1. IPv4 Fragmentation, PMTU Discovery and NAT
Dr Ram P Rustagi, PESIT
Jun 06, 2013

2. IP addresses: how to get one?
Q: How does a host get IP address?
hard-coded by system admin in a file
Windows: control-panel->network->configuration->tcp/ip- >properties
UNIX: /etc/rc.config
DHCP: Dynamic Host Configuration Protocol: dynamically get address from as server
“plug-and-play”
Network Layer
4-44

3. DHCP: Dynamic Host Configuration Protocol
goal: allow host to dynamically obtain its IP address from network server when it joins network
can renew its lease on address in use, after lease expiry
preferably gets the same address, not guaranteed
client can reuse its address
support for mobile users who want to join network
guarantee one address will be assigned to only one client
retain DHCP client address across reboots
not guaranteed
retain DHCP client configs across server reboots
must coexist with statically assigned addresses
interoperate with BOOTP relay agents
Network Layer
4-45

4. DHCP: Dynamic Host Configuration Protocol
DHCP overview:
an extension of BOOTP mechanism
host broadcasts “DHCP discover” msg [optional]
DHCP server responds with “DHCP offer” msg [optional]
more than one server can make the offer
client can choose which server to send request to
host requests IP address: “DHCP request” msg
DHCP server sends address: “DHCP ack” msg
Renewal happens with DHCP request/ack
On completion, client sends DHCP release msg
practically not seen

5. DHCP: more than IP addresses
DHCP can return more than just allocated IP address on subnet:
address of first-hop router for client
name and IP address of DNS sever
network mask (indicating network versus host portion of address)
Network Layer
4-48

6. DHCP: example
DHCP
UDP IP
Eth
Phy
DHCP
DHCP
connecting laptop needs its IP address, addr of first-hop router, addr of DNS server: use DHCP
DHCP
DHCP request encapsulated in UDP, encapsulated in IP, encapsulated in Ethernet
DHCP
DHCP
UDP IP
Eth
Phy
DHCP
Ethernet frame broadcast (dest: FFFFFFFFFFFF) on LAN, received at router running DHCP server
router with DHCP
server built into
router
Ethernet demuxed to IP demuxed, UDP demuxed to DHCP
src: Kurose & Ross

7. DHCP: example
DHCP
DCP server formulates DHCP ACK containing client’s IP address, IP address of first-hop router for client, name & IP address of DNS server
DHCP
UDP IP
Eth
Phy
encapsulation of DHCP server, frame forwarded to client, demuxing up to DHCP at client
DHCP
UDP IP
Eth
Phy
DHCP
DHCP
router with DHCP
server built into
router
client now knows its IP address, name and IP address of DSN server, IP address of its first-hop router
DHCP
src: Kurose & Ross

8. 32 bit destination IP address
IP datagram format
IP protocol version
number
ver
length
32 bits
data
(variable length,
typically a TCP
or UDP segment)
16-bit identifier
header
checksum
time to
live
32 bit source IP address
head.
len
type of
service
flgs
fragment
offset
upper
layer
32 bit destination IP address
options (if any)
total datagram
length (bytes)
header length
(bytes)
“type” of data
for
fragmentation/
reassembly
max number
remaining hops
(decremented at
each router)
upper layer protocol
to deliver payload to
e.g. timestamp,
record route
taken, specify
list of routers
to visit.
how much overhead?
20 bytes of TCP
20 bytes of IP
= 40 bytes + app layer overhead
src: Kurose & Ross

9. IP fragmentation, reassembly
network links have MTU (max.transfer size) - largest possible link-level frame
different link types, different MTUs
large IP datagram divided (“fragmented”) within net
one datagram becomes several datagrams
“reassembled” only at final destination
IP header bits used to identify, order related fragments
can be refragmented …
fragmentation:
in: one large datagram
out: 3 smaller datagrams
reassembly …
src: Kurose & Ross

10. IP fragmentation, reassembly
MTU for some networks
Hyperchannel
Token Ring (16Mbps)
Token Ring (4Mbps)
FDDI
Ethernet X
PPP - 296

11. IP fragmentation, reassemblyID =x
offset =0
fragflag
length
=4000 x D M
Flags:
D - Don’t Fragment
M - More Fragments
example:
4000 byte datagram
MTU = 1500 bytes ID =x
offset =0
fragflag =1
length
=1500
=185
=370
=1040
one large datagram becomes
several smaller datagrams
1480 bytes in data field
offset =
1480/8
Network Layer
4-36

12. Example : Re-fragmentation
Src: Forouzan - Data Communication and Networking

13. Example : Re-fragmentation
Src: Forouzan - Data Communication and Networking

14. IP fragmentation, reassembly
Questions
A pkt has arrived with an M bit of 0. Is this the first, last of middle segment? Can we say if pkt was fragmented?
A pkt has arrived with an M bit of 1. Is this the first, last of middle segment? Can we say if pkt was fragmented?
A pkt has arrived with an M bit of 1 and offset value 0. Is this the first, last of middle segment? Can we say if pkt was fragmented?
A pkt has arrived with offset value 100, and HLEN value of 5, value of total length field is 100. What are the numbers of the first byte and last byte?

15. IP Fragmentation IP Fragmentation is avoided in general
Path MTU (PMTU) discovery is used
to find the max segment size for a given path
Fragment is a costly process
especially for NAT address
is configured by default in any enterprise

16. ICMP: internet control message protocol RFC 792
Query Messages:
Type Code description
0/ Echo reply/request (ping)
13/ Timestamp request/reply
10/ Router solicitation/advt
Error Reporting Messages
dest. network unreachable
dest host unreachable
dest protocol unreachable
dest port unreachable
dest network unknown
dest host unknown
source quench (congestion
control - not used)
Redirect
TTL expired
bad IP header
used by hosts & routers to communicate network- level information
error reporting: unreachable host, network, port, protocol
echo request/reply (used by ping)
network-layer “above” IP:
ICMP msgs carried in IP datagrams
ICMP message: type, code plus first 8 bytes of IP datagram causing error
Network Layer
4-65

17. ICMP Redirect
Src: Forouzan - Data Communication and Networking

18. Traceroute and ICMP source sends series of UDP segments to dest
first set has TTL =1
second set has TTL=2, etc.
unlikely port number
when nth set of datagrams arrives to nth router:
router discards datagrams
and sends source ICMP messages (type 11, code 0)
ICMP messages includes name of router & IP address
when ICMP messages arrives, source records RTTs
stopping criteria:
UDP segment eventually arrives at destination host
destination returns ICMP “port unreachable” message (type 3, code 3)
source stops
3 probes
3 probes
3 probes
src: Kurose & Ross

19. PMTU Discovery PMTU Discovery
source finds out the max fragment size on the path to destination
it is actually a misnomer
no actual discovery packet is sent from source
source sets ‘D’ (don’t fragment) bit in IP header
when a router is required to fragment
i.e. MTU of o/g link is small
packet is dropped and ICMP error is sent to source
with the supported fragment size
source retransmits the packet with smaller fragment size
PMTU discovery continues till packet reaches destination

20. PMTU Discovery Implementation on Linux machines enabled by default
$ sysctl net.ipv4.ip_no_pmtu_disc
net.ipv4.ip_no_pmtu_disc = 0
no PMTU discovery is disabled i.e.
PMTU discovery is enabled
To disable PMTU discovery
$sudo sysctl -w net.ipv4.ip_no_pmtu_disc=1

21. NAT: network address translation
rest of
Internet
local network
(e.g., home network)
10.0.0/24
all datagrams leaving local
network have same single source NAT IP address: ,different source port numbers
datagrams with source or
destination in this network
have /24 address for
source, destination (as usual)
src: Kurose & Ross

22. NAT: network address translation
motivation: local network uses just one IP address as far as outside world is concerned:
range of addresses not needed from ISP: just one IP address for all devices
can change addresses of devices in local network without notifying outside world
can change ISP without changing addresses of devices in local network
devices inside local net not explicitly addressable, visible by outside world (a security plus)
Network Layer
4-57

23. NAT: network address translation
implementation: NAT router must:
outgoing datagrams: replace (source IP address, port #) of every outgoing datagram to (NAT IP address, new port #)
. . . remote clients/servers will respond using (NAT IP address, new port #) as destination addr
remember (in NAT translation table) every (source IP address, port #) to (NAT IP address, new port #) translation pair
incoming datagrams: replace (NAT IP address, new port #) in dest fields of every incoming datagram with corresponding (source IP address, port #) stored in NAT table
Network Layer
4-58

24. Address Allocation for private networks
RFC 1918
One Class A network
(224)
16 Class B Networks
(220)
256 Class C Networks
(216)

25. Private addresses without NAT

26. Special purpose addresses
Loopback
address

27. NAT: network address translation
NAT translation table
WAN side addr LAN side addr
2: NAT router
changes datagram
source addr from
, 3345 to
, 5001,
updates table
+Protocol
1: host
sends datagram to
, 80
, , 3345
…… ……
S: , 3345
D: , 80 1
S: , 80
D: , 3345 4
S: , 5001
D: , 80 2
S: , 80
D: , 5001 3
4: NAT router
changes datagram
dest addr from
, 5001 to , 3345
3: reply arrives
dest. address:
, 5001
src: Kurose & Ross
Network Layer
4-59

28. NAT: network address translation
16-bit port-number field:
simultaneous connections on one NAT address!
65,535
NAT is controversial:
routers should only process up to layer 3
violates end-to-end argument
NAT possibility must be taken into account by app designers, e.g., P2P applications
address shortage should instead be solved by IPv6
gets complicated with IP fragmentation
Network Layer
4-60

29. NAT traversal problem
client wants to connect to server with address
server address local to LAN (client can’t use it as destination addr)
only one externally visible NATed address:
solution1: statically configure NAT to forward incoming connection requests at given port to server
e.g., ( , port 2500) always forwarded to port 25000
client ?
NAT
router
src: Kurose & Ross
Network Layer
4-61

30. NAT traversal problem
solution 2: Universal Plug and Play (UPnP) Internet Gateway Device (IGD) Protocol. Allows NATed host to:
learn public IP address ( )
add/remove port mappings (with lease times)
i.e., automate static NAT port map configuration
NAT
router
IGD
src: Kurose & Ross
Network Layer
4-62

31. NAT traversal problem solution 3: relaying (used in Skype)
NATed client establishes connection to relay
external client connects to relay
relay bridges packets between to connections
2. connection to
relay initiated
by client
NAT
router
1. connection to
relay initiated
by NATed host
3. relaying
established
client
src: Kurose & Ross
Network Layer
4-63

32. NAT: network address translation
Two type of NAT
SNAT (Source NAT)
Destination NAT
NAT for TCP
Data entering from outside is allowed after connection establishment
NAT Entry is removed for connection close
NAT for UDP
makes a short entry for small duration (30s ?)
Good Tutorial for Linux NAT
tutorial.html#NATINTRO

33. Thank You