Internet Engineering Task Force P. Savola
Internet-Draft CSC/FUNET
Expires: November 10, 2005 May 9, 2005
MTU and Fragmentation Issues with In-the-Network Tunneling
draft-savola-mtufrag-network-tunneling-03.txt
Status of this Memo
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 November 10, 2005.
Copyright Notice
Copyright (C) The Internet Society (2005).
Abstract
Tunneling techniques such as IP-in-IP when deployed in the middle of
the network, typically between routers, have certain issues regarding
how large packets can be handled: whether such packets would be
fragmented and reassembled (and how), whether Path MTU Discovery
would be used, or how this scenario could be operationally avoided.
This memo justifies why this is a common, non-trivial problem, and
goes on to describe the different solutions and their characteristics
at some length.
Savola Expires November 10, 2005 [Page 1]
Internet-Draft Packet Size Issues in Network Tunnels May 2005
Table of Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . 3
2. Problem Statement . . . . . . . . . . . . . . . . . . . . . . 4
3. Description of Solutions . . . . . . . . . . . . . . . . . . . 5
3.1 Fragmentation and Reassembly by the Tunnel Endpoints . . . 5
3.2 Signalling the Lower MTU to the Sources . . . . . . . . . 6
3.3 Encapsulate Only When There is Free MTU . . . . . . . . . 6
3.4 Fragmentation of the Inner Packet . . . . . . . . . . . . 8
4. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . 8
5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 9
6. Security Considerations . . . . . . . . . . . . . . . . . . . 9
7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . 10
8. References . . . . . . . . . . . . . . . . . . . . . . . . . . 10
8.1 Normative References . . . . . . . . . . . . . . . . . . . 10
8.2 Informative References . . . . . . . . . . . . . . . . . . 11
Author's Address . . . . . . . . . . . . . . . . . . . . . . . 11
A. MTU of the Tunnel . . . . . . . . . . . . . . . . . . . . . . 11
Intellectual Property and Copyright Statements . . . . . . . . 12
Savola Expires November 10, 2005 [Page 2]
Internet-Draft Packet Size Issues in Network Tunnels May 2005
1. Introduction
A large number of ways to encapsulate datagrams in other packets,
i.e., tunneling mechanisms, have been specified over the years: for
example, IP-in-IP (e.g., [1], [2]), GRE [3], L2TP [4], or IPsec [5]
in tunnel mode -- any of which might run on top of IPv4, IPv6, or
some other protocol and carrying the same or a different protocol.
All of these can be run so that the endpoints of the inner protocol
are co-located with the endpoints of the outer protocol; in a typical
scenario, this would correspond to "host-to-host" tunneling. It is
also possible to have one set of endpoints co-located, i.e., host-to-
router or router-to-host tunneling. Finally, many of these
mechanisms are also employed between the routers for all or a part of
the traffic that passes between them, resulting in router-to-router
tunneling.
All these protocols and scenarios have one issue in common: how does
the source select the maximum packet size so that the packets will
fit, even encapsulated, in the largest Maximum Transfer Unit (MTU) of
the traversed path in the network; and if you cannot affect the
packet sizes, what do you do to be able to encapsulate them in any
case? The four main solutions are (these will be elaborated in
Section 3):
1. Fragmenting all too big encapsulated packets to fit in the paths,
and reassembling them at the tunnel end-points.
2. Signal to all the sources whose traffic must be encapsulated, and
is larger than that fits, to send smaller packets, e.g., using
Path MTU Discovery [6] [7].
3. Ensure that in the specific environment, the encapsulated packets
will fit in all the paths in the network, e.g., by using MTU
bigger than 1500 in the backbone used for encapsulation.
4. Fragmenting the original too big packets so that their fragments
will fit, even encapsulated, in the paths, and reassembling them
at the destination nodes. Note that this approach is only
available for IPv4 under certain assumptions (see Section 3.4).
The tunneling packet size issues are relatively straightforward in
host-to-host tunneling or host-to-router tunneling where Path MTU
Discovery only needs to signal to one source node. The issues are
signficantly more difficult in router-to-router and certain router-
to-host scenarios, which are the focus of this memo.
It is worth noting that most of this discussion applies to a more
Savola Expires November 10, 2005 [Page 3]
Internet-Draft Packet Size Issues in Network Tunnels May 2005
generic case, where there exists a link with lower MTU in the path.
A concrete and widely deployed example of this is the usage of PPP
over Ethernet (PPPoE) [9] at the customers' access link. These
lower-MTU links, and particularly PPPoE links, are typically not
deployed in topologies where fragmentation and reassembly might be
unfeasible (e.g., a backbone), so this may be a slightly easier
problem. However, this more generic case is considered out of scope
of this memo.
There are also known challenges in specifying and implementing a
mechanism which would be used at the tunnel end-point to obtain the
best suitable packet size to use for encapsulation; if a static value
is chosen, a lot of fragmentation might end up being performed; if
PMTUD is used, the implementation would need to use or relay the
received Packet Too Big messages, and assume that sufficient data has
been piggybacked on the ICMP messages (beyond the required 64 bits
for ICMPv4) to make this possible. However, this problem is
described elsewhere (e.g., in [3] and [1]) and is out of scope of
this memo.
Section 2 includes a problem statement, section 3 describes the
different solutions with their drawbacks and advantages, and section
4 presents conclusions.
2. Problem Statement
It is worth considering why exactly this is considered a problem.
It is possible to fix all the packet size issues using the solution
1, fragmenting the resulting encapsulated packet, and reassembling it
by the tunnel endpoint. However, this is considered problematic for
at least three reasons, as described in Section 3.1.
Therefore it is desirable to avoid fragmentation and reassembly if
possible. On the other hand, the other solutions may not be
practical either: especially in router-to-router or router-to-host
tunneling, Path MTU Discovery might be very disadvantageous --
consider the case where a backbone router would send an ICMP Packet
Too Big messages to every source who would try to send packets
through it. Fragmenting before encapsulation is also not available
in IPv6, and not available when the Don't Fragment (DF) bit has been
set (unless the implementation ignores the DF bit). Ensuring high
enough MTU so encapsulation is always possible is of course a valid
approach, but requires careful operational planning, and may not be a
feasible assumption for implementors.
This yields that there is no trivial solution to this problem, and it
needs to be further explored to consider the tradeoffs, as is done in
Savola Expires November 10, 2005 [Page 4]
Internet-Draft Packet Size Issues in Network Tunnels May 2005
this memo.
3. Description of Solutions
This section describes the potential solutions in a bit more detail.
3.1 Fragmentation and Reassembly by the Tunnel Endpoints
The seemingly simplest solution to tunneling packet size issues is
fragmentation of the outer packet by the encapsulator, and reassembly
by the decapsulator. However, this is highly problematic for at
least three reasons:
o Fragmentation causes overhead: every fragment requires the IP
header (20 or 40 bytes), and with IPv6, additional 8 bytes for the
Fragment Header.
o Fragmentation and reassembly require computation: splitting
datagrams to fragments is a non-trivial procedure, and so is their
reassembly. For example, software router forwarding
implementations may not be able to be perform these operations at
line rate.
o Reassembling requires buffers: fragments might get lost, be
reordered or delayed; when that happens, the reassembly engine has
to wait with the partial packet for some time. When this would
have to be done at the line rate, with e.g., 10 Gbit/s speed, the
length of the buffers that reassembly might require, especially in
the worst case, might be considerable.
When examining router-to-router tunneling, the third problem is
likely the worst; certainly, a hardware computation and
implementation requirement would also be significant, but not all
that difficult in the end -- and the link capacity wasted in the
backbones by additional overhead might not be a huge problem either.
However, IPv4 identification header length is only 16 bits (compared
to 32 bits in IPv6), and if a larger number of packets are being
tunneled between two IP addresses, the ID is very likely to wrap and
cause data misassociation. This reassembly wrongly combining data
from two unrelated packets causes data integrity and potentially a
confidentiality violation. This problem is further described in
[10].
So, if reassembly could be made to work sufficiently reliably, this
would be one acceptable fallback solution but only for IPv6.
Savola Expires November 10, 2005 [Page 5]
Internet-Draft Packet Size Issues in Network Tunnels May 2005
3.2 Signalling the Lower MTU to the Sources
Another approach is to use techniques like Path MTU Discovery (or
potentially a future derivative [11]) to signal to the sources whose
packets will be encapsulated in the network to send smaller packets
so that they can be encapsulated.
Note that this only works if the MTU of the tunnel is of reasonable
size, and not e.g., 64 kilobytes: see Appendix A for more.
This approach would presuppose that PMTUD works. While it is
currently working for IPv6, and critical for its operation, there is
ample evidence that in IPv4, PMTUD is far from reliable due to e.g.,
firewalls and other boxes being configured to inappropriately drop
all the ICMP packets [12], or software bugs rendering PMTUD
inoperational.
Further, there are two scenarios where signalling from the network
would be highly undesirable: when the encapsulation would be done in
such a prominent place in the network that a very large number of
sources would need to be signalled with this information (possibly
even multiple times, depending on how long they keep their PMTUD
state), or when the encapsulation is done for passive monitoring
purposes (network management, lawful interception, etc.) -- when it's
critical that the sources whose traffic is being encapsulated are not
aware of this happening.
A related technique, which works with TCP under specific scenarios
only is so-called "MSS clamping". With that technique or rather a
"hack", the TCP packets' Maximum Segment Size (MSS) is reduced by
tunnel endpoints so that the TCP connection automatically restricts
itself to the maximum available packet size. Obviously this does not
work for UDP or other protocols which have no MSS. This approach is
most applicable and used with PPPoE, but could be applied otherwise
as well; the approach also assumes that all the traffic goes through
tunnel endpoints which do MSS clamping -- this is trivial for the
single-homed access links, but could be a challenge otherwise.
A new approach to PMTUD is in the works [11], but it is uncertain
whether that would fix the problems -- at least not the passive
monitoring requirements.
3.3 Encapsulate Only When There is Free MTU
The third approach is an operational one, depending on the
environment where encapsulation and decapsulation is being performed.
That is, if an ISP would deploy tunneling in its backbone, which
would consist only of links supporting high MTUs (e.g., Gigabit
Savola Expires November 10, 2005 [Page 6]
Internet-Draft Packet Size Issues in Network Tunnels May 2005
Ethernet or SDH/SONET), but all its customers and peers would have a
lower MTU (e.g., 1500, or the backbone MTU minus the encapsulation
overhead), this would imply that no packets (with the encapsulation
overhead added) would have larger MTU than the "backbone MTU", and
all the encapsulated packets would always fit MTU-wise in the
backbone links.
This approach is highly assumptive of the deployment scenario. It
may be desirable to build a tunnel to/from another ISP (for example),
where this might no longer hold; or there might be links in the
network which cannot support the higher MTUs to satisfy the tunneling
requirements; or customers themselves might try to tunnel fragmented
packets to the ISP, requiring the reassembly capability from the
ISP's equipment (in this last case, it might be possible to get the
MTU at the customer's end lowered, eliminating the fragmentation, but
it might not always be an option).
To restate, this approach can only be considered when tunneling is
done inside a part of specific kind of ISP's own network, not e.g.,
transiting an ISP.
Another, related approach might be having the sources use only a low
enough MTU which would fit in all the physical MTUs; for example,
IPv6 specifies the minimum MTU of 1280 bytes. For example, if all
the sources whose traffic would be encapsulated would use this as the
maximum packet size, there would probably always be enough free MTU
for encapsulation in the network. However, this is not the case
today, and it would be completely unrealistic to assume that this
kind of approach could be made to work in general.
It is worth remembering that while the IPv6 minimum MTU is 1280 bytes
[8], there are scenarios where the tunnel implementation must
implement fragmentation and reassembly [2]: for example, when having
an IPv6-in-IPv6 tunnel on top of a physical interface with MTU of
1280 bytes, or when having two layers of IPv6 tunneling. This can
only be avoided by ensuring that links on top of which IPv6 is being
tunneled have a somewhat larger MTU (e.g., 40 bytes) than 1280 bytes.
This conclusion can be generalized: because IP can be tunneled on top
of IP, no single minimum or maximum MTU can be found such that
fragmentation or signalling to the sources would never be needed.
All in all, while in certain operational environments it might be
possible to avoid any problems by deployment choices, or limiting the
MTU that the sources use, this is probably not a sufficiently good
general solution for the equipment vendors, and other solutions must
also be provided.
Savola Expires November 10, 2005 [Page 7]
Internet-Draft Packet Size Issues in Network Tunnels May 2005
3.4 Fragmentation of the Inner Packet
A final possibility is fragmenting the inner packet, before
encapsulation, in such a manner that the encapsulated packet fits in
the the path MTU. However, one should note that only IPv4 supports
this "in-flight" fragmentation; further, it isn't allowed for packets
where Don't Fragment -bit has been set. Even if one could ignore
IPv6 completely, so many IPv4 host stacks send packets with DF bit
set that this would seem unfeasible.
Regardless of what the specifications say, there are implementations
that perform fragmentation if required regardless of the DF bit:
either ignoring the DF bit completely, either for all or specified
interfaces, or clearing the DF bit in egress of the specified
interfaces. This is non-compliant behaviour, but there are certainly
uses for it especially in certain tightly controlled passive
monitoring scenarios, and has potential for more generic
applicability as well, to work around PMTUD issues.
As this is an implemented and desired (by some) behaviour, the full
impacts e.g., for the functioning of PMTUD (for example) should be
analyzed, and the use of fragmentation-related IPv4 bits should be
re-evaluated.
In summary, this approach provides a relatively easy fix for IPv4
problems, with potential for causing problems for PMTUD; as this
would not work with IPv6, it could not be considered a generic
solution.
4. Conclusions
Fragmentation and reassembly by the tunnel endpoints is a clear
solution to the problem, but the hardware reassembly when the packets
get lost may face significant implementation challenges. Whether
these challenges are practically insurmountable or not should be
evaluated. This approach does not seem feasible especially for IPv4
with high data rates due to problems with wrapping fragment
identification field [10]. Constant wrapping may occur when the data
rate is in the order of MB/s for IPv4 and in the order of dozens of
GB/s for IPv6. However, this reassembly approach is probably not a
problem for passive monitoring applications.
PMTUD techniques, at least at the moment and especially for IPv4,
appear to be too unreliable or unscalable to be used in the
backbones. It is an open question whether a future solution might
work better in this aspect.
It is clear that in some environments the operational approach to the
Savola Expires November 10, 2005 [Page 8]
Internet-Draft Packet Size Issues in Network Tunnels May 2005
problem, ensuring that fragmentation is never necessary by keeping
higher MTUs in the networks where encapsulated packets traverse, is
sufficient. But this is unlikely to be enough in general, and for
vendors which may not be able to make assumptions about the
operators' deployments.
Fragmentation of the inner packet is only possible with IPv4, and is
sufficient only if standards-incompliant behaviour, with potential
for bad side-effects e.g., for PMTUD, is adopted. It should not be
used if there are alternatives; fragmentation of the outer packet
seems a better option for passive monitoring.
An interesting thing to explicitly note is that when tunneling is
done in a high-speed backbone, typically one may be able to make
assumptions on the environment; however, when reassembly is not
performed in such a network, it might be done in software or with
lower requirements, and there either exists a reassembly
implementation, using PMTUD, or using a separate approach for passive
monitoring -- so this might not be a real problem.
In consequence, the critical questions at this point appear to be 1)
whether a higher MTU can be assumed in the high-speed networks that
deploy tunneling, and 2) whether "slower-speed" networks could cope
with a software-based reassembly, a less capable hardware-based
reassembly, or the other workarounds. An important future task would
be analyzing the observed incompliant behaviour about DF-bit to note
whether it has any unanticipated drawbacks.
5. IANA Considerations
This document makes no request of IANA.
Note to RFC Editor: this section may be removed on publication as an
RFC.
6. Security Considerations
This document describes different issues with packet sizes and in-
the-network tunneling; this does not have security considerations on
its own.
However, different solutions might have characteristics which may
make them more susceptible to attacks -- for example, a router-based
fragment reassembly could easily lead to (reassembly) buffer memory
exhaustion if the attacker would send a sufficient number of
fragments without sending all of them, so that the reassembly would
be stalled until a timeout; these and other fragment attacks (e.g.,
[13]) have already been used against e.g., firewalls and host stacks,
Savola Expires November 10, 2005 [Page 9]
Internet-Draft Packet Size Issues in Network Tunnels May 2005
and need to be taken into consideration in the implementations.
It is worth considering the cryptographic expense (which is typically
more significant than the reassembly, if done in software) with
fragmentation of the inner or outer packet. If an outer fragment
goes missing, no cryptographic operations have been yet performed; if
an inner fragment goes missing, cryptographic operations have already
been performed. Therefore, which of these approaches is preferable
also depends on whether cryptography or reassembly are already
provided in hardware; for high-speed routers, at least, one should be
able to assume that if it is performing relatively heavy
cryptography, hardware support is already required.
7. Acknowledgements
While the topic is far from new, recent discussions with W. Mark
Townsley on L2TP fragmentation issues caused the author to sit down
and write up the issues in more general. Michael Richardson and Mika
Joutsenvirta provided useful feedback on the first draft. When
soliciting comments from NANOG list, Carsten Bormann, Kevin Miller,
Warren Kumari, Iljitsch van Beijnum, Alok Dube, and Stephen J. Wilcox
provided useful feedback.
8. References
8.1 Normative References
[1] Nordmark, E. and R. Gilligan, "Basic Transition Mechanisms for
IPv6 Hosts and Routers", draft-ietf-v6ops-mech-v2-07 (work in
progress), March 2005.
[2] Conta, A. and S. Deering, "Generic Packet Tunneling in IPv6
Specification", RFC 2473, December 1998.
[3] Farinacci, D., Li, T., Hanks, S., Meyer, D., and P. Traina,
"Generic Routing Encapsulation (GRE)", RFC 2784, March 2000.
[4] Lau, J., Townsley, M., and I. Goyret, "Layer Two Tunneling
Protocol - Version 3 (L2TPv3)", RFC 3931, March 2005.
[5] Kent, S. and R. Atkinson, "Security Architecture for the
Internet Protocol", RFC 2401, November 1998.
[6] Mogul, J. and S. Deering, "Path MTU discovery", RFC 1191,
November 1990.
[7] McCann, J., Deering, S., and J. Mogul, "Path MTU Discovery for
IP version 6", RFC 1981, August 1996.
Savola Expires November 10, 2005 [Page 10]
Internet-Draft Packet Size Issues in Network Tunnels May 2005
[8] Deering, S. and R. Hinden, "Internet Protocol, Version 6 (IPv6)
Specification", RFC 2460, December 1998.
8.2 Informative References
[9] Mamakos, L., Lidl, K., Evarts, J., Carrel, D., Simone, D., and
R. Wheeler, "A Method for Transmitting PPP Over Ethernet
(PPPoE)", RFC 2516, February 1999.
[10] Mathis, M., "Fragmentation Considered Very Harmful",
draft-mathis-frag-harmful-00 (work in progress), July 2004.
[11] Mathis, M., "Path MTU Discovery", draft-ietf-pmtud-method-04
(work in progress), February 2005.
[12] Medina, A., Allman, M., and S. Floyd, "Measuring the Evolution
of Transport Protocols in the Internet", Computer
Communications Review, Apr 2005, <http://www.icir.org/tbit/>.
[13] Miller, I., "Protection Against a Variant of the Tiny Fragment
Attack (RFC 1858)", RFC 3128, June 2001.
Author's Address
Pekka Savola
CSC/FUNET
Espoo
Finland
Email: psavola@funet.fi
Appendix A. MTU of the Tunnel
Different tunneling mechanisms may treat the tunnel links as having
different kind of MTU values. Some might use the same default MTU as
for other interfaces; some others might use the default MTU minus the
expected IP overhead (e.g., 20, 28, or 40 bytes); some others might
even treat the tunnel as having "infinite MTU", e.g., 64 kilobytes.
As [1] describes, having an infinite MTU, i.e., fragmenting the outer
packet (and never the inner packet) and never performing PMTUD is a
very bad idea, especially in host-to-router scenarios. (It could be
argued that if the nodes are sure that this is a host-to-host tunnel,
a larger MTU might make sense if fragmentation and reassembly is more
efficient than just sending properly sized packets -- but this seems
like a stretch.)
Savola Expires November 10, 2005 [Page 11]
Internet-Draft Packet Size Issues in Network Tunnels May 2005
Intellectual Property Statement
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
ietf-ipr@ietf.org.
Disclaimer of Validity
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 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.
Copyright Statement
Copyright (C) The Internet Society (2005). 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.
Acknowledgment
Funding for the RFC Editor function is currently provided by the
Internet Society.
Savola Expires November 10, 2005 [Page 12]