man.bsd.lv manual page server

NAME

ipfw3 — IP firewall and traffic shaper control program

SYNOPSIS

DESCRIPTION

The ipfw3 utility is the user interface for controlling the ipfw(4) firewall and the dummynet(4) traffic shaper in DragonFly.

An ipfw3 configuration, or ruleset, is made of a list of rules numbered from 1 to 65535. Packets are passed to ipfw3 from a number of different places in the protocol stack (depending on the source and destination of the packet, it is possible that ipfw3 is invoked multiple times on the same packet). The packet passed to the firewall is compared against each of the rules in the firewall ruleset. When a match is found, the action corresponding to the matching rule is performed.

Depending on the action and certain system settings, packets can be reinjected into the firewall at some rule after the matching one for further processing.

An ipfw3 ruleset always includes a default rule (numbered 65535) which cannot be modified, and matches all packets. The action associated with the default rule can be either deny or allow depending on how the kernel is configured.

If the ruleset includes one or more rules with the keep-state or limit option, then ipfw3 assumes a stateful behaviour, i.e. upon a match it will create dynamic rules matching the exact parameters (addresses and ports) of the matching packet.

These dynamic rules, which have a limited lifetime, are checked at the first occurrence of a check-state, keep-state or limit rule, and are typically used to open the firewall on-demand to legitimate traffic only. See the STATEFUL FIREWALL and EXAMPLES Sections below for more information on the stateful behaviour of ipfw3.

All rules (including dynamic ones) have a few associated counters: a packet count, a byte count, a log count and a timestamp indicating the time of the last match. Counters can be displayed or reset with ipfw3 commands.

Rules can be added with the add command; deleted individually or in groups with the delete command, and globally with the flush command; displayed, optionally with the content of the counters, using the show and list commands. Finally, counters can be reset with the zero and resetlog commands.

Also, each rule belongs to one of 32 different sets , and there are ipfw3 commands to atomically manipulate sets, such as enable, disable, swap sets, move all rules in a set to another one, delete all rules in a set. These can be useful to install temporary configurations, or to test them. See Section SETS OF RULES for more information on sets.

The following options are available:

-a

While listing, show counter values. The show command just implies this option.

-c

When entering or showing rules, print them in compact form, i.e. without the optional "ip " string when this does not carry any additional information.

-d

While listing, show states in addition to static ones.

-D

While listing, show states only without static ones.

-e

While listing, if the -d option was specified, also show expired dynamic rules.

-f

Don't ask for confirmation for commands that can cause problems if misused, i.e. flush. If there is no tty associated with the process, this is implied.

-N

Try to resolve addresses and service names in output.

-S

While listing rules, show the set each rule belongs to. If this flag is not specified, disabled rules will not be listed.

-s [field]

While listing pipes, sort according to one of the four counters (total or current packets or bytes).

-t

While listing, show last match timestamp.

-T

While listing, show last match timestamp in unix format.

-v

With verbose information, it will like the ipfw in FreeBSD.

To ease configuration, rules can be put into a file which is processed using ipfw3 as shown in the last synopsis line. An absolute pathname must be used. The file will be read line by line and applied as arguments to the ipfw3 utility.

Optionally, a preprocessor can be specified using -p preproc where pathname is to be piped through. Useful preprocessors include cpp(1) and m4(1). If preproc doesn't start with a slash (‘/’) as its first character, the usual PATH name search is performed. Care should be taken with this in environments where not all file systems are mounted (yet) by the time ipfw3 is being run (e.g. when they are mounted over NFS). Once -p has been specified, optional -D and -U specifications can follow and will be passed on to the preprocessor. This allows for flexible configuration files (like conditionalizing them on the local hostname) and the use of macros to centralize frequently required arguments like IP addresses.

The ipfw3 pipe and queue commands are used to configure the traffic shaper, as shown in the TRAFFIC SHAPER (DUMMYNET) CONFIGURATION Section below.

If the world and the kernel get out of sync the ipfw3 ABI may break, preventing you from being able to add any rules. This can adversely affect the booting process. You can use ipfw3 disable firewall to temporarily disable the firewall to regain access to the network, allowing you to fix the problem.

PACKET FLOW

A packet is checked against the active ruleset in multiple places in the protocol stack, under control of several sysctl variables. These places and variables are shown below, and it is important to have this picture in mind in order to design a correct ruleset.

         ^     to upper layers     V
         |                         |
         +------------>------------+
         ^                         V
    [ip_input]                [ip_output]   net.inet.ip.fw.enable=1
         |                         |
         ^                         V
[ether_demux_oncpu]   [ether_output_frame]  net.link.ether.ipfw=1
         ^                         V
         |       to devices        |

As can be noted from the above picture, the number of times the same packet goes through the firewall can vary between 0 and 4 depending on packet source and destination, and system configuration.

Note that as packets flow through the stack, headers can be stripped or added to it, and so they may or may not be available for inspection. E.g., incoming packets will include the MAC header when ipfw3 is invoked from ether_demux_oncpu(), but the same packets will have the MAC header stripped off when ipfw3 is invoked from ip_input().

Also note that each packet is always checked against the complete ruleset, irrespective of the place where the check occurs, or the source of the packet. If a rule contains some match patterns or actions which are not valid for the place of invocation (e.g. trying to match a MAC header within ip_input()), the match pattern will not match, but a not operator in front of such patterns will cause the pattern to always match on those packets. It is thus the responsibility of the programmer, if necessary, to write a suitable ruleset to differentiate among the possible places. skipto rules can be useful here, as an example:

# packets from ether_demux_oncpu
ipfw add 10 skipto 1000 all layer2 in
# packets from ip_input
ipfw add 10 skipto 2000 all not layer2 in
# packets from ip_output
ipfw add 10 skipto 3000 all not layer2 out
# packets from ether_output_frame
ipfw add 10 skipto 4000 all layer2 out

RULE FORMAT

The format of ipfw3 rules is the following:

where the body of the rule specifies which information is used for filtering packets, among the following:

Note that some of the above information, e.g. source MAC or IP addresses and TCP/UDP ports, could easily be spoofed, so filtering on those fields alone might not guarantee the desired results.

rule_number

Each rule is associated with a rule_number in the range 1..65535, with the latter reserved for the default rule. Rules are checked sequentially by rule number. Multiple rules can have the same number, in which case they are checked (and listed) according to the order in which they have been added. If a rule is entered without specifying a number, the kernel will assign one in such a way that the rule becomes the last one before the default rule. Automatic rule numbers are assigned by incrementing the last non-default rule number by the value of the sysctl variable net.inet.ip.fw.autoinc_step which defaults to 100. If this is not possible (e.g. because we would go beyond the maximum allowed rule number), the number of the last non-default value is used instead.

set set_number

Each rule is associated with a set_number in the range 0..31, with the latter reserved for the default rule. Sets can be individually disabled and enabled, so this parameter is of fundamental importance for atomic ruleset manipulation. It can be also used to simplify deletion of groups of rules. If a rule is entered without specifying a set number, set 0 will be used.

prob match_probability

A match is only declared with the specified probability (floating point number between 0 and 100). This can be useful for a number of applications such as random packet drop or (in conjunction with dummynet(4)) to simulate the effect of multiple paths leading to out-of-order packet delivery.

log [logamount number]

When a packet matches a rule with the log keyword, a message will be logged to syslogd(8) with a LOG_SECURITY facility. The logging only occurs if the sysctl variable net.inet.ip.fw.verbose is set to 1 (which is the default when the kernel is compiled with IPFIREWALL_VERBOSE) and the number of packets logged so far for that particular rule does not exceed the logamount parameter. If no logamount is specified, the limit is taken from the sysctl variable net.inet.ip.fw.verbose_limit. In both cases, a value of 0 removes the logging limit.

Once the limit is reached, logging can be re-enabled by clearing the logging counter or the packet counter for that entry, see the resetlog command.

SETS OF RULES

Each rule belongs to one of 32 different sets , numbered 0 to 31. Set 31 is reserved for the default rule.

By default, rules are put in set 0, unless you use the set N attribute when entering a new rule. Sets can be individually and atomically enabled or disabled, so this mechanism permits an easy way to store multiple configurations of the firewall and quickly (and atomically) switch between them. The command to enable/disable sets is

where multiple enable or disable sections can be specified. Command execution is atomic on all the sets specified in the command. By default, all sets are enabled.

When you disable a set, its rules behave as if they do not exist in the firewall configuration, with only one exception:

The set number of rules can be changed with the command

Also, you can atomically swap two rulesets with the command

See the EXAMPLES Section on some possible uses of sets of rules.

STATEFUL FIREWALL

Stateful operation is a way for the firewall to dynamically create rules for specific flows when packets that match a given pattern are detected. Support for stateful operation comes through the check-state, keep-state and limit options of ipfw3 rules.

Dynamic rules are created when a packet matches a keep-state or limit rule, causing the creation of a dynamic rule which will match all and only packets with a given protocol between a src-ip/src-port dst-ip/dst-port pair of addresses ( src and dst are used here only to denote the initial match addresses, but they are completely equivalent afterwards). Dynamic rules will be checked at the first check-state, keep-state or limit occurrence, and the action performed upon a match will be the same as in the parent rule.

Note that no additional attributes other than protocol and IP addresses and ports are checked on dynamic rules.

The typical use of dynamic rules is to keep a closed firewall configuration, but let the first TCP SYN packet from the inside network install a dynamic rule for the flow so that packets belonging to that session will be allowed through the firewall:

A similar approach can be used for UDP, where an UDP packet coming from the inside will install a dynamic rule to let the response through the firewall:

Dynamic rules expire after some time, which depends on the status of the flow and the setting of some sysctl variables. See Section SYSCTL VARIABLES for more details. For TCP sessions, dynamic rules can be instructed to periodically send keepalive packets to refresh the state of the rule when it is about to expire.

See Section EXAMPLES for more examples on how to use dynamic rules.

TRAFFIC SHAPER (DUMMYNET) CONFIGURATION

ipfw3 is also the user interface for the dummynet(4) traffic shaper.

dummynet(4) operates by first using the firewall to classify packets and divide them into flows, using any match pattern that can be used in ipfw3 rules. Depending on local policies, a flow can contain packets for a single TCP connection, or from/to a given host, or entire subnet, or a protocol type, etc.

Packets belonging to the same flow are then passed to either of two different objects, which implement the traffic regulation:

pipe

A pipe emulates a link with given bandwidth, propagation delay, queue size and packet loss rate. Packets are queued in front of the pipe as they come out from the classifier, and then transferred to the pipe according to the pipe's parameters.

queue

A queue is an abstraction used to implement the WF2Q+ (Worst-case Fair Weighted Fair Queueing) policy, which is an efficient variant of the WFQ policy.
The queue associates a weight and a reference pipe to each flow, and then all backlogged (i.e., with packets queued) flows linked to the same pipe share the pipe's bandwidth proportionally to their weights. Note that weights are not priorities; a flow with a lower weight is still guaranteed to get its fraction of the bandwidth even if a flow with a higher weight is permanently backlogged.

The pipe and queue configuration commands are the following:

The following parameters can be configured for a pipe:

bw bandwidth

Bandwidth, measured in [K|M]{bit/s|Byte/s}.

A value of 0 (default) means unlimited bandwidth. The unit must immediately follow the number, as in

ipfw pipe 1 config bw 300Kbit/s

delay ms-delay

Propagation delay, measured in milliseconds. The value is rounded to the next multiple of the clock tick (typically 10ms, but it is a good practice to run kernels with options HZ=1000 to reduce the granularity to 1ms or less). Default value is 0, meaning no delay.

The following parameters can be configured for a queue:

pipe pipe_nr

Connects a queue to the specified pipe. Multiple queues (with the same or different weights) can be connected to the same pipe, which specifies the aggregate rate for the set of queues.

weight weight

Specifies the weight to be used for flows matching this queue. The weight must be in the range 1..100, and defaults to 1.

Finally, the following parameters can be configured for both pipes and queues:

buckets hash-table-size

Specifies the size of the hash table used for storing the various queues. Default value is 64 controlled by the sysctl(8) variable net.inet.ip.dummynet.hash_size, allowed range is 16 to 65536.

mask mask-specifier

Packets sent to a given pipe or queue by an ipfw3 rule can be further classified into multiple flows, each of which is then sent to a different dynamic pipe or queue. A flow identifier is constructed by masking the IP addresses, ports and protocol types as specified with the mask options in the configuration of the pipe or queue. For each different flow identifier, a new pipe or queue is created with the same parameters as the original object, and matching packets are sent to it.

Thus, when dynamic pipes are used, each flow will get the same bandwidth as defined by the pipe, whereas when dynamic queues are used, each flow will share the parent's pipe bandwidth evenly with other flows generated by the same queue (note that other queues with different weights might be connected to the same pipe).
Available mask specifiers are a combination of one or more of the following:

dst-ip mask, src-ip mask, dst-port mask, src-port mask, proto mask or all,

where the latter means all bits in all fields are significant.

noerror

When a packet is dropped by a dummynet queue or pipe, the error is normally reported to the caller routine in the kernel, in the same way as it happens when a device queue fills up. Setting this option reports the packet as successfully delivered, which can be needed for some experimental setups where you want to simulate loss or congestion at a remote router.

NOTE: This option is always on, since DragonFly 1.11.

plr packet-loss-rate

Packet loss rate. Argument packet-loss-rate is a floating-point number between 0 and 1, with 0 meaning no loss, 1 meaning 100% loss. The loss rate is internally represented on 31 bits.

queue {slots | sizeKbytes}

Queue size, in slots or KBytes. Default value is 50 slots, which is the typical queue size for Ethernet devices. Note that for slow speed links you should keep the queue size short or your traffic might be affected by a significant queueing delay. E.g., 50 max-sized ethernet packets (1500 bytes) mean 600Kbit or 20s of queue on a 30Kbit/s pipe. Even worse effect can result if you get packets from an interface with a much larger MTU, e.g. the loopback interface with its 16KB packets.

red | gred w_q/min_th/max_th/max_p

Make use of the RED (Random Early Detection) queue management algorithm. w_q and max_p are floating point numbers between 0 and 1 (0 not included), while min_th and max_th are integer numbers specifying thresholds for queue management (thresholds are computed in bytes if the queue has been defined in bytes, in slots otherwise). The dummynet(4) also supports the gentle RED variant (gred). Three sysctl(8) variables can be used to control the RED behaviour:

net.inet.ip.dummynet.red_lookup_depth

specifies the accuracy in computing the average queue when the link is idle (defaults to 256, must be greater than zero)

net.inet.ip.dummynet.red_avg_pkt_size

specifies the expected average packet size (defaults to 512, must be greater than zero)

net.inet.ip.dummynet.red_max_pkt_size

specifies the expected maximum packet size, only used when queue thresholds are in bytes (defaults to 1500, must be greater than zero).

CHECKLIST

Here are some important points to consider when designing your rules:

• Remember that you filter both packets going in and out. Most connections need packets going in both directions.
• Remember to test very carefully. It is a good idea to be near the console when doing this. If you cannot be near the console, use an auto-recovery script such as the one in /usr/share/examples/ipfw/change_rules.sh.
• Don't forget the loopback interface.

FINE POINTS

• There are circumstances where fragmented datagrams are unconditionally dropped. TCP packets are dropped if they do not contain at least 20 bytes of TCP header, UDP packets are dropped if they do not contain a full 8 byte UDP header, and ICMP packets are dropped if they do not contain 4 bytes of ICMP header, enough to specify the ICMP type, code, and checksum. These packets are simply logged as “pullup failed” since there may not be enough good data in the packet to produce a meaningful log entry.
• Another type of packet is unconditionally dropped, a TCP packet with a fragment offset of one. This is a valid packet, but it only has one use, to try to circumvent firewalls. When logging is enabled, these packets are reported as being dropped by rule -1.
• If you are logged in over a network, loading the kld(4) version of ipfw3 is probably not as straightforward as you would think. I recommend the following command line:

  kldload /boot/modules/ipfw3.ko && \
  ipfw add 32000 allow ip

  Along the same lines, doing an

  ipfw flush

  in similar surroundings is also a bad idea.

• The ipfw3 filter list may not be modified if the system security level is set to 3 or higher (see init(8) for information on system security levels).

PACKET DIVERSION

A divert(4) socket bound to the specified port will receive all packets diverted to that port. If no socket is bound to the destination port, or if the kernel wasn't compiled with divert socket support, the packets are dropped.

NETWORK ADDRESS TRANSLATION (NAT)

ipfw3 comes with an ipfw3_nat module which supports the in-kernel NAT. This kernel module should be loaded before you start the configuration.

The nat configuration command is the following:

More than one address can be configured, and the packet will be randomly translated to these addresses.

All the translation records (NAT states) will be displayed in the format below.

In DragonFly, the sample packets will be always handled by the same CPU. Due to the nature of the network address translation (NAT), the outgoing and returning packets will be handled by another CPU. To ensure the returned traffic can be translated back to the original address, ipfw3 will automatically prepare the translation records into the correct CPU.

SYSCTL VARIABLES

A set of sysctl(8) variables controls the behaviour of the firewall and associated modules (dummynet(4)). These are shown below together with their default value (but always check with the sysctl(8) command what value is actually in use) and meaning:

net.filters_default_to_accept: 0

If set prior to loading the ipfw3 kernel module, the filter will default to allowing all packets through. If not set the filter will likely default to not allowing any packets through.

net.inet.ip.dummynet.expire: 1

Lazily delete dynamic pipes/queue once they have no pending traffic. You can disable this by setting the variable to 0, in which case the pipes/queues will only be deleted when the threshold is reached.

net.inet.ip.dummynet.hash_size: 64

Default size of the hash table used for dynamic pipes/queues. This value is used when no buckets option is specified when configuring a pipe/queue.

net.inet.ip.dummynet.max_chain_len: 16

Target value for the maximum number of pipes/queues in a hash bucket. The product max_chain_len*hash_size is used to determine the threshold over which empty pipes/queues will be expired even when net.inet.ip.dummynet.expire=0.

net.inet.ip.dummynet.red_lookup_depth: 256

 

net.inet.ip.dummynet.red_avg_pkt_size: 512

 

net.inet.ip.dummynet.red_max_pkt_size: 1500

Parameters used in the computations of the drop probability for the RED algorithm.

net.inet.ip.fw.autoinc_step: 100

Delta between rule numbers when auto-generating them. The value must be in the range 1..1000.

net.inet.ip.fw.curr_dyn_buckets: net.inet.ip.fw.dyn_buckets

The current number of buckets in the hash table for dynamic rules (readonly).

net.inet.ip.fw.debug: 1

Controls debugging messages produced by ipfw3.

net.inet.ip.fw.dyn_buckets: 256

The number of buckets in the hash table for dynamic rules. Must be a power of 2, up to 65536. It only takes effect when all dynamic rules have expired, so you are advised to use a flush command to make sure that the hash table is resized.

net.inet.ip.fw.dyn_count: 3

Current number of dynamic rules (read-only).

net.inet.ip.fw.dyn_keepalive: 1

Enables generation of keepalive packets for keep-state rules on TCP sessions. A keepalive is generated to both sides of the connection every 5 seconds for the last 20 seconds of the lifetime of the rule.

net.inet.ip.fw.dyn_max: 8192

Maximum number of dynamic rules. When you hit this limit, no more dynamic rules can be installed until old ones expire.

net.inet.ip.fw.dyn_ack_lifetime: 300

 

net.inet.ip.fw.dyn_syn_lifetime: 20

 

net.inet.ip.fw.dyn_fin_lifetime: 1

 

net.inet.ip.fw.dyn_rst_lifetime: 1

 

net.inet.ip.fw.dyn_udp_lifetime: 5

 

net.inet.ip.fw.dyn_short_lifetime: 30

These variables control the lifetime, in seconds, of dynamic rules. Upon the initial SYN exchange the lifetime is kept short, then increased after both SYN have been seen, then decreased again during the final FIN exchange or when a RST is received. Both dyn_fin_lifetime and dyn_rst_lifetime must be strictly lower than 5 seconds, the period of repetition of keepalives. The firewall enforces that.

net.inet.ip.fw.enable: 1

Enables the firewall. Setting this variable to 0 lets you run your machine without firewall even if compiled in.

net.inet.ip.fw.one_pass: 1

When set, the packet exiting from the dummynet(4) pipe is not passed though the firewall again. Otherwise, after a pipe action, the packet is reinjected into the firewall at the next rule.

Note: layer 2 packets coming out of a pipe are never reinjected in the firewall irrespective of the value of this variable.

net.inet.ip.fw.verbose: 1

Enables verbose messages.

net.inet.ip.fw.verbose_limit: 0

Limits the number of messages produced by a verbose firewall.

net.link.ether.ipfw: 0

Controls whether layer-2 packets are passed to ipfw3. Default is no.

IPFW3 ENHANCEMENTS

This Section lists the features that have been introduced in ipfw3 of DragonFly which were not present in ipfw of FreeBSD. We list them in order of the potential impact that they can have in writing your rulesets. You might want to consider using these features in order to write your rulesets in a more efficient way.

Modular Design

 

Lockless

 

Stateful

In ipfw of DragonFly, the state links to the rule which created it. all packets will be filtered during the action ‘check-state’. And states can be manipulated by using the ipfw utility.

ipfw state add rule 1000 udp 192.168.1.100:0 8.8.8.8:53 expiry 600

EXAMPLES

There are far too many possible uses of ipfw3 so this Section will only give a small set of examples.

SEE ALSO

cpp(1), m4(1), divert(4), dummynet(4), ip(4), ipfirewall(4), protocols(5), services(5), init(8), kldload(8), reboot(8), sysctl(8), syslogd(8)

HISTORY

The ipfw3 utility first appeared in FreeBSD 2.0. dummynet(4) was introduced in FreeBSD 2.2.8. Stateful extensions were introduced in FreeBSD 4.0. ipfw3 was introduced in Summer 2002.

AUTHORS

Ugen J. S. Antsilevich,
Poul-Henning Kamp,
Alex Nash,
Archie Cobbs,
Luigi Rizzo,
Bill Yuan.

API based upon code written by Daniel Boulet for BSDI.

Work on dummynet(4) traffic shaper supported by Akamba Corp.

BUGS

The syntax has grown over the years and sometimes it might be confusing. Unfortunately, backward compatibility prevents cleaning up mistakes made in the definition of the syntax.

!!! WARNING !!!

Misconfiguring the firewall can put your computer in an unusable state, possibly shutting down network services and requiring console access to regain control of it.

Incoming packet fragments diverted by divert or tee are reassembled before delivery to the socket. The action used on those packet is the one from the rule which matches the first fragment of the packet.

Packets that match a tee rule should not be immediately accepted, but should continue going through the rule list. This may be fixed in a later version.

Packets diverted to userland, and then reinserted by a userland process (such as natd(8)) will lose various packet attributes, including their source interface. If a packet is reinserted in this manner, later rules may be incorrectly applied, making the order of divert rules in the rule sequence very important.