a reply is not expected (last message of the exchange), retransmit only on receiving a retransmission of the previous message. Notably, in case (1) we do NOT retransmit on receiving a retransmission, which avoids possible congestion problems arising from packet duplication, at the price of slowing response to packet loss. The timeout for case (1) is 10 seconds for the first retry, 20 Spencer & Redelmeier [Page 5] Internet Draft IKE Implementation Issues 21 Nov 2001 seconds for the second, and 40 seconds for all subsequent retries (normally only one, except when configuration settings call for persistence and the message is the first message of Main Mode with a new peer). These retransmission rules have been entirely successful. (Michael Thomas of Cisco has pointed out that the retry timeouts should include some random jitter, to de-synchronize hosts which are initially synchronized by, e.g., a power outage. We already jitter our rekeying times, as noted in section 4.2, but that does not help with initial startup. We're implementing jittered retries, but cannot yet report on experience with this.) There is a deeper problem, of course, when an entire "exchange" consists of a single message, e.g. the ISAKMP Informational Exchange. Then there is no way to decide whether or when a retransmission is warranted at all. This seems like poor design, to put it mildly (and there is now talk of fixing it). We have no experience in dealing with this problem at this time, although it is part of the reason why we have delayed implementing Notification messages. 3.3. Replay Prevention The unsequenced nature of UDP transmission is also troublesome, because it means that higher levels must consider the possibility of replay attacks. FreeS/WAN takes the position that systematically eliminating this possibility at a low level is strongly preferable to forcing careful consideration of possible impacts at every step of an exchange. RFC 2408 [ISAKMP] section 3.1 states that the Message ID of an ISAKMP message must be "unique". FreeS/WAN interprets this literally, as forbidding duplication of Message IDs within the set of all messages sent via a single ISAKMP SA. This requires remembering all Message IDs until the ISAKMP SA is superseded by rekeying, but that is not costly (four bytes per sent or received message), and it ELIMINATES replay attacks from consideration; we believe this investment of resources is well worthwhile. If the resource consumption becomes excessive--in our experience it has not--the ISAKMP SA can be rekeyed early to collect the garbage. There is theoretically an interoperability problem when talking to implementations which interpret "unique" more loosely and may re-use Message IDs, but it has not been encountered in practice. This approach appears to be completely interoperable. The proposal by Andrew Krywaniuk [REPLAY], which advocates turning the Message ID into an anti-replay counter, would achieve the same goal without the minor per-message memory overhead. This may be Spencer & Redelmeier [Page 6] Internet Draft IKE Implementation Issues 21 Nov 2001 preferable, although it means an actual protocol change and more study is needed. 4. Basic Keying and Rekeying 4.1. When to Create SAs As Tim Jenkins [REKEY] pointed out, there is a potential race condition in Quick Mode: a fast lightly-loaded Initiator might start using IPsec SAs very shortly after sending QM3 (the third and last message of Quick Mode), while a slow heavily-loaded Responder might not be ready to receive them until after spending a significant amount of time creating its inbound SAs. The problem is even worse if QM3 gets delayed or lost. FreeS/WAN's approach to this is what Jenkins called "Responder Pre- Setup": the Responder creates its inbound IPsec SAs before it sends QM2, so they are always ready and waiting when the Initiator sends QM3 and begins sending traffic. This approach is simple and reliable, and in our experience it interoperates with everybody. (There is potentially still a problem if FreeS/WAN is the Initiator and the Responder does not use Responder Pre-Setup, but no such problems have been seen.) The only real weakness of Responder Pre- Setup is the possibility of replay attacks, which we have eliminated by other means (see section 3.3). With this approach, the Commit Bit is useless, and we ignore it. In fact, until quite recently we discarded any IKE message containing it, and this caused surprisingly few interoperability problems; apparently it is not widely used. We have recently been persuaded that simply ignoring it is preferable; preliminary experience with this indicates that the result is successful interoperation with implementations which set it. 4.2. When to Rekey To preserve connectivity for user traffic, rekeying of a connection (that is, creation of new IPsec SAs to supersede the current ones) must begin before its current IPsec SAs expire. Preferably one end should predictably start rekeying negotiations first, to avoid the extra overhead of two simultaneous negotiations, although either end should be prepared to rekey if the other does not. There is also a problem with "convoys" of keying negotiations: for example, a "hub" gateway with many IPsec connections can be inundated with rekeying negotiations exactly one connection-expiry time after it reboots, and the massive overload this induces tends to make this situation self- perpetuating, so it recurs regularly. (Convoys can also evolve gradually from initially-unsynchronized negotiations.) Spencer & Redelmeier [Page 7] Internet Draft IKE Implementation Issues 21 Nov 2001 FreeS/WAN has the concept of a "rekeying margin", measured in seconds. If FreeS/WAN was the Initiator for the previous rekeying (or the startup, if none) of the connection, it nominally starts rekeying negotiations at expiry time minus one rekeying margin. Some random jitter is added to break up convoys: rather than starting rekeying exactly at minus one margin, it starts at a random time between minus one margin and minus two margins. If FreeS/WAN was the Responder for the previous rekeying/startup, and nothing has been heard from the previous Initiator at expiry time minus one-half the rekeying margin, FreeS/WAN will initiate rekeying negotiations. No jitter is applied (perhaps it should be). Having the Initiator lead the way is an obvious way of deciding who should speak first, since there is already an Initiator/Responder asymmetry in the connection. Moreover, our experience has been that Initiator lead gives a significantly higher probability of successful negotiation! The negotiation process itself is asymmetric, because the Initiator must make a few specific proposals which the Responder can only accept or reject, so the Initiator must try to guess where its "acceptable" region (in parameter space) might overlap with the Responder's. We have seen situations where negotiations would succeed or fail depending on which end initiated them, because one end was making better guesses. Given an existing connection, we KNOW that the previous Initiator WAS able to initiate a successful negotiation, so it should (if at all possible) take the lead again. FreeS/WAN defaults the rekeying margin to 9 minutes, although this can be changed by configuration. There is also a configuration option to adjust the permissible range of jitter. The defaults were chosen somewhat arbitrarily, but they work extremely well and the configuration options are rarely used. 4.3. Choosing an SA Once rekeying has occurred, both old and new IPsec SAs for the connection exist, at least momentarily. FreeS/WAN accepts incoming traffic on either old or new inbound SAs, but sends outgoing traffic only on the new outbound ones. This approach appears to be significantly more robust than using the old ones until they expire, notably in cases where renegotiation has occurred because something has gone wrong on the other end. It avoids having to pay meticulous attention to the state of the other end, state which is difficult to learn reliably given the limitations of IKE. This approach has interoperated successfully with ALMOST all other implementations. The only (well-characterized) problem cases have been implementations which rely on receiving a Delete message for the Spencer & Redelmeier [Page 8] Internet Draft IKE Implementation Issues 21 Nov 2001 old SAs to tell them to switch over to the new ones. Since delivery of Delete is unreliable, and support for Delete is optional, this reliance seems like a serious mistake. This is all the more true because Delete announces that the deletion has already occurred [ISAKMP, section 3.15], not that it is about to occur, so packets already in transit in the other direction could be lost. Delete should be used for resource cleanup, not for switchover control. (These matters are discussed further in section 5.) 4.4. Why to Rekey FreeS/WAN currently implements only time-based expiry (life in seconds), although we are working toward supporting volume-based expiry (life in kilobytes) as well. The lack of volume-based expiry has not been an interoperability problem so far. Volume-based expiry does add some minor complications. In particular, it makes explicit Delete of now-disused SAs more important, because once an SA stops being used, it might not expire on its own. We believe this lacks robustness and is generally unwise, and expect to use volume-based expiry only as a supplement to time-based expiry. However, Delete support (see section 5) does seem advisable for use with volume-based expiry. We do not believe that volume-based expiry alters the desirability of switching immediately to the new SAs after rekeying. Rekeying margins are normally a small fraction of the total life of an SA, so we feel there is no great need to "use it all up". 4.5. Rekeying ISAKMP SAs The above discussion has focused on rekeying for IPsec SAs, but FreeS/WAN applies the same approaches to rekeying for ISAKMP SAs, with similar success. 4.6. Bulk Negotiation Quick Mode nominally provides for negotiating possibly-large numbers of similar but unrelated IPsec SAs simultaneously [IKE, section 9]. Nobody appears to do this. FreeS/WAN does not support it, and its absence has caused no problems. 5. Deletions, Teardowns, Crashes FreeS/WAN currently ignores all Notifications and Deletes, and never generates them. This has caused little difficulty in interoperability, which shouldn't be surprising (since Notification and Delete support is officially entirely optional) but does seem to Spencer & Redelmeier [Page 9] Internet Draft IKE Implementation Issues 21 Nov 2001 surprise some people. Nevertheless, we do plan some changes to this approach based on past experience. 5.1. Deletions As hinted at above, we plan to implement Delete support. Shortly after rekeying, the Responder will issue a Delete for its old inbound SAs. The Responder initiates this because the Initiator started using the new SAs on sending QM3, while the Responder started using them only on receiving QM3, so there is less chance of old-SA packets still being in transit from the Initiator. The Initiator will issue a Delete only if it does not hear one from the Responder after a longer delay. Either party, on receiving a Delete for the old outbound SAs of a connection, will respond with a Delete for the old inbound SAs. Either party, on NOT receiving a Delete in response to a Delete it sent, will re-send after a timeout (although only a few retries will be attempted). As noted in section 4.3, these Deletes are intended for resource cleanup, not to control switching between SAs. But we expect that they will improve interoperability with some broken implementations. We believe strongly that connections need to be considered as a whole, rather than treating each SA as an independent entity. We will issue Deletes only for the full set of inbound SAs of a connection, and will treat a Delete for any outbound SA as equivalent to deletion of all the outbound SAs for the associated connection. 5.2. Teardowns and Shutdowns When a connection is not intended to be up permanently, there is a need to coordinate teardown, so that both ends are aware that the connection is down. This is both for recovery of resources, and to avoid routing packets through dangling SAs which can no longer deliver them. Connection teardown will use the same bidirectional exchange of Deletes as discussed in section 5.1: a Delete received for current SAs (not yet obsoleted by rekeying) indicates that the other host wishes to tear down the associated connection. When a host with current connections knows that it is about to shut down, it will issue Deletes for all SAs involved, advising its peers (as per the meaning of Delete [ISAKMP, section 3.15]) that the SAs have become useless. It will ignore attempts at rekeying or connection startup thereafter, until it shuts down. Spencer & Redelmeier [Page 10] Internet Draft IKE Implementation Issues 21 Nov 2001 It would be better to have a Final-Contact notification, analogous to Initial-Contact but indicating that no new negotiations should be attempted until further notice. Initial-Contact actually could be used for this (!), but in networks where connections are intended to exist permanently, it seems likely to provoke unwanted attempts to renegotiate the lost connections. 5.3. Crashes Systems sometimes crash. Coping with the resulting loss of information is easily the most difficult problem we have found in implementing robust IPsec systems. When connections are intended to be permanent, it is simple to specify renegotiation on reboot. With our approach to SA selection (see section 4.3), this handles such cases robustly and well. We do have to tell users that BOTH hosts should be set this way. In cases where crashes are synchronized (e.g. by power interruptions), this may result in simultaneous negotiations at reboot. We currently allow both negotiations to proceed to completion, but our use-newest selection method effectively ignores one connection or the other, and when one of them rekeys, we notice that the new SAs replace those of both old connections, and we then refrain from rekeying the other. (This duplicate detection is desirable in any event, for robustness, to ensure that the system converges on a reasonable state eventually after it is perturbed by difficulties or bugs.) When connections are not permanent, the situation is less happy. One particular situation in which we see problems is when a number of "Road Warrior" hosts occasionally call in to a central server. The server is normally configured not to initiate such connections, since it does not know when the Road Warrior is available (or what IP address it is using). Unfortunately, if the server crashes and reboots, any Road Warriors then connected have a problem: they don't know that the server has crashed, so they can't renegotiate, and the server has forgotten both the connections and their (transient) IP addresses, so it cannot renegotiate. We believe that the simplest answer to this problem is what John Denker has dubbed "address inertia": the server makes a best-effort attempt to remember (in nonvolatile storage) which connections were active and what the far- end addresses were, so that it can attempt renegotiation on reboot. We have not implemented this yet, but intend to; Denker has implemented it himself, although in a somewhat messy way, and reports excellent results. Spencer & Redelmeier [Page 11] Internet Draft IKE Implementation Issues 21 Nov 2001 5.4. Unknown SAs A more complete solution to crashes would be for an IPsec host to note the arrival of ESP packets on an unknown IPsec SA, and report it somehow to the other host, which can then decide to renegotiate. This arguably might be preferable in any case--if the non-rebooted host has no traffic to send, it does not care whether the connection is intact--but delays and packet loss will be reduced if the connection is renegotiated BEFORE there is traffic for it. So unknown-SA detection is best reserved as a fallback method, with address inertia used to deal with most such cases. A difficulty with unknown-SA detection is, just HOW should the other host be notified? IKE provides no good way to do the notification: Notification payloads (e.g., Initial-Contact) are unauthenticated unless they are sent under protection of an ISAKMP SA. A "Security Failures - Bad SPI" ICMP message [SECFAIL] is an interesting alternative, but has the disadvantage of likewise being unauthenticated. It's fundamentally unlikely that there is a simple solution to this, given that almost any way of arranging or checking authentication for such a notification is costly. We think the best answer to this is a two-step approach. An unauthenticated Initial-Contact or Security Failures - Bad SPI cannot be taken as a reliable report of a problem, but can be taken as a hint that a problem MIGHT exist. There needs to be some reliable way of checking such hints, subject to rate limiting since the checks are likely to be costly (and checking the same connection repeatedly at short intervals is unlikely to be worthwhile anyway). Then the rebooted host sends the notification, and the non-rebooted host-- which still thinks it has a connection--checks whether the connection still works, and renegotiates if not. We have not yet decided what form the notification should take. IKE Initial-Contact is an obvious possibility, but has some disadvantages. It does not specify which connection has had difficulties. Also, the specification [IKE section 4.6.3.3] refers to "remote system" and "sending system" without clearly specifying just what those terms mean; in the case of a multi-homed host using multiple forms of identification, the question is not trivial. Initial-Contact does have the fairly-decisive advantage that it is likely to convey the right general meaning even to an implementation which does not do things exactly the way ours does. A more fundamental difficulty is what form the reliable check takes. What is wanted is an "IKE ping", verifying that the ISAKMP SA is still intact (it being unlikely that IPsec SAs have been lost while the ISAKMP SA has not). The lack of such a facility is a serious Spencer & Redelmeier [Page 12] Internet Draft IKE Implementation Issues 21 Nov 2001 failing of IKE. An acknowledged Notification of some sort would be ideal, but there is none at present. Some existing implementations are known to use the private Notification values 30000 as ping and 30002 as ping reply, and that seems the most attractive choice at present. If it is not recognized, there will probably be no reply, and the result will be an unnecessary renegotiation, so this needs strict rate limiting. While we think this facility is desirable, and is about the best that can be done with the poor tools available, we have not gotten very far in implementation and cannot comment intelligently about how well it works or interoperates. 6. Misc. IKE Issues 6.1. Group 1 We have dropped support for the first Oakley Group (group 1), despite it being officially mandatory, on the grounds that it is grossly too weak to provide enough randomness for 3DES. There have been some interoperability problems, mostly quite minor: ALMOST everyone supports group 2 as well, although sometimes it has to be explicitly configured. We also support the quasi-standard group 5 [GROUPS]. This has not been seriously exercised yet, because at the moment we offer group 2 first and almost everyone accepts it. We plan to change to offering group 5 first. 6.2. To PFS Or Not To PFS A persistent small interoperability problem is that the presence or absence of PFS (for keys) is neither negotiated nor announced. We have it enabled by default, and successful interoperation often requires having the other end turn it on in their implementation, or having the FreeS/WAN end disable it. Almost everyone supports it, but it's usually not the default, and interoperability is often impossible unless the two ends somehow reach prior agreement on it. We do not explicitly support the other flavor of PFS, for identities, and this has caused no interoperability problems. 6.3. Debugging Tools, Lack Thereof We find IKE lacking in basic debugging tools. Section 5.4, above, notes that an IKE ping would be useful for connectivity verification. It would also be extremely helpful for determining that UDP/500 packets get back and forth successfully between the two ends, which Spencer & Redelmeier [Page 13] Internet Draft IKE Implementation Issues 21 Nov 2001 is often an important first step in debugging. It's also quite common to have IKE negotiate a connection successfully, but to have some firewall along the way blocking ESP, and users find this mysterious and difficult to diagnose. We have no immediate suggestions on what could be done about it. 6.4. Terminology, Vagueness Thereof The terminology of IPsec needs work. We feel that both the specifications and more-user-oriented documentation would be greatly clarified by concise, intelligible names for certain concepts. We semi-consistently use "group" for the set of IPsec SAs which are established in one direction by a single Quick Mode negotiation and are used together to process a packet (e.g., an ESP SA plus an AH SA), "connection" for the logical packet path provided by a succession of pairs of groups (each rekeying providing a new pair, one group in each direction), and "keying channel" for the corresponding supervisory path provided by a sequence of ISAKMP SAs. We think it's a botch that "PFS" refers to two very different things, but we have no specific new terms to suggest, since we only implement one kind of PFS and thus can just ignore the other. 6.5. A Question of Identity One specification problem deserves note: exactly when can an existing phase 1 negotiation be re-used for a new phase 2 negotiation, as IKE [IKE, section 4] specifies? Presumably, when it connects the same two "parties"... but exactly what is a "party"? As noted in section 5.4, in cases involving multi-homing and multiple identities, it's not clear exactly what criteria are used for deciding whether the intended far end for a new negotiation is the same one as for a previous negotiation. Is it by Identification Payload? By IP address? Or what? We currently use a somewhat-vague notion of "identity", basically what gets sent in Identification Payloads, for this, and this seems to be successful, but we think this needs better specification. 6.6. Opportunistic Encryption Further IKE challenges appear in the context of Opportunistic Encryption [OE], but operational experience with it is too limited as yet for us to comment usefully right now. Spencer & Redelmeier [Page 14] Internet Draft IKE Implementation Issues 21 Nov 2001 7. Security Considerations Since this document discusses aspects of building robust and interoperable IPsec implementations, security considerations permeate it. 8. References [AH] Kent, S., and Atkinson, R., "IP Authentication Header", RFC 2402, Nov 1998. [CIPHERS] Pereira, R., and Adams, R., "The ESP CBC-Mode Cipher Algorithms", RFC 2451, Nov 1998. [CRACK] Electronic Frontier Foundation, "Cracking DES: Secrets of Encryption Research, Wiretap Politics and Chip Design", O'Reilly 1998, ISBN 1-56592-520-3. [DES] Madson, C., and Doraswamy, N., "The ESP DES-CBC Cipher Algorithm", RFC 2405, Nov 1998. [ESP] Kent, S., and Atkinson, R., "IP Encapsulating Security Payload (ESP)", RFC 2406, Nov 1998. [GROUPS] Kivinen, T., and Kojo, M., "More MODP Diffie-Hellman groups for IKE", , 29 Aug 2001 (work in progress). [IKE] Harkins, D., and Carrel, D., "The Internet Key Exchange (IKE)", RFC 2409, Nov 1998. [IPSEC] Kent, S., and Atkinson, R., "Security Architecture for the Internet Protocol", RFC 2401, Nov 1998. [ISAKMP] Maughan, D., Schertler, M., Schneider, M., and Turner, J., "Internet Security Association and Key Management Protocol (ISAKMP)", RFC 2408, Nov 1998. [OE] Richardson, M., Redelmeier, D. H., and Spencer, H., "A method for doing opportunistic encryption with IKE", , Oct 2001 (work in progress). [PFKEY] McDonald, D., Metz, C., and Phan, B., "PF_KEY Key Management API, Version 2", RFC 2367, July 1998. Spencer & Redelmeier [Page 15] Internet Draft IKE Implementation Issues 21 Nov 2001 [REKEY] Tim Jenkins, "IPsec Re-keying Issues", , 2 May 2000 (draft expired, work no longer in progress). [REPLAY] Krywaniuk, A., "Using Isakmp Message Ids for Replay Protection", , 12 July 2001 (work in progress). [SECFAIL] Karn, P., and Simpson, W., "ICMP Security Failures Messages", RFC 2521, March 1999. Author's Addresses Henry Spencer SP Systems Box 280 Stn. A Toronto, Ont. M5W1B2 Canada henry@spsystems.net 416-690-6561 D. Hugh Redelmeier Mimosa Systems Inc. 29 Donino Ave. Toronto, Ont. M4N2W6 Canada hugh@mimosa.com 416-482-8253 Spencer & Redelmeier [Page 16] Internet Draft IKE Implementation Issues 21 Nov 2001 Full Copyright Statement Copyright (C) The Internet Society 2001. All Rights Reserved. This document and translations of it may be copied and furnished to others, and derivative works that comment on or otherwise explain it or assist in its implmentation may be prepared, copied, published and distributed, in whole or in part, without restriction of any kind, provided that the above copyright notice and this paragraph are included on all such copies and derivative works. However, this document itself may not be modified in any way, such as by removing the copyright notice or references to the Internet Society or other Internet organizations, except as needed for the purpose of developing Internet standards in which case the procedures for copyrights defined in the Internet Standards process must be followed, or as required to translate it into languages other than English. The limited permissions granted above are perpetual and will not be revoked by the Internet Society or its successors or assigns. This document and the information contained herein is provided on an "AS IS" basis and THE INTERNET SOCIETY AND THE INTERNET ENGINEERING TASK FORCE DISCLAIMS 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. Spencer & Redelmeier [Page 17]