iptables and dhcp

I am implementing a firewall and am seeing the following problem:

DHCP requests are passing through my firewall to the next higher server in
the network. These requests seem to be alternatively serviced by my
firewall and the server above my firewall.

I've tried blocking ports 67 and 68 with iptables, and it still gets passed
through and serviced. I've tried blocking everything in both directions and
it still gets passed through.

Any ideas?

In article <(E-Mail Removed) m>,
Dave Lister wrote:
> DHCP requests are passing through my firewall to the next higher server in
> the network. These requests seem to be alternatively serviced by my

Are you running a DHCP relay? Or are these would-be DHCP clients on the
same physical network segment as the firewall and the remote DHCP
server?

> I've tried blocking ports 67 and 68 with iptables, and it still gets passed
> through and serviced. I've tried blocking everything in both directions and
> it still gets passed through.
>
> Any ideas?

I'll bet if you showed us your rules we could show you exactly where and
why they're wrong. Here are some guesses in advance:
1. You used INPUT and not FORWARD chain
2. You used -p tcp and not -p udp
--
/dev/rob0 - preferred_email=i$((28*28+28))@softhome.net
or put "not-spam" or "/dev/rob0" in Subject header to reply

/dev/rob0 <(E-Mail Removed)> wrote in
news:(E-Mail Removed):

> In article <(E-Mail Removed) m>,
> Dave Lister wrote:
>> DHCP requests are passing through my firewall to the next higher
>> server in the network. These requests seem to be alternatively
>> serviced by my
>
> Are you running a DHCP relay? Or are these would-be DHCP clients on
> the same physical network segment as the firewall and the remote DHCP
> server?

No DHCP relay.

They are not on the same physical segment: my firewall sits in between.

>> I've tried blocking ports 67 and 68 with iptables, and it still gets
>> passed through and serviced. I've tried blocking everything in both
>> directions and it still gets passed through.
>>
>> Any ideas?
>
> I'll bet if you showed us your rules we could show you exactly where
> and why they're wrong. Here are some guesses in advance:
> 1. You used INPUT and not FORWARD chain
> 2. You used -p tcp and not -p udp

No, but I will post the rules. I also tried it with no rules in place, a
drop policy, and the dhcp still went through.

#!/bin/sh
#
# Generated iptables firewall script for the Linux 2.4 kernel
# Script generated by Easy Firewall Generator for IPTables
# copyright 2002 Timothy Scott Morizot
#
# Redhat chkconfig comments - firewall applied early,
# removed late
# chkconfig: 2345 08 92
# description: This script applies or removes iptables firewall rules
#
# This generator is primarily designed for RedHat installations,
# although it should be adaptable for others.
#
# It can be executed with the typical start and stop arguments.
# If used with stop, it will stop after flushing the firewall.
# The save and restore arguments will save or restore the rules
# from the /etc/sysconfig/iptables file. The save and restore
# arguments are included to preserve compatibility with
# Redhat's init.d script (at least in 7.x) if you prefer to use it.

# Redhat installation instructions
#
# 1. Ensure that ipchains will not automatically start.
# chkconfig --level 0123456 ipchains off
# This will make sure that the ipchains init.d script
# is not linked to an S file in any of the rc directories.
#
# 2. Stop ipchains if it's running.
# service ipchains stop
#
# 3. Execute lsmod to see if the ipchains kernel module is still loaded.
# If it is, use rmmod to unload it. -- rmmod ipchains
#
# 4. Have the system link the iptables init.d startup script into run
states
# 2, 3, and 5.
# chkconfig --level 235 iptables on
#
# 5. Save this script and execute it to load the ruleset from this file.
# You may need to run the dos2unix command on it to remove carraige
returns.
#
# 6. To have it applied at startup, copy this script to
# /etc/init.d/iptables. It accepts stop, start, save, and restore
# arguments. (You may wish to save the existing one first.)
#
# 7. For non-Redhat systems (or Redhat systems if you have a problem),
you
# may want to append the command to execute this script to rc.local.
# rc.local is typically located in /etc and /etc/rc.d and is usually
# the last thing executed on startup. Simply add
/path/to/script/script_name
# on its own line in the rc.local file.

################################################## #######################
######
#
# Local Settings
#

# sysctl location. If set, it will use sysctl to adjust the kernel
parameters.
# If this is set to the empty string (or is unset), the use of sysctl
# is disabled.

# SYSCTL="/sbin/sysctl -w"

# To echo the value directly to the /proc file instead
SYSCTL=""

# IPTables Location - adjust if needed

IPT="/sbin/iptables"
IPTS="/sbin/iptables-save"
IPTR="/sbin/iptables-restore"

# Internet Interface
INET_IFACE="ixp0"

# Local Interface Information
LOCAL_IFACE="br0"
LOCAL_IP="192.168.15.1"
LOCAL_NET="192.168.15.0/24"
LOCAL_BCAST="192.168.15.255"

# Localhost Interface

LO_IFACE="lo"
LO_IP="127.0.0.1"

# Save and Restore arguments handled here
if [ "$1" = "save" ]
then
echo -n "Saving firewall to /etc/sysconfig/iptables ... "
$IPTS > /etc/sysconfig/iptables
echo "done"
exit 0
elif [ "$1" = "restore" ]
then
echo -n "Restoring firewall from /etc/sysconfig/iptables ... "
$IPTR < /etc/sysconfig/iptables
echo "done"
exit 0
fi

################################################## #######################
######
#
# Load Modules
#

echo "Loading kernel modules ..."

# You should uncomment the line below and run it the first time just to
# ensure all kernel module dependencies are OK. There is no need to run
# every time, however.

# /sbin/depmod -a

# Unless you have kernel module auto-loading disabled, you should not
# need to manually load each of these modules. Other than ip_tables,
# ip_conntrack, and some of the optional modules, I've left these
# commented by default. Uncomment if you have any problems or if
# you have disabled module autoload. Note that some modules must
# be loaded by another kernel module.

# core netfilter module
# /sbin/modprobe ip_tables

# the stateful connection tracking module
# /sbin/modprobe ip_conntrack

# filter table module
# /sbin/modprobe iptable_filter

# mangle table module
# /sbin/modprobe iptable_mangle

# nat table module
# /sbin/modprobe iptable_nat

# LOG target module
# /sbin/modprobe ipt_LOG

# This is used to limit the number of packets per sec/min/hr
# /sbin/modprobe ipt_limit

# masquerade target module
# /sbin/modprobe ipt_MASQUERADE

# filter using owner as part of the match
# /sbin/modprobe ipt_owner

# REJECT target drops the packet and returns an ICMP response.
# The response is configurable. By default, connection refused.
# /sbin/modprobe ipt_REJECT

# This target allows packets to be marked in the mangle table
# /sbin/modprobe ipt_mark

# This target affects the TCP MSS
# /sbin/modprobe ipt_tcpmss

# This match allows multiple ports instead of a single port or range
# /sbin/modprobe multiport

# This match checks against the TCP flags
# /sbin/modprobe ipt_state

# This match catches packets with invalid flags
# /sbin/modprobe ipt_unclean

# The ftp nat module is required for non-PASV ftp support
# /sbin/modprobe ip_nat_ftp

# the module for full ftp connection tracking
# /sbin/modprobe ip_conntrack_ftp

# the module for full irc connection tracking
# /sbin/modprobe ip_conntrack_irc


################################################## #######################
######
#
# Kernel Parameter Configuration
#
# See http://ipsysctl-tutorial.frozentux.n...tml/index.html
# for a detailed tutorial on sysctl and the various settings
# available.

# Required to enable IPv4 forwarding.
# Redhat users can try setting FORWARD_IPV4 in /etc/sysconfig/network to
true
# Alternatively, it can be set in /etc/sysctl.conf
if [ "$SYSCTL" = "" ]
then
echo "1" > /proc/sys/net/ipv4/ip_forward
else
$SYSCTL net.ipv4.ip_forward="1"
fi

# This enables dynamic address hacking.
# This may help if you have a dynamic IP address \(e.g. slip, ppp, dhcp
\).
#if [ "$SYSCTL" = "" ]
#then
# echo "1" > /proc/sys/net/ipv4/ip_dynaddr
#else
# $SYSCTL net.ipv4.ip_dynaddr="1"
#fi

# This enables SYN flood protection.
# The SYN cookies activation allows your system to accept an unlimited
# number of TCP connections while still trying to give reasonable
# service during a denial of service attack.
if [ "$SYSCTL" = "" ]
then
echo "1" > /proc/sys/net/ipv4/tcp_syncookies
else
$SYSCTL net.ipv4.tcp_syncookies="1"
fi

# This enables source validation by reversed path according to RFC1812.
# In other words, did the response packet originate from the same
interface
# through which the source packet was sent? It's recommended for single-
homed
# systems and routers on stub networks. Since those are the
configurations
# this firewall is designed to support, I turn it on by default.
# Turn it off if you use multiple NICs connected to the same network.
if [ "$SYSCTL" = "" ]
then
echo "1" > /proc/sys/net/ipv4/conf/all/rp_filter
else
$SYSCTL net.ipv4.conf.all.rp_filter="1"
fi

# This option allows a subnet to be firewalled with a single IP address.
# It's used to build a DMZ. Since that's not a focus of this firewall
# script, it's not enabled by default, but is included for reference.
# See: http://www.sjdjweis.com/linux/proxyarp/
#if [ "$SYSCTL" = "" ]
#then
# echo "1" > /proc/sys/net/ipv4/conf/all/proxy_arp
#else
# $SYSCTL net.ipv4.conf.all.proxy_arp="1"
#fi

# The following kernel settings were suggested by Alex Weeks. Thanks!

# This kernel parameter instructs the kernel to ignore all ICMP
# echo requests sent to the broadcast address. This prevents
# a number of smurfs and similar DoS nasty attacks.
if [ "$SYSCTL" = "" ]
then
echo "1" > /proc/sys/net/ipv4/icmp_echo_ignore_broadcasts
else
$SYSCTL net.ipv4.icmp_echo_ignore_broadcasts="1"
fi

# This option can be used to accept or refuse source routed
# packets. It is usually on by default, but is generally
# considered a security risk. This option turns it off.
if [ "$SYSCTL" = "" ]
then
echo "0" > /proc/sys/net/ipv4/conf/all/accept_source_route
else
$SYSCTL net.ipv4.conf.all.accept_source_route="0"
fi

# This option can disable ICMP redirects. ICMP redirects
# are generally considered a security risk and shouldn't be
# needed by most systems using this generator.
#if [ "$SYSCTL" = "" ]
#then
# echo "0" > /proc/sys/net/ipv4/conf/all/accept_redirects
#else
# $SYSCTL net.ipv4.conf.all.accept_redirects="0"
#fi

# However, we'll ensure the secure_redirects option is on instead.
# This option accepts only from gateways in the default gateways list.
if [ "$SYSCTL" = "" ]
then
echo "1" > /proc/sys/net/ipv4/conf/all/secure_redirects
else
$SYSCTL net.ipv4.conf.all.secure_redirects="1"
fi

# This option logs packets from impossible addresses.
if [ "$SYSCTL" = "" ]
then
echo "1" > /proc/sys/net/ipv4/conf/all/log_martians
else
$SYSCTL net.ipv4.conf.all.log_martians="1"
fi


################################################## #######################
######
#
# Flush Any Existing Rules or Chains
#

echo "Flushing Tables ..."

# Reset Default Policies
$IPT -P INPUT ACCEPT
$IPT -P FORWARD ACCEPT
$IPT -P OUTPUT ACCEPT
$IPT -t nat -P PREROUTING ACCEPT
$IPT -t nat -P POSTROUTING ACCEPT
$IPT -t nat -P OUTPUT ACCEPT
$IPT -t mangle -P PREROUTING ACCEPT
$IPT -t mangle -P OUTPUT ACCEPT

# Flush all rules
$IPT -F
$IPT -t nat -F
$IPT -t mangle -F

# Erase all non-default chains
$IPT -X
$IPT -t nat -X
$IPT -t mangle -X

if [ "$1" = "stop" ]
then
echo "Firewall completely flushed! Now running with no firewall."
exit 0
fi

################################################## #######################
######
#
# Rules Configuration
#

################################################## #######################
######
#
# Filter Table
#
################################################## #######################
######

# Set Policies

$IPT -P INPUT DROP
$IPT -P OUTPUT DROP
$IPT -P FORWARD DROP

################################################## #######################
######
#
# User-Specified Chains
#
# Create user chains to reduce the number of rules each packet
# must traverse.

echo "Create and populate custom rule chains ..."

# Create a chain to filter INVALID packets

$IPT -N bad_packets

# Create another chain to filter bad tcp packets

$IPT -N bad_tcp_packets

# Create separate chains for icmp, tcp (incoming and outgoing),
# and incoming udp packets.

$IPT -N icmp_packets

# Used for UDP packets inbound from the Internet
$IPT -N udp_inbound

# Used to block outbound UDP services from internal network
# Default to allow all
$IPT -N udp_outbound

# Used to allow inbound services if desired
# Default fail except for established sessions
$IPT -N tcp_inbound

# Used to block outbound services from internal network
# Default to allow all
$IPT -N tcp_outbound

################################################## #######################
######
#
# Populate User Chains
#

# bad_packets chain
#
# Drop INVALID packets immediately

$IPT -A bad_packets -p ALL -m state --state INVALID -j LOG \
--log-prefix "Invalid packet: "

$IPT -A bad_packets -p ALL -m state --state INVALID -j DROP

# Then check the tcp packets for additional problems
$IPT -A bad_packets -p tcp -j bad_tcp_packets

# All good, so return
$IPT -A bad_packets -p ALL -j RETURN

# bad_tcp_packets chain
#
# All tcp packets will traverse this chain.
# Every new connection attempt should begin with
# a syn packet. If it doesn't, it is likely a
# port scan. This drops packets in state
# NEW that are not flagged as syn packets.

# Return to the calling chain if the bad packets originate
# from the local interface. This maintains the approach
# throughout this firewall of a largely trusted internal
# network.
$IPT -A bad_tcp_packets -p tcp -i $LOCAL_IFACE -j RETURN

# However, I originally did apply this filter to the forward chain
# for packets originating from the internal network. While I have
# not conclusively determined its effect, it appears to have the
# interesting side effect of blocking some of the ad systems.
# Apparently some ad systems have the browser initiate a NEW
# connection that is not flagged as a syn packet to retrieve
# the ad image. If you wish to experiment further comment the
# rule above. If you try it, you may also wish to uncomment the
# rule below. It will keep those packets from being logged.
# There are a lot of them.
# $IPT -A bad_tcp_packets -p tcp -i $LOCAL_IFACE ! --syn -m state \
# --state NEW -j DROP

$IPT -A bad_tcp_packets -p tcp ! --syn -m state --state NEW -j LOG \
--log-prefix "New not syn: "
$IPT -A bad_tcp_packets -p tcp ! --syn -m state --state NEW -j DROP

# All good, so return
$IPT -A bad_tcp_packets -p tcp -j RETURN

# icmp_packets chain
#
# This chain is for inbound (from the Internet) icmp packets only.
# Type 8 (Echo Request) is not accepted by default
# Enable it if you want remote hosts to be able to reach you.
# 11 (Time Exceeded) is the only one accepted
# that would not already be covered by the established
# connection rule. Applied to INPUT on the external interface.
#
# See: http://www.ee.siue.edu/~rwalden/networking/icmp.html
# for more info on ICMP types.
#
# Note that the stateful settings allow replies to ICMP packets.
# These rules allow new packets of the specified types.

# ICMP packets should fit in a Layer 2 frame, thus they should
# never be fragmented. Fragmented ICMP packets are a typical sign
# of a denial of service attack.
$IPT -A icmp_packets --fragment -p ICMP -j LOG \
--log-prefix "ICMP Fragment: "
$IPT -A icmp_packets --fragment -p ICMP -j DROP

# Echo - uncomment to allow your system to be pinged.
# Uncomment the LOG command if you also want to log PING attempts
#
# $IPT -A icmp_packets -p ICMP -s 0/0 --icmp-type 8 -j LOG \
# --log-prefix "Ping detected: "
# $IPT -A icmp_packets -p ICMP -s 0/0 --icmp-type 8 -j ACCEPT

# Time Exceeded
$IPT -A icmp_packets -p ICMP -s 0/0 --icmp-type 11 -j ACCEPT

# Not matched, so return so it will be logged
$IPT -A icmp_packets -p ICMP -j RETURN

# TCP & UDP
# Identify ports at:
# http://www.chebucto.ns.ca/~rakerman/port-table.html
# http://www.iana.org/assignments/port-numbers

# udp_inbound chain
#
# This chain describes the inbound UDP packets it will accept.
# It's applied to INPUT on the external or Internet interface.
# Note that the stateful settings allow replies.
# These rules are for new requests.
# It drops netbios packets (windows) immediately without logging.

# Drop netbios calls
# Please note that these rules do not really change the way the firewall
# treats netbios connections. Connections from the localhost and
# internal interface (if one exists) are accepted by default.
# Responses from the Internet to requests initiated by or through
# the firewall are also accepted by default. To get here, the
# packets would have to be part of a new request received by the
# Internet interface. You would have to manually add rules to
# accept these. I added these rules because some network connections,
# such as those via cable modems, tend to be filled with noise from
# unprotected Windows machines. These rules drop those packets
# quickly and without logging them. This prevents them from traversing
# the whole chain and keeps the log from getting cluttered with
# chatter from Windows systems.
$IPT -A udp_inbound -p UDP -s 0/0 --destination-port 137 -j DROP
$IPT -A udp_inbound -p UDP -s 0/0 --destination-port 138 -j DROP

# Dynamic Address
# If DHCP, the initial request is a broadcast. The response
# doesn't exactly match the outbound packet. This explicitly
# allow the DHCP ports to alleviate this problem.
# If you receive your dynamic address by a different means, you
# can probably comment this line.
$IPT -A udp_inbound -p UDP -s 0/0 --source-port 67 --destination-port 68
\
-j ACCEPT


# Not matched, so return for logging
$IPT -A udp_inbound -p UDP -j RETURN

# udp_outbound chain
#
# This chain is used with a private network to prevent forwarding for
# UDP requests on specific protocols. Applied to the FORWARD rule from
# the internal network. Ends with an ACCEPT

$IPT -A udp_outbound -p UDP -s 0/0 --destination-port 67 -j REJECT
$IPT -A udp_outbound -p UDP -s 0/0 --destination-port 68 -j REJECT

# No match, so ACCEPT
$IPT -A udp_outbound -p UDP -s 0/0 -j ACCEPT

# tcp_inbound chain
#
# This chain is used to allow inbound connections to the
# system/gateway. Use with care. It defaults to none.
# It's applied on INPUT from the external or Internet interface.


# Not matched, so return so it will be logged
$IPT -A tcp_inbound -p TCP -j RETURN

# tcp_outbound chain
#
# This chain is used with a private network to prevent forwarding for
# requests on specific protocols. Applied to the FORWARD rule from
# the internal network. Ends with an ACCEPT

# Block Outbound Telnet
$IPT -A tcp_outbound -p TCP -s 0/0 --destination-port 67 -j REJECT
$IPT -A tcp_outbound -p TCP -s 0/0 --destination-port 68 -j REJECT


# No match, so ACCEPT
$IPT -A tcp_outbound -p TCP -s 0/0 -j ACCEPT

################################################## #######################
######
#
# INPUT Chain
#

echo "Process INPUT chain ..."

# Allow all on localhost interface
$IPT -A INPUT -p ALL -i $LO_IFACE -j ACCEPT

# Drop bad packets
$IPT -A INPUT -p ALL -j bad_packets

# DOCSIS compliant cable modems
# Some DOCSIS compliant cable modems send IGMP multicasts to find
# connected PCs. The multicast packets have the destination address
# 224.0.0.1. You can accept them. If you choose to do so,
# Uncomment the rule to ACCEPT them and comment the rule to DROP
# them The firewall will drop them here by default to avoid
# cluttering the log. The firewall will drop all multicasts
# to the entire subnet (224.0.0.1) by default. To only affect
# IGMP multicasts, change '-p ALL' to '-p 2'. Of course,
# if they aren't accepted elsewhere, it will only ensure that
# multicasts on other protocols are logged.
# Drop them without logging.
$IPT -A INPUT -p ALL -d 224.0.0.1 -j DROP
# The rule to accept the packets.
# $IPT -A INPUT -p ALL -d 224.0.0.1 -j ACCEPT

# Rules for the private network (accessing gateway system itself)
$IPT -A INPUT -p ALL -i $LOCAL_IFACE -s $LOCAL_NET -j ACCEPT
$IPT -A INPUT -p ALL -i $LOCAL_IFACE -d $LOCAL_BCAST -j ACCEPT

# Allow DHCP client request packets inbound from internal network
$IPT -A INPUT -p UDP -i $LOCAL_IFACE --source-port 68 --destination-port
67 \
-j ACCEPT


# Inbound Internet Packet Rules

# Accept Established Connections
$IPT -A INPUT -p ALL -i $INET_IFACE -m state --state ESTABLISHED,RELATED
\
-j ACCEPT

# Route the rest to the appropriate user chain
$IPT -A INPUT -p TCP -i $INET_IFACE -j tcp_inbound
$IPT -A INPUT -p UDP -i $INET_IFACE -j udp_inbound
$IPT -A INPUT -p ICMP -i $INET_IFACE -j icmp_packets

# Drop without logging broadcasts that get this far.
# Cuts down on log clutter.
# Comment this line if testing new rules that impact
# broadcast protocols.
$IPT -A INPUT -p ALL -d 255.255.255.255 -j DROP

# Log packets that still don't match
$IPT -A INPUT -m limit --limit 3/minute --limit-burst 3 -j LOG \
--log-prefix "INPUT packet died: "

################################################## #######################
######
#
# FORWARD Chain
#

echo "Process FORWARD chain ..."

# Used if forwarding for a private network

# Drop bad packets
$IPT -A FORWARD -p ALL -j bad_packets

# Accept TCP packets we want to forward from internal sources
$IPT -A FORWARD -p tcp -i $LOCAL_IFACE -j tcp_outbound

# Accept UDP packets we want to forward from internal sources
$IPT -A FORWARD -p udp -i $LOCAL_IFACE -j udp_outbound

# If not blocked, accept any other packets from the internal interface
$IPT -A FORWARD -p ALL -i $LOCAL_IFACE -j ACCEPT

# Deal with responses from the internet
$IPT -A FORWARD -i $INET_IFACE -m state --state ESTABLISHED,RELATED \
-j ACCEPT

# Log packets that still don't match
$IPT -A FORWARD -m limit --limit 3/minute --limit-burst 3 -j LOG \
--log-prefix "FORWARD packet died: "

################################################## #######################
######
#
# OUTPUT Chain
#

echo "Process OUTPUT chain ..."

# Generally trust the firewall on output

# However, invalid icmp packets need to be dropped
# to prevent a possible exploit.
$IPT -A OUTPUT -m state -p icmp --state INVALID -j DROP

# Localhost
$IPT -A OUTPUT -p ALL -s $LO_IP -j ACCEPT
$IPT -A OUTPUT -p ALL -o $LO_IFACE -j ACCEPT

# To internal network
$IPT -A OUTPUT -p ALL -s $LOCAL_IP -j ACCEPT
$IPT -A OUTPUT -p ALL -o $LOCAL_IFACE -j ACCEPT

# To internet
$IPT -A OUTPUT -p ALL -o $INET_IFACE -j ACCEPT

# Log packets that still don't match
$IPT -A OUTPUT -m limit --limit 3/minute --limit-burst 3 -j LOG \
--log-prefix "OUTPUT packet died: "

################################################## #######################
######
#
# nat table
#
################################################## #######################
######

# The nat table is where network address translation occurs if there
# is a private network. If the gateway is connected to the Internet
# with a static IP, snat is used. If the gateway has a dynamic address,
# masquerade must be used instead. There is more overhead associated
# with masquerade, so snat is better when it can be used.
# The nat table has a builtin chain, PREROUTING, for dnat and redirects.
# Another, POSTROUTING, handles snat and masquerade.

echo "Load rules for nat table ..."

################################################## #######################
######
#
# PREROUTING chain
#


################################################## #######################
######
#
# POSTROUTING chain
#

$IPT -t nat -A POSTROUTING -o $INET_IFACE -j MASQUERADE

################################################## #######################
######
#
# mangle table
#
################################################## #######################
######

# The mangle table is used to alter packets. It can alter or mangle them
in
# several ways. For the purposes of this generator, we only use its
ability
# to alter the TTL in packets. However, it can be used to set netfilter
# mark values on specific packets. Those marks could then be used in
another
# table like filter, to limit activities associated with a specific host,
for
# instance. The TOS target can be used to set the Type of Service field
in
# the IP header. Note that the TTL target might not be included in the
# distribution on your system. If it is not and you require it, you will
# have to add it. That may require that you build from source.

echo "Load rules for mangle table ..."

>
> They are not on the same physical segment: my firewall sits in between.

Since you're using a bridging firewall they are on the same IP broadcast
domain. If you use a routing firewall, DHCP broadcasts won't pass through.

>
> >> I've tried blocking ports 67 and 68 with iptables, and it still gets
> >> passed through and serviced. I've tried blocking everything in both
> >> directions and it still gets passed through.
> >>
> >> Any ideas?
> >

Iptables can't filter DHCP packets for some reason that I've never seen
adequately explained. If you can't switch to a routing firewall, you might
have to resort to MAC address matching at the DHCP server to prevent it
assigning addresses to machines beyond the firewall.

"W Cardwell" <(E-Mail Removed)> wrote in
news:lHE9b.3805$(E-Mail Removed):

>>
>> They are not on the same physical segment: my firewall sits in
>> between.
>
> Since you're using a bridging firewall they are on the same IP
> broadcast domain. If you use a routing firewall, DHCP broadcasts won't
> pass through.

How do I switch? How can you tell that I'm doing a bridging firewall?

In article <(E-Mail Removed) m>,
Dave Lister wrote:
>> Since you're using a bridging firewall they are on the same IP
>> broadcast domain. If you use a routing firewall, DHCP broadcasts won't
>> pass through.
>
> How do I switch?

I don't know! Who set it up? Maybe you could scrap it and go for
something like smoothwall or Clark Connect.

> How can you tell that I'm doing a bridging firewall?

That huge firewall script you posted said that the internal interface
was "br0".
--
/dev/rob0 - preferred_email=i$((28*28+28))@softhome.net
or put "not-spam" or "/dev/rob0" in Subject header to reply

> How do I switch? How can you tell that I'm doing a bridging firewall?

The local interface is of type 'br' (bridge). Bridging requires a customized
kernel, and I'm not sure you wouldn't have to use a different kernel and
alter your boot loader configuration to switch back to using your NICs in
the standard way. Perhaps someone else could help out here.

In any case, you would want to not load the bridge service or bring up the
bridge interface at startup, and you would instead assign the IP address
information to 'eth0' or whatever real LAN interface you use on the local
side. You would also use different IP network address ranges on opposite
sides of the firewall.

--

wcardwell at nc dot rr dot com

"W Cardwell" <(E-Mail Removed)> wrote in
news:y0I9b.3853$(E-Mail Removed):

>> How do I switch? How can you tell that I'm doing a bridging firewall?
>
> The local interface is of type 'br' (bridge). Bridging requires a
> customized kernel, and I'm not sure you wouldn't have to use a
> different kernel and alter your boot loader configuration to switch
> back to using your NICs in the standard way. Perhaps someone else
> could help out here.

Why doesn't it just route between ixp0 and br0? I have two wireless and one
wired LAN bridged together to form br0. Won't the router work between br0
and ixp0?

> In any case, you would want to not load the bridge service or bring up
> the bridge interface at startup, and you would instead assign the IP
> address information to 'eth0' or whatever real LAN interface you use
> on the local side. You would also use different IP network address
> ranges on opposite sides of the firewall.

I am using different IP network addresses on both sides of the firewall. I
am puzzled as to why the router doesn't see br0 as a single LAN interface,
and not route dhcp broadcast packets between the WAN (ixp0) and the LAN.

/dev/rob0 <(E-Mail Removed)> wrote in news:(E-Mail Removed):

>> How can you tell that I'm doing a bridging firewall?
>
> That huge firewall script you posted said that the internal interface
> was "br0".

I'm bridging the LAN side together, yes. But it should route properly
between br0 and ixp0, no?

On 2003-09-16, Dave Lister <(E-Mail Removed)> wrote:

> I am implementing a firewall and am seeing the following problem:
>
> DHCP requests are passing through my firewall to the next higher server in
> the network. These requests seem to be alternatively serviced by my
> firewall and the server above my firewall.

Can you restrict the subnet mask in your dhcpd.conf to only the
machines behind the firewall?

--

-John ((E-Mail Removed))