shorewall_code/manpages6/shorewall6-tcrules.xml

627 lines
27 KiB
XML
Raw Normal View History

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE refentry PUBLIC "-//OASIS//DTD DocBook XML V4.4//EN"
"http://www.oasis-open.org/docbook/xml/4.4/docbookx.dtd">
<refentry>
<refmeta>
<refentrytitle>shorewall6-tcrules</refentrytitle>
<manvolnum>5</manvolnum>
</refmeta>
<refnamediv>
<refname>tcrules</refname>
<refpurpose>Shorewall6 Packet Marking rules file</refpurpose>
</refnamediv>
<refsynopsisdiv>
<cmdsynopsis>
<command>/etc/shorewall6/rules</command>
</cmdsynopsis>
</refsynopsisdiv>
<refsect1>
<title>Description</title>
<para>Entries in this file cause packets to be marked as a means of
classifying them for traffic control or policy routing.</para>
<important>
<para>Unlike rules in the <ulink
url="shorewall6-rules.html">shorewall6-rules</ulink>(5) file, evaluation
of rules in this file will continue after a match. So the final mark for
each packet will be the one assigned by the LAST tcrule that
matches.</para>
<para>If you use multiple internet providers with the 'track' option, in
/etc/shorewall6/providers be sure to read the restrictions at <ulink
url="http://shorewall.net/MultiISP.html">http://shorewall.net/MultiISP.html</ulink>.</para>
</important>
<para>The columns in the file are as follows.</para>
<variablelist>
<varlistentry>
<term><emphasis role="bold">MARK/CLASSIFY</emphasis> -
{<emphasis>value</emphasis>|<emphasis>major</emphasis><emphasis
role="bold">:</emphasis><emphasis>minor</emphasis>|<emphasis
role="bold">RESTORE</emphasis>[<emphasis
role="bold">/</emphasis><emphasis>mask</emphasis>]|<emphasis
role="bold">SAVE</emphasis>[<emphasis
role="bold">/</emphasis><emphasis>mask</emphasis>]|<emphasis
role="bold">CONTINUE</emphasis>|<emphasis
role="bold">COMMENT</emphasis>}[<emphasis
role="bold">:</emphasis>{<emphasis role="bold">C</emphasis>|<emphasis
role="bold">F</emphasis>|<emphasis role="bold">P</emphasis>|<emphasis
role="bold">T</emphasis>|<emphasis role="bold">CF</emphasis>|<emphasis
role="bold">CP</emphasis>|<emphasis role="bold">CT</emphasis>}]</term>
<listitem>
<para>May assume one of the following values.</para>
<orderedlist numeration="arabic">
<listitem>
<para>A mark <emphasis>value</emphasis> which is an integer in
the range 1-255.</para>
<para>Normally will set the mark value. If preceded by a
vertical bar ("|"), the mark value will be logically ORed with
the current mark value to produce a new mark value. If preceded
by an ampersand ("&amp;"), will be logically ANDed with the
current mark value to produce a new mark value.</para>
<para>Both "|" and "&amp;" require Extended MARK Target support
in your kernel and ip6tables; neither may be used with connection
marks (see below).</para>
<para>May optionally be followed by <emphasis
role="bold">:P</emphasis>, <emphasis role="bold">:F</emphasis>
or <emphasis role="bold">:T</emphasis> where<emphasis
role="bold"> :P</emphasis> indicates that marking should occur
in the PREROUTING chain, <emphasis role="bold">:F</emphasis>
indicates that marking should occur in the FORWARD chain and
<emphasis role="bold">:T</emphasis> indicates that marking
should occur in the POSTROUTING chain. If neither <emphasis
role="bold">:P</emphasis>, <emphasis role="bold">:F</emphasis>
nor <emphasis role="bold">:T</emphasis> follow the mark value
then the chain is determined as follows:</para>
<para>- If the SOURCE is <emphasis
role="bold">$FW</emphasis>[<emphasis
role="bold">:</emphasis><emphasis>address-or-range</emphasis>[,<emphasis>address-or-range</emphasis>]...],
then the rule is inserted into the OUTPUT chain. The behavior
changed in Shorewall6-perl 4.1. Only high mark values may be
assigned in this case. Packet marking rules for traffic shaping
of packets originating on the firewall must be coded in the
POSTROUTING chain (see below).</para>
<para>- Otherwise, the chain is determined by the setting of
MARK_IN_FORWARD_CHAIN in <ulink
url="shorewall6.conf.html">shorewall6.conf</ulink>(5).</para>
<para>If your kernel and ip6tables include CONNMARK support then
you can also mark the connection rather than the packet.</para>
<para>The mark value may be optionally followed by "/" and a
mask value (used to determine those bits of the connection mark
to actually be set). The mark and optional mask are then
followed by one of:+</para>
<variablelist>
<varlistentry>
<term><emphasis role="bold">C</emphasis></term>
<listitem>
<para>Mark the connection in the chain determined by the
setting of MARK_IN_FORWARD_CHAIN</para>
</listitem>
</varlistentry>
<varlistentry>
<term><emphasis role="bold">CF</emphasis></term>
<listitem>
<para>Mark the connection in the FORWARD chain</para>
</listitem>
</varlistentry>
<varlistentry>
<term><emphasis role="bold">CP</emphasis></term>
<listitem>
<para>Mark the connection in the PREROUTING chain.</para>
</listitem>
</varlistentry>
<varlistentry>
<term>CT</term>
<listitem>
<para>Mark the connecdtion in the POSTROUTING chain</para>
</listitem>
</varlistentry>
</variablelist>
<para><emphasis role="bold">Special considerations for If
HIGH_ROUTE_MARKS=Yes in <ulink
url="shorewall6.conf.html">shorewall6.conf</ulink>(5</emphasis>).</para>
<para>If HIGH_ROUTE_MARKS=Yes, then you may also specify a value
in the range 0x0100-0xFF00 with the low-order byte being zero.
Such values may only be used in the PREROUTING chain (value
followed by <emphasis role="bold">:P</emphasis> or you have set
MARK_IN_FORWARD_CHAIN=No in <ulink
url="shorewall6.conf.html">shorewall6.conf</ulink>(5) and have
not followed the value with <option>:F</option>) or the OUTPUT
chain (SOURCE is <emphasis role="bold">$FW</emphasis>). With
HIGH_ROUTE_MARKS=Yes, non-zero mark values less that 256 are not
permitted. Shorewall6 4.1 and later versions prohibit non-zero
mark values less that 256 in the OUTPUT chain when
HIGH_ROUTE_MARKS=Yes. While earlier versions allow such values
in the OUTPUT chain, it is strongly recommended that with
HIGH_ROUTE_MARKS=Yes, you use the POSTROUTING chain to apply
traffic shaping marks/classification.</para>
</listitem>
<listitem>
<para>A classification Id (classid) of the form
<emphasis>major</emphasis>:<emphasis>minor</emphasis> where
<emphasis>major</emphasis> and <emphasis>minor</emphasis> are
integers. Corresponds to the 'class' specification in these
traffic shaping modules:</para>
<programlisting> atm
cbq
dsmark
pfifo_fast
htb
prio</programlisting>
<para>Classification occurs in the POSTROUTING chain except when
the <emphasis role="bold">SOURCE</emphasis> is <emphasis
role="bold">$FW</emphasis>[:<emphasis>address</emphasis>] in
which case classification occurs in the OUTPUT chain.</para>
<para>When using Shorewall6's built-in traffic shaping tool, the
<emphasis>major</emphasis> class is the device number (the first
device in <ulink
url="shorewall6-tcdevices.html">shorewall6-tcdevices</ulink>(5)
is major class 1, the second device is major class 2, and so on)
and the <emphasis>minor</emphasis> class is the class's MARK
value in <ulink
url="shorewall6-tcclasses.html">shorewall6-tcclasses</ulink>(5)
preceded by the number 1 (MARK 1 corresponds to minor class 11,
MARK 5 corresponds to minor class 15, MARK 22 corresponds to
minor class 122, etc.).</para>
</listitem>
<listitem>
<para><emphasis
role="bold">RESTORE</emphasis>[/<emphasis>mask</emphasis>] --
restore the packet's mark from the connection's mark using the
supplied mask if any. Your kernel and ip6tables must include
CONNMARK support.</para>
<para>As in 1) above, may be followed by <emphasis
role="bold">:P</emphasis> or <emphasis
role="bold">:F</emphasis></para>
</listitem>
<listitem>
<para><emphasis
role="bold">SAVE</emphasis>[/<emphasis>mask</emphasis>] -- save
the packet's mark to the connection's mark using the supplied
mask if any. Your kernel and ip6tables must include CONNMARK
support.</para>
<para>As in 1) above, may be followed by <emphasis
role="bold">:P</emphasis> or <emphasis
role="bold">:F</emphasis></para>
</listitem>
<listitem>
<para><emphasis role="bold">CONTINUE</emphasis> Don't process
any more marking rules in the table.</para>
<para>As in 1) above, may be followed by <emphasis
role="bold">:P</emphasis> or <emphasis
role="bold">:F</emphasis>. Currently, CONTINUE may not be used
with <emphasis>exclusion</emphasis> (see the SOURCE and DEST
columns below); that restriction will be removed when
ip6tables/Netfilter provides the necessary support.</para>
</listitem>
<listitem>
<para><emphasis role="bold">COMMENT</emphasis> -- the rest of
the line will be attached as a comment to the Netfilter rule(s)
generated by the following entries. The comment will appear
delimited by "/* ... */" in the output of <command>shorewall6
show mangle</command></para>
<para>To stop the comment from being attached to further rules,
simply include COMMENT on a line by itself.</para>
</listitem>
</orderedlist>
</listitem>
</varlistentry>
<varlistentry>
<term><emphasis role="bold">SOURCE</emphasis> - {<emphasis
role="bold">-</emphasis>|{<emphasis>interface</emphasis>|<emphasis
role="bold">$FW</emphasis>}|[{<emphasis>interface</emphasis>|<emphasis
role="bold">$FW</emphasis>}:]&lt;<emphasis>address-or-range</emphasis>[<emphasis
role="bold">,</emphasis><emphasis>address-or-range</emphasis>]...}[<emphasis>exclusion</emphasis>]&gt;</term>
<listitem>
<para>Source of the packet. A comma-separated list of interface
names, IP addresses, MAC addresses and/or subnets for packets being
routed through a common path. List elements may also consist of an
interface name followed by ":" and an address (e.g.,
eth1:192.168.1.0/24). For example, all packets for connections
masqueraded to eth0 from other interfaces can be matched in a single
rule with several alternative SOURCE criteria. However, a connection
whose packets gets to eth0 in a different way, e.g., direct from the
firewall itself, needs a different rule.</para>
<para>Accordingly, use $<emphasis role="bold">FW</emphasis> in its
own separate rule for packets originating on the firewall. In such a
rule, the MARK column may NOT specify either <emphasis
role="bold">:P</emphasis> or <emphasis role="bold">:F</emphasis>
because marking for firewall-originated packets always occurs in the
OUTPUT chain.</para>
<para>MAC addresses must be prefixed with "~" and use "-" as a
separator.</para>
<para>Example: ~00-A0-C9-15-39-78</para>
<para>When an interface is not specified, the angled brackets
('&lt;' and '&gt;') surrounding the address(es) may be
omitted.</para>
<para>You may exclude certain hosts from the set already defined
through use of an <emphasis>exclusion</emphasis> (see <ulink
url="shorewall6-exclusion.html">shorewall6-exclusion</ulink>(5)).</para>
</listitem>
</varlistentry>
<varlistentry>
<term><emphasis role="bold">DEST</emphasis> - {<emphasis
role="bold">-</emphasis>|{<emphasis>interface</emphasis>|[<emphasis>interface</emphasis>:]&lt;<emphasis>address-or-range</emphasis>[<emphasis
role="bold">,</emphasis><emphasis>address-or-range</emphasis>]...}[<emphasis>exclusion</emphasis>]&gt;</term>
<listitem>
<para>Destination of the packet. Comma separated list of IP
addresses and/or subnets. If your kernel and ip6tables include
iprange match support, IP address ranges are also allowed. List
elements may also consist of an interface name followed by ":" and
an address (e.g., eth1:192.168.1.0/24). If the <emphasis
role="bold">MARK</emphasis> column specificies a classification of
the form <emphasis>major</emphasis>:<emphasis>minor</emphasis> then
this column may also contain an interface name.</para>
<para>When an interface is not specified, the angled brackets
('&lt;' and '&gt;') surrounding the address(es) may be
omitted.</para>
<para>You may exclude certain hosts from the set already defined
through use of an <emphasis>exclusion</emphasis> (see <ulink
url="shorewall6-exclusion.html">shorewall6-exclusion</ulink>(5)).</para>
</listitem>
</varlistentry>
<varlistentry>
<term><emphasis role="bold">PROTO</emphasis> - {<emphasis
role="bold">-</emphasis>|<emphasis
role="bold">tcp:syn</emphasis>|<emphasis
role="bold">ipp2p</emphasis>|<emphasis
role="bold">ipp2p:udp</emphasis>|<emphasis
role="bold">ipp2p:all</emphasis>|<emphasis>protocol-number</emphasis>|<emphasis>protocol-name</emphasis>|<emphasis
role="bold">all}</emphasis></term>
<listitem>
<para>Protocol - <emphasis role="bold">ipp2p</emphasis> requires
ipp2p match support in your kernel and ip6tables.</para>
</listitem>
</varlistentry>
<varlistentry>
<term><emphasis role="bold">PORT(S)</emphasis> (Optional) - [<emphasis
role="bold">-</emphasis>|<emphasis>port-name-number-or-range</emphasis>[<emphasis
role="bold">,</emphasis><emphasis>port-name-number-or-range</emphasis>]...]</term>
<listitem>
<para>Destination Ports. A comma-separated list of Port names (from
services(5)), <emphasis>port number</emphasis>s or <emphasis>port
range</emphasis>s; if the protocol is <emphasis
role="bold">icmp</emphasis>, this column is interpreted as the
destination icmp-type(s).</para>
<para>If the protocol is <emphasis role="bold">ipp2p</emphasis>,
this column is interpreted as an ipp2p option without the leading
"--" (example <emphasis role="bold">bit</emphasis> for bit-torrent).
If no PORT is given, <emphasis role="bold">ipp2p</emphasis> is
assumed.</para>
<para>This column is ignored if PROTOCOL = all but must be entered
if any of the following field is supplied. In that case, it is
suggested that this field contain "-"</para>
</listitem>
</varlistentry>
<varlistentry>
<term><emphasis role="bold">SOURCE PORT(S)</emphasis> (Optional) -
[<emphasis
role="bold">-</emphasis>|<emphasis>port-name-number-or-range</emphasis>[<emphasis
role="bold">,</emphasis><emphasis>port-name-number-or-range</emphasis>]...]</term>
<listitem>
<para>Source port(s). If omitted, any source port is acceptable.
Specified as a comma-separated list of port names, port numbers or
port ranges.</para>
</listitem>
</varlistentry>
<varlistentry>
<term><emphasis role="bold">USER</emphasis> (Optional) - [<emphasis
role="bold">!</emphasis>][<emphasis>user-name-or-number</emphasis>][<emphasis
role="bold">:</emphasis><emphasis>group-name-or-number</emphasis>]</term>
<listitem>
<para>This column may only be non-empty if the SOURCE is the
firewall itself.</para>
<para>When this column is non-empty, the rule applies only if the
program generating the output is running under the effective
<emphasis>user</emphasis> and/or <emphasis>group</emphasis>
specified (or is NOT running under that id if "!" is given).</para>
<para>Examples:</para>
<variablelist>
<varlistentry>
<term>joe</term>
<listitem>
<para>program must be run by joe</para>
</listitem>
</varlistentry>
<varlistentry>
<term>:kids</term>
<listitem>
<para>program must be run by a member of the 'kids'
group</para>
</listitem>
</varlistentry>
<varlistentry>
<term>!:kids</term>
<listitem>
<para>program must not be run by a member of the 'kids'
group</para>
</listitem>
</varlistentry>
</variablelist>
</listitem>
</varlistentry>
<varlistentry>
<term><emphasis role="bold">TEST</emphasis> - [<emphasis
role="bold">!</emphasis>]<emphasis>value</emphasis>[/<emphasis>mask</emphasis>][<emphasis
role="bold">:C</emphasis>]</term>
<listitem>
<para>Defines a test on the existing packet or connection mark. The
rule will match only if the test returns true.</para>
<para>If you don't want to define a test but need to specify
anything in the following columns, place a "-" in this field.</para>
<variablelist>
<varlistentry>
<term>!</term>
<listitem>
<para>Inverts the test (not equal)</para>
</listitem>
</varlistentry>
<varlistentry>
<term><emphasis>value</emphasis></term>
<listitem>
<para>Value of the packet or connection mark.</para>
</listitem>
</varlistentry>
<varlistentry>
<term><emphasis>mask</emphasis></term>
<listitem>
<para>A mask to be applied to the mark before testing.</para>
</listitem>
</varlistentry>
<varlistentry>
<term><emphasis role="bold">:C</emphasis></term>
<listitem>
<para>Designates a connection mark. If omitted, the packet
mark's value is tested.</para>
</listitem>
</varlistentry>
</variablelist>
</listitem>
</varlistentry>
<varlistentry>
<term><emphasis role="bold">LENGTH</emphasis> (Optional) -
[<emphasis>length</emphasis>|[<emphasis>min</emphasis>]<emphasis
role="bold">:</emphasis>[<emphasis>max</emphasis>]]</term>
<listitem>
<para>Packet Length. This field, if present allow you to match the
length of a packet against a specific value or range of values. You
must have ip6tables length support for this to work. A range is
specified in the form
<emphasis>min</emphasis>:<emphasis>max</emphasis> where either
<emphasis>min</emphasis> or <emphasis>max</emphasis> (but not both)
may be omitted. If <emphasis>min</emphasis> is omitted, then 0 is
assumed; if <emphasis>max</emphasis> is omitted, than any packet
that is <emphasis>min</emphasis> or longer will match.</para>
</listitem>
</varlistentry>
<varlistentry>
<term><emphasis role="bold">TOS</emphasis> -
<emphasis>tos</emphasis></term>
<listitem>
<para>Type of service. Either a standard name, or a numeric value to
match.</para>
<programlisting> <emphasis role="bold">Minimize-Delay</emphasis> (16)
<emphasis role="bold">Maximize-Throughput</emphasis> (8)
<emphasis role="bold">Maximize-Reliability</emphasis> (4)
<emphasis role="bold">Minimize-Cost</emphasis> (2)
<emphasis role="bold">Normal-Service</emphasis> (0)</programlisting>
</listitem>
</varlistentry>
<varlistentry>
<term><emphasis role="bold">CONNBYTES</emphasis> -
[!]<emphasis>min</emphasis>:[<emphasis>max</emphasis>[:{<emphasis
role="bold">O</emphasis>|<emphasis role="bold">R</emphasis>|<emphasis
role="bold">B</emphasis>}[:{<emphasis
role="bold">B</emphasis>|<emphasis role="bold">P</emphasis>|<emphasis
role="bold">A</emphasis>}]]]</term>
<listitem>
<para>Connection Bytes; defines a byte or packet range that the
connection must fall within in order for the rule to match.</para>
<para>A packet matches if the the packet/byte count is within the
range defined by <emphasis>min</emphasis> and
<emphasis>max</emphasis> (unless ! is given in which case, a packet
matches if the packet/byte count is not within the range).
<emphasis>min</emphasis> is an integer which defines the beginning
of the byte/packet range. <emphasis>max</emphasis> is an integer
which defines the end of the byte/packet range; if omitted, only the
beginning of the range is checked. The first letter gives the
direction which the range refers to:<blockquote>
<para><emphasis role="bold">O</emphasis> - The original
direction of the connection.</para>
<para><emphasis role="bold">R</emphasis> - The opposite
direction from the original connection.</para>
<para><emphasis role="bold">B</emphasis> - The total of both
directions.</para>
</blockquote></para>
<para>If omitted, <emphasis role="bold">B</emphasis> is
assumed.</para>
<para>The second letter determines what the range refers
to.<blockquote>
<para><emphasis role="bold">B</emphasis> - Bytes</para>
<para><emphasis role="bold">P</emphasis> - Packets</para>
<para><emphasis role="bold">A</emphasis> - Average packet
size.</para>
</blockquote>If omitted, <emphasis role="bold">B</emphasis> is
assumed.</para>
</listitem>
</varlistentry>
<varlistentry>
<term><emphasis role="bold">HELPER -
</emphasis><emphasis>helper</emphasis></term>
<listitem>
<para>Names a Netfiler protocol <firstterm>helper</firstterm> module
such as <option>ftp</option>, <option>sip</option>,
<option>amanda</option>, etc. A packet will match if it was accepted
by the named helper module. You can also append "-" and a port
number to the helper module name (e.g., <emphasis
role="bold">ftp-21</emphasis>) to specify the port number that the
original connection was made on.</para>
<para>Example: Mark all FTP data connections with mark
4:<programlisting>#MARK/ SOURCE DEST PROTO PORT(S) SOURCE USER TEST LENGTH TOS CONNBYTES HELPER
#CLASSIFY PORT(S)
4 ::/ ::/ TCP - - - - - - - ftp</programlisting></para>
</listitem>
</varlistentry>
</variablelist>
</refsect1>
<refsect1>
<title>Example</title>
<variablelist>
<varlistentry>
<term>Example 1:</term>
<listitem>
<para>Mark all forwarded ICMP echo traffic with packet mark 1. Mark
all forwarded peer to peer traffic with packet mark 4.</para>
<para>This is a little more complex than otherwise expected. Since
the ipp2p module is unable to determine all packets in a connection
are P2P packets, we mark the entire connection as P2P if any of the
packets are determined to match.</para>
<para>We assume packet/connection mark 0 means unclassified.</para>
<programlisting> #MARK/ SOURCE DEST PROTO PORT(S) SOURCE USER TEST
#CLASSIFY PORT(S)
1 ::/ ::/ icmp echo-request
1 ::/ ::/ icmp echo-reply
RESTORE ::/ ::/ all - - - 0
CONTINUE ::/ ::/ all - - - !0
4 ::/ ::/ ipp2p:all
SAVE ::/ ::/ all - - - !0</programlisting>
<para>If a packet hasn't been classifed (packet mark is 0), copy the
connection mark to the packet mark. If the packet mark is set, we're
done. If the packet is P2P, set the packet mark to 4. If the packet
mark has been set, save it to the connection mark.</para>
</listitem>
</varlistentry>
</variablelist>
</refsect1>
<refsect1>
<title>FILES</title>
<para>/etc/shorewall6/tcrules</para>
</refsect1>
<refsect1>
<title>See ALSO</title>
<para><ulink
url="http://shorewall.net/traffic_shaping.htm">http://shorewall.net/traffic_shaping.htm</ulink></para>
<para><ulink
url="http://shorewall.net/MultiISP.html">http://shorewall.net/MultiISP.html</ulink></para>
<para><ulink
url="http://shorewall.net/PacketMarking.html">http://shorewall.net/PacketMarking.html</ulink></para>
<para>shorewall6(8), shorewall6-accounting(5), shorewall6-actions(5),
shorewall6-blacklist(5), shorewall6-ecn(5), shorewall6-exclusion(5),
shorewall6-hosts(5), shorewall6-interfaces(5), shorewall6-maclist(5),
shorewall6-params(5), shorewall6-policy(5), shorewall6-providers(5),
shorewall6-route_rules(5), shorewall6-routestopped(5),
shorewall6-rules(5), shorewall6.conf(5), shorewall6-tcclasses(5),
shorewall6-tcdevices(5), shorewall6-tos(5), shorewall6-tunnels(5),
shorewall6-zones(5)</para>
</refsect1>
</refentry>