Child pages
  • 4.3 Firewall
Skip to end of metadata
Go to start of metadata

4.3.1 Introduction

Firewalls are used to filter network traffic based on rules set by the system administrator. Firewall can protect your personal computer or whole company's network from unauthorized network while allowing passage of legitimate network traffic.

4.3.2 IP Filter

OpenIndiana comes with built-in firewall, IP Filter. IP Filter is the stateful packet filtering and network address translation (NAT) mechanism. IP filter can filter any kind of traffic based on source or destination IP address or pool of IP addresses, source or destination ports, interface or direction of the network traffic. OpenIndiana IP Filter is derived from open source IPFilter software.


To manage IP Filter rules one must assume a role that includes IP Filter Management profile rights or superuser (root).

4.3.3 IP Filter configuration

IP Filter is configured by loadable configurations files stored in /etc/ipf. One can create several configuration files in /etc/ipf for firewall configuration:

  • ipf.conf - stores packet filtering rules
  • ipnat.conf - defines NAT rules
  • ippol.conf - address pool configuration

If IP Filter SMF service is enabled then the configured rules will be automatically loaded at every boot time of the operating system. Packet Filtering

IP Filter ruleset can be configured with the ipf(1M) or /etc/ipf/ipf.conf file. Rules are processed by the "the last matching rule wins" logic. This means that packet passing the IP Filter ruleset from the beggining and the action of the last rule that matched the packet is applied. There are two exceptions, which change this processing. The first one is the use of quick keyword, which will apply the rule on the packet and stop further filter rules checks. Another exception is the group keyword, which matches packet. Only rules with group keyword are used for packet processing. Filtering rules syntax

The following format is used to create filtering rules:

action [in|out] option keyword, keyword..

Every rule begins with the action. Action can be one of these:

  • block - denies packets from passing the filter
  • pass - allows packets to pass the filter
  • log - logs the packet. ipmon(1M) is used to view the log file.
  • count - counts packet into the filter statistics. Use ipfstat(1M) to display the statistics.
  • skip number - skips the filter over number filtering rules
  • auth - user program is requested to perform packet authentification in order to decide if the packet should be passed or not

Following the action, the next word is in or out. This determines in which direction rules are applied, e.g incoming or outgoing packets.

The option keyword is next. One can choose from:

  • log - logs the packet is the packet matched the rule. Use ipmon(1M) to view the log.
  • quick - rule with quick keyword is executed if packet matches it. No further rules checking is done.
  • on interface - rule is applied only on interface in both directions
  • dup-to interface - packet is copied and sent out on interface to specified IP address
  • to interface - packet is moved to an outbound queue on interface

Next are the keywords that determine if the packet matches the rule. The following keywords shown here can be used:

  • tos - packet is filtered based on the type-of-service value written as decimal or hexadecimal integer.
  • ttl - packet is matched based on its time-to-live value.
  • proto - used to match a specific protocol. Any protocol name from /etc/protocols or its decimal representation can be used.
  • from/to/all/any - matches either source or destination IP address of the packet and the port number. All accepts packet from any source to any destination.
  • with - matches specified attributes associated with the packet. Inserted not/no in front of the keyword matches the packet only if the option is not present.
  • flags - filters based on TCP flags that are set.
  • icmp-type - filters based on ICMP type.
  • keep keep-options - determines whetever state should be kept for a packet. state stores information about the session and can be kept on TCP, UDP, and ICMP packets. The frags keeps information about packet fragments and apllies them to later fragments. This option allows matching packets to pass without further ruleset evalution.
  • head number - creates new group for filtering rules denoted by number.
  • group number - adds the rule to group number. The default group value is 0.

In the following example we will block all incoming packet on igb0 from This rule should be included in one's ruleset:

block in quick on igb0 from to any Network Address Translation (NAT)

NAT is used in case when one needs to do address or port translation. This happens when one wants to connect multiple computers at home and share the network connection or when one wants to do port forwarding. NAT on OpenIndiana is set up in /etc/ipf/ipnat.conf and work regarding NATs is done with ipnat(1M). NAT rules syntax

To create NAT rules use the following syntax:

command interface-name parameters

Every rule begins with command from one of these:

  • map - maps one IP address or network to another IP address or network.
  • rdr - redirects packet from one IP address and port to another IP address and port.
  • bimap - creates bidirectorial NAT between an external and an internal IP address.
  • map-block - establishes static IP address-based translation.

Interface named is used after command, e.g. igb0.

To determine NAT configuration one has to use one of the following parameters:

  • ipmask - designates the network mask.
  • dstipmask - designates the address ipmask is translated to.
  • mapport - designates TCP or UDP protocols along with range of ports.

Assuming we have an external IP address on interface eg1000 and an internal range of The example NAT rule would look like this:

map eg1000 ->


NAT is not usable with IPv6 IP filter as NAT is deprecated in IPv6. NAT can be only used with IPv4 addresses. Address pools

Address pools group multiple IP addresses/networks into a single reference that can be used in IP Filter rules.