{ "mode": "man", "parameter": "iptables-extensions", "section": "8", "url": "https://www.chedong.com/phpMan.php/man/iptables-extensions/8/json", "generated": "2026-06-12T16:23:42Z", "synopsis": "ip6tables [-m name [module-options...]] [-j target-name [target-options...]\niptables [-m name [module-options...]] [-j target-name [target-options...]", "sections": { "NAME": { "content": "iptables-extensions — list of extensions in the standard iptables distribution\n", "subsections": [] }, "SYNOPSIS": { "content": "ip6tables [-m name [module-options...]] [-j target-name [target-options...]\n\niptables [-m name [module-options...]] [-j target-name [target-options...]\n", "subsections": [] }, "MATCH EXTENSIONS": { "content": "iptables can use extended packet matching modules with the -m or --match options, followed by\nthe matching module name; after these, various extra command line options become available,\ndepending on the specific module. You can specify multiple extended match modules in one\nline, and you can use the -h or --help options after the module has been specified to receive\nhelp specific to that module. The extended match modules are evaluated in the order they are\nspecified in the rule.\n\nIf the -p or --protocol was specified and if and only if an unknown option is encountered,\niptables will try load a match module of the same name as the protocol, to try making the op‐\ntion available.\n", "subsections": [ { "name": "addrtype", "content": "This module matches packets based on their address type. Address types are used within the\nkernel networking stack and categorize addresses into various groups. The exact definition\nof that group depends on the specific layer three protocol.\n\nThe following address types are possible:\n\nUNSPEC an unspecified address (i.e. 0.0.0.0)\n\nUNICAST\nan unicast address\n\nLOCAL a local address\n\nBROADCAST\na broadcast address\n\nANYCAST\nan anycast packet\n\nMULTICAST\na multicast address\n\nBLACKHOLE\na blackhole address\n\nUNREACHABLE\nan unreachable address\n\nPROHIBIT\na prohibited address\n\nTHROW FIXME\n\nNAT FIXME\n\nXRESOLVE\n\n[!] --src-type type\nMatches if the source address is of given type\n\n[!] --dst-type type\nMatches if the destination address is of given type\n" }, { "name": "--limit-iface-in", "content": "The address type checking can be limited to the interface the packet is coming in.\nThis option is only valid in the PREROUTING, INPUT and FORWARD chains. It cannot be\nspecified with the --limit-iface-out option.\n", "long": "--limit-iface-in" }, { "name": "--limit-iface-out", "content": "The address type checking can be limited to the interface the packet is going out.\nThis option is only valid in the POSTROUTING, OUTPUT and FORWARD chains. It cannot be\nspecified with the --limit-iface-in option.\n", "long": "--limit-iface-out" }, { "name": "ah (IPv6-specific)", "content": "This module matches the parameters in Authentication header of IPsec packets.\n\n[!] --ahspi spi[:spi]\nMatches SPI.\n\n[!] --ahlen length\nTotal length of this header in octets.\n" }, { "name": "--ahres", "content": "Matches if the reserved field is filled with zero.\n", "long": "--ahres" }, { "name": "ah (IPv4-specific)", "content": "This module matches the SPIs in Authentication header of IPsec packets.\n\n[!] --ahspi spi[:spi]\n" }, { "name": "bpf", "content": "Match using Linux Socket Filter. Expects a path to an eBPF object or a cBPF program in deci‐\nmal format.\n\n--object-pinned path\nPass a path to a pinned eBPF object.\n\nApplications load eBPF programs into the kernel with the bpf() system call and BPFPROGLOAD\ncommand and can pin them in a virtual filesystem with BPFOBJPIN. To use a pinned object in\niptables, mount the bpf filesystem using\n\nmount -t bpf bpf ${BPFMOUNT}\n\nthen insert the filter in iptables by path:\n\niptables -A OUTPUT -m bpf --object-pinned ${BPFMOUNT}/{PINNEDPATH} -j ACCEPT\n\n--bytecode code\nPass the BPF byte code format as generated by the nfbpfcompile utility.\n\nThe code format is similar to the output of the tcpdump -ddd command: one line that stores\nthe number of instructions, followed by one line for each instruction. Instruction lines fol‐\nlow the pattern 'u16 u8 u8 u32' in decimal notation. Fields encode the operation, jump offset\nif true, jump offset if false and generic multiuse field 'K'. Comments are not supported.\n\nFor example, to read only packets matching 'ip proto 6', insert the following, without the\ncomments or trailing whitespace:\n\n4 # number of instructions\n48 0 0 9 # load byte ip->proto\n21 0 1 6 # jump equal IPPROTOTCP\n6 0 0 1 # return pass (non-zero)\n6 0 0 0 # return fail (zero)\n\nYou can pass this filter to the bpf match with the following command:\n\niptables -A OUTPUT -m bpf --bytecode '4,48 0 0 9,21 0 1 6,6 0 0 1,6 0 0 0' -j ACCEPT\n\nOr instead, you can invoke the nfbpfcompile utility.\n\niptables -A OUTPUT -m bpf --bytecode \"`nfbpfcompile RAW 'ip proto 6'`\" -j ACCEPT\n\nOr use tcpdump -ddd. In that case, generate BPF targeting a device with the same data link\ntype as the xtables match. Iptables passes packets from the network layer up, without mac\nlayer. Select a device with data link type RAW, such as a tun device:\n\nip tuntap add tun0 mode tun\nip link set tun0 up\ntcpdump -ddd -i tun0 ip proto 6\n\nSee tcpdump -L -i $dev for a list of known data link types for a given device.\n\nYou may want to learn more about BPF from FreeBSD's bpf(4) manpage.\n" }, { "name": "cgroup", "content": "[!] --path path\nMatch cgroup2 membership.\n\nEach socket is associated with the v2 cgroup of the creating process. This matches\npackets coming from or going to all sockets in the sub-hierarchy of the specified\npath. The path should be relative to the root of the cgroup2 hierarchy.\n\n[!] --cgroup classid\nMatch cgroup netcls classid.\n\nclassid is the marker set through the cgroup netcls controller. This option and\n--path can't be used together.\n\nExample:\n\niptables -A OUTPUT -p tcp --sport 80 -m cgroup ! --path service/http-server -j DROP\n\niptables -A OUTPUT -p tcp --sport 80 -m cgroup ! --cgroup 1 -j DROP\n\nIMPORTANT: when being used in the INPUT chain, the cgroup matcher is currently only of lim‐\nited functionality, meaning it will only match on packets that are processed for local sock‐\nets through early socket demuxing. Therefore, general usage on the INPUT chain is not advised\nunless the implications are well understood.\n\nAvailable since Linux 3.14.\n" }, { "name": "cluster", "content": "Allows you to deploy gateway and back-end load-sharing clusters without the need of load-bal‐\nancers.\n\nThis match requires that all the nodes see the same packets. Thus, the cluster match decides\nif this node has to handle a packet given the following options:\n\n--cluster-total-nodes num\nSet number of total nodes in cluster.\n\n[!] --cluster-local-node num\nSet the local node number ID.\n\n[!] --cluster-local-nodemask mask\nSet the local node number ID mask. You can use this option instead of --cluster-lo‐‐\ncal-node.\n\n--cluster-hash-seed value\nSet seed value of the Jenkins hash.\n\nExample:\n\niptables -A PREROUTING -t mangle -i eth1 -m cluster --cluster-total-nodes 2 --clus‐\nter-local-node 1 --cluster-hash-seed 0xdeadbeef -j MARK --set-mark 0xffff\n\niptables -A PREROUTING -t mangle -i eth2 -m cluster --cluster-total-nodes 2 --clus‐\nter-local-node 1 --cluster-hash-seed 0xdeadbeef -j MARK --set-mark 0xffff\n\niptables -A PREROUTING -t mangle -i eth1 -m mark ! --mark 0xffff -j DROP\n\niptables -A PREROUTING -t mangle -i eth2 -m mark ! --mark 0xffff -j DROP\n\nAnd the following commands to make all nodes see the same packets:\n\nip maddr add 01:00:5e:00:01:01 dev eth1\n\nip maddr add 01:00:5e:00:01:02 dev eth2\n\narptables -A OUTPUT -o eth1 --h-length 6 -j mangle --mangle-mac-s 01:00:5e:00:01:01\n\narptables -A INPUT -i eth1 --h-length 6 --destination-mac 01:00:5e:00:01:01 -j mangle\n--mangle-mac-d 00:zz:yy:xx:5a:27\n\narptables -A OUTPUT -o eth2 --h-length 6 -j mangle --mangle-mac-s 01:00:5e:00:01:02\n\narptables -A INPUT -i eth2 --h-length 6 --destination-mac 01:00:5e:00:01:02 -j mangle\n--mangle-mac-d 00:zz:yy:xx:5a:27\n\nNOTE: the arptables commands above use mainstream syntax. If you are using arptables-jf in‐\ncluded in some RedHat, CentOS and Fedora versions, you will hit syntax errors. Therefore,\nyou'll have to adapt these to the arptables-jf syntax to get them working.\n\nIn the case of TCP connections, pickup facility has to be disabled to avoid marking TCP ACK\npackets coming in the reply direction as valid.\n\necho 0 > /proc/sys/net/netfilter/nfconntracktcploose\n" }, { "name": "comment", "content": "Allows you to add comments (up to 256 characters) to any rule.\n\n--comment comment\n\nExample:\niptables -A INPUT -i eth1 -m comment --comment \"my local LAN\"\n" }, { "name": "connbytes", "content": "Match by how many bytes or packets a connection (or one of the two flows constituting the\nconnection) has transferred so far, or by average bytes per packet.\n\nThe counters are 64-bit and are thus not expected to overflow ;)\n\nThe primary use is to detect long-lived downloads and mark them to be scheduled using a lower\npriority band in traffic control.\n\nThe transferred bytes per connection can also be viewed through `conntrack -L` and accessed\nvia ctnetlink.\n\nNOTE that for connections which have no accounting information, the match will always return\nfalse. The \"net.netfilter.nfconntrackacct\" sysctl flag controls whether new connections\nwill be byte/packet counted. Existing connection flows will not be gaining/losing a/the ac‐\ncounting structure when be sysctl flag is flipped.\n\n[!] --connbytes from[:to]\nmatch packets from a connection whose packets/bytes/average packet size is more than\nFROM and less than TO bytes/packets. if TO is omitted only FROM check is done. \"!\" is\nused to match packets not falling in the range.\n\n--connbytes-dir {original|reply|both}\nwhich packets to consider\n\n--connbytes-mode {packets|bytes|avgpkt}\nwhether to check the amount of packets, number of bytes transferred or the average\nsize (in bytes) of all packets received so far. Note that when \"both\" is used together\nwith \"avgpkt\", and data is going (mainly) only in one direction (for example HTTP),\nthe average packet size will be about half of the actual data packets.\n\nExample:\niptables .. -m connbytes --connbytes 10000:100000 --connbytes-dir both\n--connbytes-mode bytes ...\n" }, { "name": "connlabel", "content": "Module matches or adds connlabels to a connection. connlabels are similar to connmarks, ex‐\ncept labels are bit-based; i.e. all labels may be attached to a flow at the same time. Up\nto 128 unique labels are currently supported.\n\n[!] --label name\nmatches if label name has been set on a connection. Instead of a name (which will be\ntranslated to a number, see EXAMPLE below), a number may be used instead. Using a\nnumber always overrides connlabel.conf.\n\n--set if the label has not been set on the connection, set it. Note that setting a label\ncan fail. This is because the kernel allocates the conntrack label storage area when\nthe connection is created, and it only reserves the amount of memory required by the\nruleset that exists at the time the connection is created. In this case, the match\nwill fail (or succeed, in case --label option was negated).\n\nThis match depends on libnetfilterconntrack 1.0.4 or later. Label translation is done via\nthe /etc/xtables/connlabel.conf configuration file.\n\nExample:\n\n0 eth0-in\n1 eth0-out\n2 ppp-in\n3 ppp-out\n4 bulk-traffic\n5 interactive\n" }, { "name": "connlimit", "content": "Allows you to restrict the number of parallel connections to a server per client IP address\n(or client address block).\n\n--connlimit-upto n\nMatch if the number of existing connections is below or equal n.\n\n--connlimit-above n\nMatch if the number of existing connections is above n.\n\n--connlimit-mask prefixlength\nGroup hosts using the prefix length. For IPv4, this must be a number between (includ‐\ning) 0 and 32. For IPv6, between 0 and 128. If not specified, the maximum prefix\nlength for the applicable protocol is used.\n" }, { "name": "--connlimit-saddr", "content": "Apply the limit onto the source group. This is the default if --connlimit-daddr is not\nspecified.\n", "long": "--connlimit-saddr" }, { "name": "--connlimit-daddr", "content": "Apply the limit onto the destination group.\n\nExamples:\n\n# allow 2 telnet connections per client host\niptables -A INPUT -p tcp --syn --dport 23 -m connlimit --connlimit-above 2 -j REJECT\n\n# you can also match the other way around:\niptables -A INPUT -p tcp --syn --dport 23 -m connlimit --connlimit-upto 2 -j ACCEPT\n\n# limit the number of parallel HTTP requests to 16 per class C sized source network (24 bit\nnetmask)\niptables -p tcp --syn --dport 80 -m connlimit --connlimit-above 16 --connlimit-mask 24\n-j REJECT\n\n# limit the number of parallel HTTP requests to 16 for the link local network\n(ipv6) ip6tables -p tcp --syn --dport 80 -s fe80::/64 -m connlimit --connlimit-above\n16 --connlimit-mask 64 -j REJECT\n\n# Limit the number of connections to a particular host:\nip6tables -p tcp --syn --dport 49152:65535 -d 2001:db8::1 -m connlimit --connlimit-\nabove 100 -j REJECT\n", "long": "--connlimit-daddr" }, { "name": "connmark", "content": "This module matches the netfilter mark field associated with a connection (which can be set\nusing the CONNMARK target below).\n\n[!] --mark value[/mask]\nMatches packets in connections with the given mark value (if a mask is specified, this\nis logically ANDed with the mark before the comparison).\n" }, { "name": "conntrack", "content": "This module, when combined with connection tracking, allows access to the connection tracking\nstate for this packet/connection.\n\n[!] --ctstate statelist\nstatelist is a comma separated list of the connection states to match. Possible\nstates are listed below.\n\n[!] --ctproto l4proto\nLayer-4 protocol to match (by number or name)\n\n[!] --ctorigsrc address[/mask]\n\n[!] --ctorigdst address[/mask]\n\n[!] --ctreplsrc address[/mask]\n\n[!] --ctrepldst address[/mask]\nMatch against original/reply source/destination address\n\n[!] --ctorigsrcport port[:port]\n\n[!] --ctorigdstport port[:port]\n\n[!] --ctreplsrcport port[:port]\n\n[!] --ctrepldstport port[:port]\nMatch against original/reply source/destination port (TCP/UDP/etc.) or GRE key.\nMatching against port ranges is only supported in kernel versions above 2.6.38.\n\n[!] --ctstatus statelist\nstatuslist is a comma separated list of the connection statuses to match. Possible\nstatuses are listed below.\n\n[!] --ctexpire time[:time]\nMatch remaining lifetime in seconds against given value or range of values (inclusive)\n\n--ctdir {ORIGINAL|REPLY}\nMatch packets that are flowing in the specified direction. If this flag is not speci‐\nfied at all, matches packets in both directions.\n\nStates for --ctstate:\n\nINVALID\nThe packet is associated with no known connection.\n\nNEW The packet has started a new connection or otherwise associated with a connection\nwhich has not seen packets in both directions.\n\nESTABLISHED\nThe packet is associated with a connection which has seen packets in both directions.\n\nRELATED\nThe packet is starting a new connection, but is associated with an existing connec‐\ntion, such as an FTP data transfer or an ICMP error.\n\nUNTRACKED\nThe packet is not tracked at all, which happens if you explicitly untrack it by using\n-j CT --notrack in the raw table.\n\nSNAT A virtual state, matching if the original source address differs from the reply desti‐\nnation.\n\nDNAT A virtual state, matching if the original destination differs from the reply source.\n\nStatuses for --ctstatus:\n\nNONE None of the below.\n\nEXPECTED\nThis is an expected connection (i.e. a conntrack helper set it up).\n\nSEENREPLY\nConntrack has seen packets in both directions.\n\nASSURED\nConntrack entry should never be early-expired.\n\nCONFIRMED\nConnection is confirmed: originating packet has left box.\n" }, { "name": "cpu", "content": "[!] --cpu number\nMatch cpu handling this packet. cpus are numbered from 0 to NRCPUS-1 Can be used in\ncombination with RPS (Remote Packet Steering) or multiqueue NICs to spread network\ntraffic on different queues.\n\nExample:\n\niptables -t nat -A PREROUTING -p tcp --dport 80 -m cpu --cpu 0 -j REDIRECT --to-port 8080\n\niptables -t nat -A PREROUTING -p tcp --dport 80 -m cpu --cpu 1 -j REDIRECT --to-port 8081\n\nAvailable since Linux 2.6.36.\n" }, { "name": "dccp", "content": "[!] --source-port,--sport port[:port]\n\n[!] --destination-port,--dport port[:port]\n\n[!] --dccp-types mask\nMatch when the DCCP packet type is one of 'mask'. 'mask' is a comma-separated list of\npacket types. Packet types are: REQUEST RESPONSE DATA ACK DATAACK CLOSEREQ CLOSE RE‐‐\nSET SYNC SYNCACK INVALID.\n\n[!] --dccp-option number\nMatch if DCCP option set.\n" }, { "name": "devgroup", "content": "Match device group of a packets incoming/outgoing interface.\n\n[!] --src-group name\nMatch device group of incoming device\n\n[!] --dst-group name\nMatch device group of outgoing device\n" }, { "name": "dscp", "content": "This module matches the 6 bit DSCP field within the TOS field in the IP header. DSCP has su‐\nperseded TOS within the IETF.\n\n[!] --dscp value\nMatch against a numeric (decimal or hex) value [0-63].\n\n[!] --dscp-class class\nMatch the DiffServ class. This value may be any of the BE, EF, AFxx or CSx classes.\nIt will then be converted into its according numeric value.\n" }, { "name": "dst (IPv6-specific)", "content": "This module matches the parameters in Destination Options header\n\n[!] --dst-len length\nTotal length of this header in octets.\n\n--dst-opts type[:length][,type[:length]...]\nnumeric type of option and the length of the option data in octets.\n" }, { "name": "ecn", "content": "This allows you to match the ECN bits of the IPv4/IPv6 and TCP header. ECN is the Explicit\nCongestion Notification mechanism as specified in RFC3168\n\n[!] --ecn-tcp-cwr\nThis matches if the TCP ECN CWR (Congestion Window Received) bit is set.\n\n[!] --ecn-tcp-ece\nThis matches if the TCP ECN ECE (ECN Echo) bit is set.\n\n[!] --ecn-ip-ect num\nThis matches a particular IPv4/IPv6 ECT (ECN-Capable Transport). You have to specify a\nnumber between `0' and `3'.\n" }, { "name": "esp", "content": "This module matches the SPIs in ESP header of IPsec packets.\n\n[!] --espspi spi[:spi]\n" }, { "name": "eui64 (IPv6-specific)", "content": "This module matches the EUI-64 part of a stateless autoconfigured IPv6 address. It compares\nthe EUI-64 derived from the source MAC address in Ethernet frame with the lower 64 bits of\nthe IPv6 source address. But \"Universal/Local\" bit is not compared. This module doesn't match\nother link layer frame, and is only valid in the PREROUTING, INPUT and FORWARD chains.\n" }, { "name": "frag (IPv6-specific)", "content": "This module matches the parameters in Fragment header.\n\n[!] --fragid id[:id]\nMatches the given Identification or range of it.\n\n[!] --fraglen length\nThis option cannot be used with kernel version 2.6.10 or later. The length of Fragment\nheader is static and this option doesn't make sense.\n" }, { "name": "--fragres", "content": "Matches if the reserved fields are filled with zero.\n", "long": "--fragres" }, { "name": "--fragfirst", "content": "Matches on the first fragment.\n", "long": "--fragfirst" }, { "name": "--fragmore", "content": "Matches if there are more fragments.\n", "long": "--fragmore" }, { "name": "--fraglast", "content": "Matches if this is the last fragment.\n", "long": "--fraglast" }, { "name": "hashlimit", "content": "hashlimit uses hash buckets to express a rate limiting match (like the limit match) for a\ngroup of connections using a single iptables rule. Grouping can be done per-hostgroup (source\nand/or destination address) and/or per-port. It gives you the ability to express \"N packets\nper time quantum per group\" or \"N bytes per seconds\" (see below for some examples).\n\nA hash limit option (--hashlimit-upto, --hashlimit-above) and --hashlimit-name are required.\n\n--hashlimit-upto amount[/second|/minute|/hour|/day]\nMatch if the rate is below or equal to amount/quantum. It is specified either as a\nnumber, with an optional time quantum suffix (the default is 3/hour), or as\namountb/second (number of bytes per second).\n\n--hashlimit-above amount[/second|/minute|/hour|/day]\nMatch if the rate is above amount/quantum.\n\n--hashlimit-burst amount\nMaximum initial number of packets to match: this number gets recharged by one every\ntime the limit specified above is not reached, up to this number; the default is 5.\nWhen byte-based rate matching is requested, this option specifies the amount of bytes\nthat can exceed the given rate. This option should be used with caution -- if the en‐\ntry expires, the burst value is reset too.\n\n--hashlimit-mode {srcip|srcport|dstip|dstport},...\nA comma-separated list of objects to take into consideration. If no --hashlimit-mode\noption is given, hashlimit acts like limit, but at the expensive of doing the hash\nhousekeeping.\n\n--hashlimit-srcmask prefix\nWhen --hashlimit-mode srcip is used, all source addresses encountered will be grouped\naccording to the given prefix length and the so-created subnet will be subject to\nhashlimit. prefix must be between (inclusive) 0 and 32. Note that --hashlimit-srcmask\n0 is basically doing the same thing as not specifying srcip for --hashlimit-mode, but\nis technically more expensive.\n\n--hashlimit-dstmask prefix\nLike --hashlimit-srcmask, but for destination addresses.\n\n--hashlimit-name foo\nThe name for the /proc/net/ipthashlimit/foo entry.\n\n--hashlimit-htable-size buckets\nThe number of buckets of the hash table\n\n--hashlimit-htable-max entries\nMaximum entries in the hash.\n\n--hashlimit-htable-expire msec\nAfter how many milliseconds do hash entries expire.\n\n--hashlimit-htable-gcinterval msec\nHow many milliseconds between garbage collection intervals.\n" }, { "name": "--hashlimit-rate-match", "content": "Classify the flow instead of rate-limiting it. This acts like a true/false match on\nwhether the rate is above/below a certain number\n\n--hashlimit-rate-interval sec\nCan be used with --hashlimit-rate-match to specify the interval at which the rate\nshould be sampled\n\nExamples:\n\nmatching on source host\n\"1000 packets per second for every host in 192.168.0.0/16\" => -s 192.168.0.0/16\n--hashlimit-mode srcip --hashlimit-upto 1000/sec\n\nmatching on source port\n\"100 packets per second for every service of 192.168.1.1\" => -s 192.168.1.1 --hash‐\nlimit-mode srcport --hashlimit-upto 100/sec\n\nmatching on subnet\n\"10000 packets per minute for every /28 subnet (groups of 8 addresses) in 10.0.0.0/8\"\n=> -s 10.0.0.0/8 --hashlimit-mask 28 --hashlimit-upto 10000/min\n\nmatching bytes per second\n\"flows exceeding 512kbyte/s\" => --hashlimit-mode srcip,dstip,srcport,dstport --hash‐\nlimit-above 512kb/s\n\nmatching bytes per second\n\"hosts that exceed 512kbyte/s, but permit up to 1Megabytes without matching\" --hash‐\nlimit-mode dstip --hashlimit-above 512kb/s --hashlimit-burst 1mb\n", "long": "--hashlimit-rate-match" }, { "name": "hbh (IPv6-specific)", "content": "This module matches the parameters in Hop-by-Hop Options header\n\n[!] --hbh-len length\nTotal length of this header in octets.\n\n--hbh-opts type[:length][,type[:length]...]\nnumeric type of option and the length of the option data in octets.\n" }, { "name": "helper", "content": "This module matches packets related to a specific conntrack-helper.\n\n[!] --helper string\nMatches packets related to the specified conntrack-helper.\n\nstring can be \"ftp\" for packets related to a ftp-session on default port. For other\nports append -portnr to the value, ie. \"ftp-2121\".\n\nSame rules apply for other conntrack-helpers.\n" }, { "name": "hl (IPv6-specific)", "content": "This module matches the Hop Limit field in the IPv6 header.\n\n[!] --hl-eq value\nMatches if Hop Limit equals value.\n\n--hl-lt value\nMatches if Hop Limit is less than value.\n\n--hl-gt value\nMatches if Hop Limit is greater than value.\n" }, { "name": "icmp (IPv4-specific)", "content": "This extension can be used if `--protocol icmp' is specified. It provides the following op‐\ntion:\n\n[!] --icmp-type {type[/code]|typename}\nThis allows specification of the ICMP type, which can be a numeric ICMP type,\ntype/code pair, or one of the ICMP type names shown by the command\niptables -p icmp -h\n" }, { "name": "icmp6 (IPv6-specific)", "content": "This extension can be used if `--protocol ipv6-icmp' or `--protocol icmpv6' is specified. It\nprovides the following option:\n\n[!] --icmpv6-type type[/code]|typename\nThis allows specification of the ICMPv6 type, which can be a numeric ICMPv6 type, type\nand code, or one of the ICMPv6 type names shown by the command\nip6tables -p ipv6-icmp -h\n" }, { "name": "iprange", "content": "This matches on a given arbitrary range of IP addresses.\n\n[!] --src-range from[-to]\nMatch source IP in the specified range.\n\n[!] --dst-range from[-to]\nMatch destination IP in the specified range.\n" }, { "name": "ipv6header (IPv6-specific)", "content": "This module matches IPv6 extension headers and/or upper layer header.\n\n--soft Matches if the packet includes any of the headers specified with --header.\n\n[!] --header header[,header...]\nMatches the packet which EXACTLY includes all specified headers. The headers encapsu‐\nlated with ESP header are out of scope. Possible header types can be:\n\nhop|hop-by-hop\nHop-by-Hop Options header\n\ndst Destination Options header\n\nroute Routing header\n\nfrag Fragment header\n\nauth Authentication header\n\nesp Encapsulating Security Payload header\n\nnone No Next header which matches 59 in the 'Next Header field' of IPv6 header or any IPv6\nextension headers\n\nprot which matches any upper layer protocol header. A protocol name from /etc/protocols and\nnumeric value also allowed. The number 255 is equivalent to prot.\n" }, { "name": "ipvs", "content": "Match IPVS connection properties.\n\n[!] --ipvs\npacket belongs to an IPVS connection\n\nAny of the following options implies --ipvs (even negated)\n\n[!] --vproto protocol\nVIP protocol to match; by number or name, e.g. \"tcp\"\n\n[!] --vaddr address[/mask]\nVIP address to match\n\n[!] --vport port\nVIP port to match; by number or name, e.g. \"http\"\n\n--vdir {ORIGINAL|REPLY}\nflow direction of packet\n\n[!] --vmethod {GATE|IPIP|MASQ}\nIPVS forwarding method used\n\n[!] --vportctl port\nVIP port of the controlling connection to match, e.g. 21 for FTP\n" }, { "name": "length", "content": "This module matches the length of the layer-3 payload (e.g. layer-4 packet) of a packet\nagainst a specific value or range of values.\n\n[!] --length length[:length]\n" }, { "name": "limit", "content": "This module matches at a limited rate using a token bucket filter. A rule using this exten‐\nsion will match until this limit is reached. It can be used in combination with the LOG tar‐\nget to give limited logging, for example.\n\nxtlimit has no negation support - you will have to use -m hashlimit ! --hashlimit rate in\nthis case whilst omitting --hashlimit-mode.\n\n--limit rate[/second|/minute|/hour|/day]\nMaximum average matching rate: specified as a number, with an optional `/second',\n`/minute', `/hour', or `/day' suffix; the default is 3/hour.\n\n--limit-burst number\nMaximum initial number of packets to match: this number gets recharged by one every\ntime the limit specified above is not reached, up to this number; the default is 5.\n" }, { "name": "mac", "content": "[!] --mac-source address\nMatch source MAC address. It must be of the form XX:XX:XX:XX:XX:XX. Note that this\nonly makes sense for packets coming from an Ethernet device and entering the PREROUT‐‐\nING, FORWARD or INPUT chains.\n" }, { "name": "mark", "content": "This module matches the netfilter mark field associated with a packet (which can be set using\nthe MARK target below).\n\n[!] --mark value[/mask]\nMatches packets with the given unsigned mark value (if a mask is specified, this is\nlogically ANDed with the mask before the comparison).\n" }, { "name": "mh (IPv6-specific)", "content": "This extension is loaded if `--protocol ipv6-mh' or `--protocol mh' is specified. It provides\nthe following option:\n\n[!] --mh-type type[:type]\nThis allows specification of the Mobility Header(MH) type, which can be a numeric MH\ntype, type or one of the MH type names shown by the command\nip6tables -p mh -h\n" }, { "name": "multiport", "content": "This module matches a set of source or destination ports. Up to 15 ports can be specified.\nA port range (port:port) counts as two ports. It can only be used in conjunction with one of\nthe following protocols: tcp, udp, udplite, dccp and sctp.\n\n[!] --source-ports,--sports port[,port|,port:port]...\nMatch if the source port is one of the given ports. The flag --sports is a convenient\nalias for this option. Multiple ports or port ranges are separated using a comma, and\na port range is specified using a colon. 53,1024:65535 would therefore match ports 53\nand all from 1024 through 65535.\n\n[!] --destination-ports,--dports port[,port|,port:port]...\nMatch if the destination port is one of the given ports. The flag --dports is a con‐\nvenient alias for this option.\n\n[!] --ports port[,port|,port:port]...\nMatch if either the source or destination ports are equal to one of the given ports.\n" }, { "name": "nfacct", "content": "The nfacct match provides the extended accounting infrastructure for iptables. You have to\nuse this match together with the standalone user-space utility nfacct(8)\n\nThe only option available for this match is the following:\n\n--nfacct-name name\nThis allows you to specify the existing object name that will be use for accounting\nthe traffic that this rule-set is matching.\n\nTo use this extension, you have to create an accounting object:\n\nnfacct add http-traffic\n\nThen, you have to attach it to the accounting object via iptables:\n\niptables -I INPUT -p tcp --sport 80 -m nfacct --nfacct-name http-traffic\n\niptables -I OUTPUT -p tcp --dport 80 -m nfacct --nfacct-name http-traffic\n\nThen, you can check for the amount of traffic that the rules match:\n\nnfacct get http-traffic\n\n{ pkts = 00000000000000000156, bytes = 00000000000000151786 } = http-traffic;\n\nYou can obtain nfacct(8) from http://www.netfilter.org or, alternatively, from the git.net‐\nfilter.org repository.\n" }, { "name": "osf", "content": "The osf module does passive operating system fingerprinting. This module compares some data\n(Window Size, MSS, options and their order, TTL, DF, and others) from packets with the SYN\nbit set.\n\n[!] --genre string\nMatch an operating system genre by using a passive fingerprinting.\n\n--ttl level\nDo additional TTL checks on the packet to determine the operating system. level can\nbe one of the following values:\n\n• 0 - True IP address and fingerprint TTL comparison. This generally works for LANs.\n\n• 1 - Check if the IP header's TTL is less than the fingerprint one. Works for globally-\nroutable addresses.\n\n• 2 - Do not compare the TTL at all.\n\n--log level\nLog determined genres into dmesg even if they do not match the desired one. level can be\none of the following values:\n\n• 0 - Log all matched or unknown signatures\n\n• 1 - Log only the first one\n\n• 2 - Log all known matched signatures\n\nYou may find something like this in syslog:\n\nWindows [2000:SP3:Windows XP Pro SP1, 2000 SP3]: 11.22.33.55:4024 -> 11.22.33.44:139 hops=3\nLinux [2.5-2.6:] : 1.2.3.4:42624 -> 1.2.3.5:22 hops=4\n\nOS fingerprints are loadable using the nfnlosf program. To load fingerprints from a file,\nuse:\n\nnfnlosf -f /usr/share/xtables/pf.os\n\nTo remove them again,\n\nnfnlosf -f /usr/share/xtables/pf.os -d\n\nThe fingerprint database can be downloaded from http://www.openbsd.org/cgi-\nbin/cvsweb/src/etc/pf.os .\n" }, { "name": "owner", "content": "This module attempts to match various characteristics of the packet creator, for locally gen‐\nerated packets. This match is only valid in the OUTPUT and POSTROUTING chains. Forwarded\npackets do not have any socket associated with them. Packets from kernel threads do have a\nsocket, but usually no owner.\n\n[!] --uid-owner username\n\n[!] --uid-owner userid[-userid]\nMatches if the packet socket's file structure (if it has one) is owned by the given\nuser. You may also specify a numerical UID, or an UID range.\n\n[!] --gid-owner groupname\n\n[!] --gid-owner groupid[-groupid]\nMatches if the packet socket's file structure is owned by the given group. You may\nalso specify a numerical GID, or a GID range.\n" }, { "name": "--suppl-groups", "content": "Causes group(s) specified with --gid-owner to be also checked in the supplementary\ngroups of a process.\n\n[!] --socket-exists\nMatches if the packet is associated with a socket.\n", "long": "--suppl-groups" }, { "name": "physdev", "content": "This module matches on the bridge port input and output devices enslaved to a bridge device.\nThis module is a part of the infrastructure that enables a transparent bridging IP firewall\nand is only useful for kernel versions above version 2.5.44.\n\n[!] --physdev-in name\nName of a bridge port via which a packet is received (only for packets entering the\nINPUT, FORWARD and PREROUTING chains). If the interface name ends in a \"+\", then any\ninterface which begins with this name will match. If the packet didn't arrive through\na bridge device, this packet won't match this option, unless '!' is used.\n\n[!] --physdev-out name\nName of a bridge port via which a packet is going to be sent (for bridged packets en‐\ntering the FORWARD and POSTROUTING chains). If the interface name ends in a \"+\", then\nany interface which begins with this name will match.\n\n[!] --physdev-is-in\nMatches if the packet has entered through a bridge interface.\n\n[!] --physdev-is-out\nMatches if the packet will leave through a bridge interface.\n\n[!] --physdev-is-bridged\nMatches if the packet is being bridged and therefore is not being routed. This is\nonly useful in the FORWARD and POSTROUTING chains.\n" }, { "name": "pkttype", "content": "This module matches the link-layer packet type.\n\n[!] --pkt-type {unicast|broadcast|multicast}\n" }, { "name": "policy", "content": "This module matches the policy used by IPsec for handling a packet.\n\n--dir {in|out}\nUsed to select whether to match the policy used for decapsulation or the policy that\nwill be used for encapsulation. in is valid in the PREROUTING, INPUT and FORWARD\nchains, out is valid in the POSTROUTING, OUTPUT and FORWARD chains.\n\n--pol {none|ipsec}\nMatches if the packet is subject to IPsec processing. --pol none cannot be combined\nwith --strict.\n" }, { "name": "--strict", "content": "Selects whether to match the exact policy or match if any rule of the policy matches\nthe given policy.\n\nFor each policy element that is to be described, one can use one or more of the following op‐\ntions. When --strict is in effect, at least one must be used per element.\n\n[!] --reqid id\nMatches the reqid of the policy rule. The reqid can be specified with setkey(8) using\nunique:id as level.\n\n[!] --spi spi\nMatches the SPI of the SA.\n\n[!] --proto {ah|esp|ipcomp}\nMatches the encapsulation protocol.\n\n[!] --mode {tunnel|transport}\nMatches the encapsulation mode.\n\n[!] --tunnel-src addr[/mask]\nMatches the source end-point address of a tunnel mode SA. Only valid with --mode tun‐‐\nnel.\n\n[!] --tunnel-dst addr[/mask]\nMatches the destination end-point address of a tunnel mode SA. Only valid with --mode\ntunnel.\n\n--next Start the next element in the policy specification. Can only be used with --strict.\n", "long": "--strict" }, { "name": "quota", "content": "Implements network quotas by decrementing a byte counter with each packet. The condition\nmatches until the byte counter reaches zero. Behavior is reversed with negation (i.e. the\ncondition does not match until the byte counter reaches zero).\n\n[!] --quota bytes\nThe quota in bytes.\n" }, { "name": "rateest", "content": "The rate estimator can match on estimated rates as collected by the RATEEST target. It sup‐\nports matching on absolute bps/pps values, comparing two rate estimators and matching on the\ndifference between two rate estimators.\n\nFor a better understanding of the available options, these are all possible combinations:\n\n• rateest operator rateest-bps\n\n• rateest operator rateest-pps\n\n• (rateest minus rateest-bps1) operator rateest-bps2\n\n• (rateest minus rateest-pps1) operator rateest-pps2\n\n• rateest1 operator rateest2 rateest-bps(without rate!)\n\n• rateest1 operator rateest2 rateest-pps(without rate!)\n\n• (rateest1 minus rateest-bps1) operator (rateest2 minus rateest-bps2)\n\n• (rateest1 minus rateest-pps1) operator (rateest2 minus rateest-pps2)\n" }, { "name": "--rateest-delta", "content": "For each estimator (either absolute or relative mode), calculate the difference between\nthe estimator-determined flow rate and the static value chosen with the BPS/PPS options.\nIf the flow rate is higher than the specified BPS/PPS, 0 will be used instead of a nega‐\ntive value. In other words, \"max(0, rateest#rate - rateest#bps)\" is used.\n\n[!] --rateest-lt\nMatch if rate is less than given rate/estimator.\n\n[!] --rateest-gt\nMatch if rate is greater than given rate/estimator.\n\n[!] --rateest-eq\nMatch if rate is equal to given rate/estimator.\n\nIn the so-called \"absolute mode\", only one rate estimator is used and compared against a\nstatic value, while in \"relative mode\", two rate estimators are compared against another.\n\n--rateest name\nName of the one rate estimator for absolute mode.\n\n--rateest1 name\n\n--rateest2 name\nThe names of the two rate estimators for relative mode.\n\n--rateest-bps [value]\n\n--rateest-pps [value]\n\n--rateest-bps1 [value]\n\n--rateest-bps2 [value]\n\n--rateest-pps1 [value]\n\n--rateest-pps2 [value]\nCompare the estimator(s) by bytes or packets per second, and compare against the cho‐\nsen value. See the above bullet list for which option is to be used in which case. A\nunit suffix may be used - available ones are: bit, [kmgt]bit, [KMGT]ibit, Bps,\n[KMGT]Bps, [KMGT]iBps.\n\nExample: This is what can be used to route outgoing data connections from an FTP server over\ntwo lines based on the available bandwidth at the time the data connection was started:\n\n# Estimate outgoing rates\n\niptables -t mangle -A POSTROUTING -o eth0 -j RATEEST --rateest-name eth0 --rateest-interval\n250ms --rateest-ewma 0.5s\n\niptables -t mangle -A POSTROUTING -o ppp0 -j RATEEST --rateest-name ppp0 --rateest-interval\n250ms --rateest-ewma 0.5s\n\n# Mark based on available bandwidth\n\niptables -t mangle -A balance -m conntrack --ctstate NEW -m helper --helper ftp -m rateest\n--rateest-delta --rateest1 eth0 --rateest-bps1 2.5mbit --rateest-gt --rateest2 ppp0 --ra‐\nteest-bps2 2mbit -j CONNMARK --set-mark 1\n\niptables -t mangle -A balance -m conntrack --ctstate NEW -m helper --helper ftp -m rateest\n--rateest-delta --rateest1 ppp0 --rateest-bps1 2mbit --rateest-gt --rateest2 eth0 --ra‐\nteest-bps2 2.5mbit -j CONNMARK --set-mark 2\n\niptables -t mangle -A balance -j CONNMARK --restore-mark\n", "long": "--rateest-delta" }, { "name": "realm (IPv4-specific)", "content": "This matches the routing realm. Routing realms are used in complex routing setups involving\ndynamic routing protocols like BGP.\n\n[!] --realm value[/mask]\nMatches a given realm number (and optionally mask). If not a number, value can be a\nnamed realm from /etc/iproute2/rtrealms (mask can not be used in that case). Both\nvalue and mask are four byte unsigned integers and may be specified in decimal, hex\n(by prefixing with \"0x\") or octal (if a leading zero is given).\n" }, { "name": "recent", "content": "Allows you to dynamically create a list of IP addresses and then match against that list in a\nfew different ways.\n\nFor example, you can create a \"badguy\" list out of people attempting to connect to port 139\non your firewall and then DROP all future packets from them without considering them.\n\n--set, --rcheck, --update and --remove are mutually exclusive.\n\n--name name\nSpecify the list to use for the commands. If no name is given then DEFAULT will be\nused.\n\n[!] --set\nThis will add the source address of the packet to the list. If the source address is\nalready in the list, this will update the existing entry. This will always return suc‐\ncess (or failure if ! is passed in).\n" }, { "name": "--rsource", "content": "Match/save the source address of each packet in the recent list table. This is the de‐\nfault.\n", "long": "--rsource" }, { "name": "--rdest", "content": "Match/save the destination address of each packet in the recent list table.\n\n--mask netmask\nNetmask that will be applied to this recent list.\n\n[!] --rcheck\nCheck if the source address of the packet is currently in the list.\n\n[!] --update\nLike --rcheck, except it will update the \"last seen\" timestamp if it matches.\n\n[!] --remove\nCheck if the source address of the packet is currently in the list and if so that ad‐\ndress will be removed from the list and the rule will return true. If the address is\nnot found, false is returned.\n\n--seconds seconds\nThis option must be used in conjunction with one of --rcheck or --update. When used,\nthis will narrow the match to only happen when the address is in the list and was seen\nwithin the last given number of seconds.\n\n--reap This option can only be used in conjunction with --seconds. When used, this will\ncause entries older than the last given number of seconds to be purged.\n\n--hitcount hits\nThis option must be used in conjunction with one of --rcheck or --update. When used,\nthis will narrow the match to only happen when the address is in the list and packets\nhad been received greater than or equal to the given value. This option may be used\nalong with --seconds to create an even narrower match requiring a certain number of\nhits within a specific time frame. The maximum value for the hitcount parameter is\ngiven by the \"ippktlisttot\" parameter of the xtrecent kernel module. Exceeding\nthis value on the command line will cause the rule to be rejected.\n\n--rttl This option may only be used in conjunction with one of --rcheck or --update. When\nused, this will narrow the match to only happen when the address is in the list and\nthe TTL of the current packet matches that of the packet which hit the --set rule.\nThis may be useful if you have problems with people faking their source address in or‐\nder to DoS you via this module by disallowing others access to your site by sending\nbogus packets to you.\n\nExamples:\n\niptables -A FORWARD -m recent --name badguy --rcheck --seconds 60 -j DROP\n\niptables -A FORWARD -p tcp -i eth0 --dport 139 -m recent --name badguy --set -j DROP\n\n/proc/net/xtrecent/* are the current lists of addresses and information about each entry of\neach list.\n\nEach file in /proc/net/xtrecent/ can be read from to see the current list or written two us‐\ning the following commands to modify the list:\n\necho +addr >/proc/net/xtrecent/DEFAULT\nto add addr to the DEFAULT list\n\necho -addr >/proc/net/xtrecent/DEFAULT\nto remove addr from the DEFAULT list\n\necho / >/proc/net/xtrecent/DEFAULT\nto flush the DEFAULT list (remove all entries).\n\nThe module itself accepts parameters, defaults shown:\n\niplisttot=100\nNumber of addresses remembered per table.\n\nippktlisttot=20\nNumber of packets per address remembered.\n\niplisthashsize=0\nHash table size. 0 means to calculate it based on iplisttot, default: 512.\n\niplistperms=0644\nPermissions for /proc/net/xtrecent/* files.\n\niplistuid=0\nNumerical UID for ownership of /proc/net/xtrecent/* files.\n\niplistgid=0\nNumerical GID for ownership of /proc/net/xtrecent/* files.\n", "long": "--rdest" }, { "name": "rpfilter", "content": "Performs a reverse path filter test on a packet. If a reply to the packet would be sent via\nthe same interface that the packet arrived on, the packet will match. Note that, unlike the\nin-kernel rpfilter, packets protected by IPSec are not treated specially. Combine this\nmatch with the policy match if you want this. Also, packets arriving via the loopback inter‐\nface are always permitted. This match can only be used in the PREROUTING chain of the raw or\nmangle table.\n" }, { "name": "--loose", "content": "Used to specify that the reverse path filter test should match even if the selected\noutput device is not the expected one.\n", "long": "--loose" }, { "name": "--validmark", "content": "Also use the packets' nfmark value when performing the reverse path route lookup.\n", "long": "--validmark" }, { "name": "--accept-local", "content": "This will permit packets arriving from the network with a source address that is also\nassigned to the local machine.\n", "long": "--accept-local" }, { "name": "--invert", "content": "This will invert the sense of the match. Instead of matching packets that passed the\nreverse path filter test, match those that have failed it.\n\nExample to log and drop packets failing the reverse path filter test:\n\niptables -t raw -N RPFILTER\n\niptables -t raw -A RPFILTER -m rpfilter -j RETURN\n\niptables -t raw -A RPFILTER -m limit --limit 10/minute -j NFLOG --nflog-prefix \"rpfilter\ndrop\"\n\niptables -t raw -A RPFILTER -j DROP\n\niptables -t raw -A PREROUTING -j RPFILTER\n\nExample to drop failed packets, without logging:\n\niptables -t raw -A RPFILTER -m rpfilter --invert -j DROP\n", "long": "--invert" }, { "name": "rt (IPv6-specific)", "content": "Match on IPv6 routing header\n\n[!] --rt-type type\nMatch the type (numeric).\n\n[!] --rt-segsleft num[:num]\nMatch the `segments left' field (range).\n\n[!] --rt-len length\nMatch the length of this header.\n" }, { "name": "--rt-0-res", "content": "Match the reserved field, too (type=0)\n\n--rt-0-addrs addr[,addr...]\nMatch type=0 addresses (list).\n", "long": "--rt-0-res" }, { "name": "--rt-0-not-strict", "content": "List of type=0 addresses is not a strict list.\n", "long": "--rt-0-not-strict" }, { "name": "sctp", "content": "This module matches Stream Control Transmission Protocol headers.\n\n[!] --source-port,--sport port[:port]\n\n[!] --destination-port,--dport port[:port]\n\n[!] --chunk-types {all|any|only} chunktype[:flags] [...]\nThe flag letter in upper case indicates that the flag is to match if set, in the lower\ncase indicates to match if unset.\n\nChunk types: DATA INIT INITACK SACK HEARTBEAT HEARTBEATACK ABORT SHUTDOWN SHUT‐\nDOWNACK ERROR COOKIEECHO COOKIEACK ECNECNE ECNCWR SHUTDOWNCOMPLETE ASCONF AS‐\nCONFACK FORWARDTSN\n\nchunk type available flags\nDATA I U B E i u b e\nABORT T t\nSHUTDOWNCOMPLETE T t\n\n(lowercase means flag should be \"off\", uppercase means \"on\")\n\nExamples:\n\niptables -A INPUT -p sctp --dport 80 -j DROP\n\niptables -A INPUT -p sctp --chunk-types any DATA,INIT -j DROP\n\niptables -A INPUT -p sctp --chunk-types any DATA:Be -j ACCEPT\n" }, { "name": "set", "content": "This module matches IP sets which can be defined by ipset(8).\n\n[!] --match-set setname flag[,flag]...\nwhere flags are the comma separated list of src and/or dst specifications and there\ncan be no more than six of them. Hence the command\n\niptables -A FORWARD -m set --match-set test src,dst\n\nwill match packets, for which (if the set type is ipportmap) the source address and\ndestination port pair can be found in the specified set. If the set type of the speci‐\nfied set is single dimension (for example ipmap), then the command will match packets\nfor which the source address can be found in the specified set.\n" }, { "name": "--return-nomatch", "content": "If the --return-nomatch option is specified and the set type supports the nomatch\nflag, then the matching is reversed: a match with an element flagged with nomatch re‐\nturns true, while a match with a plain element returns false.\n", "long": "--return-nomatch" }, { "name": "! --update-counters", "content": "If the --update-counters flag is negated, then the packet and byte counters of the\nmatching element in the set won't be updated. Default the packet and byte counters are\nupdated.\n" }, { "name": "! --update-subcounters", "content": "If the --update-subcounters flag is negated, then the packet and byte counters of the\nmatching element in the member set of a list type of set won't be updated. Default the\npacket and byte counters are updated.\n\n[!] --packets-eq value\nIf the packet is matched an element in the set, match only if the packet counter of\nthe element matches the given value too.\n\n--packets-lt value\nIf the packet is matched an element in the set, match only if the packet counter of\nthe element is less than the given value as well.\n\n--packets-gt value\nIf the packet is matched an element in the set, match only if the packet counter of\nthe element is greater than the given value as well.\n\n[!] --bytes-eq value\nIf the packet is matched an element in the set, match only if the byte counter of the\nelement matches the given value too.\n\n--bytes-lt value\nIf the packet is matched an element in the set, match only if the byte counter of the\nelement is less than the given value as well.\n\n--bytes-gt value\nIf the packet is matched an element in the set, match only if the byte counter of the\nelement is greater than the given value as well.\n\nThe packet and byte counters related options and flags are ignored when the set was defined\nwithout counter support.\n\nThe option --match-set can be replaced by --set if that does not clash with an option of\nother extensions.\n\nUse of -m set requires that ipset kernel support is provided, which, for standard kernels, is\nthe case since Linux 2.6.39.\n" }, { "name": "socket", "content": "This matches if an open TCP/UDP socket can be found by doing a socket lookup on the packet.\nIt matches if there is an established or non-zero bound listening socket (possibly with a\nnon-local address). The lookup is performed using the packet tuple of TCP/UDP packets, or the\noriginal TCP/UDP header embedded in an ICMP/ICPMv6 error packet.\n" }, { "name": "--transparent", "content": "Ignore non-transparent sockets.\n", "long": "--transparent" }, { "name": "--nowildcard", "content": "Do not ignore sockets bound to 'any' address. The socket match won't accept\nzero-bound listeners by default, since then local services could intercept traffic\nthat would otherwise be forwarded. This option therefore has security implications\nwhen used to match traffic being forwarded to redirect such packets to local machine\nwith policy routing. When using the socket match to implement fully transparent prox‐\nies bound to non-local addresses it is recommended to use the --transparent option in‐\nstead.\n\nExample (assuming packets with mark 1 are delivered locally):\n\n-t mangle -A PREROUTING -m socket --transparent -j MARK --set-mark 1\n", "long": "--nowildcard" }, { "name": "--restore-skmark", "content": "Set the packet mark to the matching socket's mark. Can be combined with the --trans‐‐\nparent and --nowildcard options to restrict the sockets to be matched when restoring\nthe packet mark.\n\nExample: An application opens 2 transparent (IPTRANSPARENT) sockets and sets a mark on them\nwith SOMARK socket option. We can filter matching packets:\n\n-t mangle -I PREROUTING -m socket --transparent --restore-skmark -j action\n\n-t mangle -A action -m mark --mark 10 -j action2\n\n-t mangle -A action -m mark --mark 11 -j action3\n", "long": "--restore-skmark" }, { "name": "state", "content": "The \"state\" extension is a subset of the \"conntrack\" module. \"state\" allows access to the\nconnection tracking state for this packet.\n\n[!] --state state\nWhere state is a comma separated list of the connection states to match. Only a subset\nof the states unterstood by \"conntrack\" are recognized: INVALID, ESTABLISHED, NEW, RE‐‐\nLATED or UNTRACKED. For their description, see the \"conntrack\" heading in this man‐\npage.\n" }, { "name": "statistic", "content": "This module matches packets based on some statistic condition. It supports two distinct\nmodes settable with the --mode option.\n\nSupported options:\n\n--mode mode\nSet the matching mode of the matching rule, supported modes are random and nth.\n\n[!] --probability p\nSet the probability for a packet to be randomly matched. It only works with the random\nmode. p must be within 0.0 and 1.0. The supported granularity is in 1/2147483648th in‐\ncrements.\n\n[!] --every n\nMatch one packet every nth packet. It works only with the nth mode (see also the\n--packet option).\n\n--packet p\nSet the initial counter value (0 <= p <= n-1, default 0) for the nth mode.\n" }, { "name": "string", "content": "This module matches a given string by using some pattern matching strategy. It requires a\nlinux kernel >= 2.6.14.\n\n--algo {bm|kmp}\nSelect the pattern matching strategy. (bm = Boyer-Moore, kmp = Knuth-Pratt-Morris)\n\n--from offset\nSet the offset from which it starts looking for any matching. If not passed, default\nis 0.\n\n--to offset\nSet the offset up to which should be scanned. That is, byte offset-1 (counting from 0)\nis the last one that is scanned. If not passed, default is the packet size.\n\n[!] --string pattern\nMatches the given pattern.\n\n[!] --hex-string pattern\nMatches the given pattern in hex notation.\n" }, { "name": "--icase", "content": "Ignore case when searching.\n\nExamples:\n\n# The string pattern can be used for simple text characters.\niptables -A INPUT -p tcp --dport 80 -m string --algo bm --string 'GET /index.html' -j\nLOG\n\n# The hex string pattern can be used for non-printable characters, like |0D 0A| or\n|0D0A|.\niptables -p udp --dport 53 -m string --algo bm --from 40 --to 57 --hex-string\n'|03|www|09|netfilter|03|org|00|'\n", "long": "--icase" }, { "name": "tcp", "content": "These extensions can be used if `--protocol tcp' is specified. It provides the following op‐\ntions:\n\n[!] --source-port,--sport port[:port]\nSource port or port range specification. This can either be a service name or a port\nnumber. An inclusive range can also be specified, using the format first:last. If the\nfirst port is omitted, \"0\" is assumed; if the last is omitted, \"65535\" is assumed.\nThe flag --sport is a convenient alias for this option.\n\n[!] --destination-port,--dport port[:port]\nDestination port or port range specification. The flag --dport is a convenient alias\nfor this option.\n\n[!] --tcp-flags mask comp\nMatch when the TCP flags are as specified. The first argument mask is the flags which\nwe should examine, written as a comma-separated list, and the second argument comp is\na comma-separated list of flags which must be set. Flags are: SYN ACK FIN RST URG PSH\nALL NONE. Hence the command\niptables -A FORWARD -p tcp --tcp-flags SYN,ACK,FIN,RST SYN\nwill only match packets with the SYN flag set, and the ACK, FIN and RST flags unset.\n\n[!] --syn\nOnly match TCP packets with the SYN bit set and the ACK,RST and FIN bits cleared.\nSuch packets are used to request TCP connection initiation; for example, blocking such\npackets coming in an interface will prevent incoming TCP connections, but outgoing TCP\nconnections will be unaffected. It is equivalent to --tcp-flags SYN,RST,ACK,FIN SYN.\nIf the \"!\" flag precedes the \"--syn\", the sense of the option is inverted.\n\n[!] --tcp-option number\nMatch if TCP option set.\n" }, { "name": "tcpmss", "content": "This matches the TCP MSS (maximum segment size) field of the TCP header. You can only use\nthis on TCP SYN or SYN/ACK packets, since the MSS is only negotiated during the TCP handshake\nat connection startup time.\n\n[!] --mss value[:value]\nMatch a given TCP MSS value or range. If a range is given, the second value must be\ngreater than or equal to the first value.\n" }, { "name": "time", "content": "This matches if the packet arrival time/date is within a given range. All options are op‐\ntional, but are ANDed when specified. All times are interpreted as UTC by default.\n\n--datestart YYYY[-MM[-DD[Thh[:mm[:ss]]]]]\n\n--datestop YYYY[-MM[-DD[Thh[:mm[:ss]]]]]\nOnly match during the given time, which must be in ISO 8601 \"T\" notation. The possi‐\nble time range is 1970-01-01T00:00:00 to 2038-01-19T04:17:07.\n\nIf --datestart or --datestop are not specified, it will default to 1970-01-01 and\n2038-01-19, respectively.\n\n--timestart hh:mm[:ss]\n\n--timestop hh:mm[:ss]\nOnly match during the given daytime. The possible time range is 00:00:00 to 23:59:59.\nLeading zeroes are allowed (e.g. \"06:03\") and correctly interpreted as base-10.\n\n[!] --monthdays day[,day...]\nOnly match on the given days of the month. Possible values are 1 to 31. Note that\nspecifying 31 will of course not match on months which do not have a 31st day; the\nsame goes for 28- or 29-day February.\n\n[!] --weekdays day[,day...]\nOnly match on the given weekdays. Possible values are Mon, Tue, Wed, Thu, Fri, Sat,\nSun, or values from 1 to 7, respectively. You may also use two-character variants (Mo,\nTu, etc.).\n" }, { "name": "--contiguous", "content": "When --timestop is smaller than --timestart value, match this as a single time period\ninstead distinct intervals. See EXAMPLES.\n", "long": "--contiguous" }, { "name": "--kerneltz", "content": "Use the kernel timezone instead of UTC to determine whether a packet meets the time\nregulations.\n\nAbout kernel timezones: Linux keeps the system time in UTC, and always does so. On boot,\nsystem time is initialized from a referential time source. Where this time source has no\ntimezone information, such as the x86 CMOS RTC, UTC will be assumed. If the time source is\nhowever not in UTC, userspace should provide the correct system time and timezone to the ker‐\nnel once it has the information.\n\nLocal time is a feature on top of the (timezone independent) system time. Each process has\nits own idea of local time, specified via the TZ environment variable. The kernel also has\nits own timezone offset variable. The TZ userspace environment variable specifies how the\nUTC-based system time is displayed, e.g. when you run date(1), or what you see on your desk‐\ntop clock. The TZ string may resolve to different offsets at different dates, which is what\nenables the automatic time-jumping in userspace. when DST changes. The kernel's timezone off‐\nset variable is used when it has to convert between non-UTC sources, such as FAT filesystems,\nto UTC (since the latter is what the rest of the system uses).\n\nThe caveat with the kernel timezone is that Linux distributions may ignore to set the kernel\ntimezone, and instead only set the system time. Even if a particular distribution does set\nthe timezone at boot, it is usually does not keep the kernel timezone offset - which is what\nchanges on DST - up to date. ntpd will not touch the kernel timezone, so running it will not\nresolve the issue. As such, one may encounter a timezone that is always +0000, or one that is\nwrong half of the time of the year. As such, using --kerneltz is highly discouraged.\n\nEXAMPLES. To match on weekends, use:\n\n-m time --weekdays Sa,Su\n\nOr, to match (once) on a national holiday block:\n\n-m time --datestart 2007-12-24 --datestop 2007-12-27\n\nSince the stop time is actually inclusive, you would need the following stop time to not\nmatch the first second of the new day:\n\n-m time --datestart 2007-01-01T17:00 --datestop 2007-01-01T23:59:59\n\nDuring lunch hour:\n\n-m time --timestart 12:30 --timestop 13:30\n\nThe fourth Friday in the month:\n\n-m time --weekdays Fr --monthdays 22,23,24,25,26,27,28\n\n(Note that this exploits a certain mathematical property. It is not possible to say \"fourth\nThursday OR fourth Friday\" in one rule. It is possible with multiple rules, though.)\n\nMatching across days might not do what is expected. For instance,\n\n-m time --weekdays Mo --timestart 23:00 --timestop 01:00 Will match Monday, for one\nhour from midnight to 1 a.m., and then again for another hour from 23:00 onwards. If\nthis is unwanted, e.g. if you would like 'match for two hours from Montay 23:00 on‐\nwards' you need to also specify the --contiguous option in the example above.\n", "long": "--kerneltz" }, { "name": "tos", "content": "This module matches the 8-bit Type of Service field in the IPv4 header (i.e. including the\n\"Precedence\" bits) or the (also 8-bit) Priority field in the IPv6 header.\n\n[!] --tos value[/mask]\nMatches packets with the given TOS mark value. If a mask is specified, it is logically\nANDed with the TOS mark before the comparison.\n\n[!] --tos symbol\nYou can specify a symbolic name when using the tos match for IPv4. The list of recog‐\nnized TOS names can be obtained by calling iptables with -m tos -h. Note that this\nimplies a mask of 0x3F, i.e. all but the ECN bits.\n" }, { "name": "ttl (IPv4-specific)", "content": "This module matches the time to live field in the IP header.\n\n[!] --ttl-eq ttl\nMatches the given TTL value.\n\n--ttl-gt ttl\nMatches if TTL is greater than the given TTL value.\n\n--ttl-lt ttl\nMatches if TTL is less than the given TTL value.\n" }, { "name": "u32", "content": "U32 tests whether quantities of up to 4 bytes extracted from a packet have specified values.\nThe specification of what to extract is general enough to find data at given offsets from tcp\nheaders or payloads.\n\n[!] --u32 tests\nThe argument amounts to a program in a small language described below.\n\ntests := location \"=\" value | tests \"&&\" location \"=\" value\n\nvalue := range | value \",\" range\n\nrange := number | number \":\" number\n\na single number, n, is interpreted the same as n:n. n:m is interpreted as the range of num‐\nbers >=n and <=m.\n\nlocation := number | location operator number\n\noperator := \"&\" | \"<<\" | \">>\" | \"@\"\n\nThe operators &, <<, >> and && mean the same as in C. The = is really a set membership oper‐\nator and the value syntax describes a set. The @ operator is what allows moving to the next\nheader and is described further below.\n\nThere are currently some artificial implementation limits on the size of the tests:\n\n* no more than 10 of \"=\" (and 9 \"&&\"s) in the u32 argument\n\n* no more than 10 ranges (and 9 commas) per value\n\n* no more than 10 numbers (and 9 operators) per location\n\nTo describe the meaning of location, imagine the following machine that interprets it. There\nare three registers:\n\nA is of type char *, initially the address of the IP header\n\nB and C are unsigned 32 bit integers, initially zero\n\nThe instructions are:\n\nnumber B = number;\n\nC = (*(A+B)<<24) + (*(A+B+1)<<16) + (*(A+B+2)<<8) + *(A+B+3)\n" }, { "name": "&number", "content": "C = C & number\n" }, { "name": "<< number", "content": "C = C << number\n" }, { "name": ">> number", "content": "C = C >> number\n" }, { "name": "@number", "content": "A = A + C; then do the instruction number\n\nAny access of memory outside [skb->data,skb->end] causes the match to fail. Otherwise the\nresult of the computation is the final value of C.\n\nWhitespace is allowed but not required in the tests. However, the characters that do occur\nthere are likely to require shell quoting, so it is a good idea to enclose the arguments in\nquotes.\n\nExample:\n\nmatch IP packets with total length >= 256\n\nThe IP header contains a total length field in bytes 2-3.\n\n--u32 \"0 & 0xFFFF = 0x100:0xFFFF\"\n\nread bytes 0-3\n\nAND that with 0xFFFF (giving bytes 2-3), and test whether that is in the range\n[0x100:0xFFFF]\n\nExample: (more realistic, hence more complicated)\n\nmatch ICMP packets with icmp type 0\n\nFirst test that it is an ICMP packet, true iff byte 9 (protocol) = 1\n\n--u32 \"6 & 0xFF = 1 && ...\n\nread bytes 6-9, use & to throw away bytes 6-8 and compare the result to 1. Next test\nthat it is not a fragment. (If so, it might be part of such a packet but we cannot al‐\nways tell.) N.B.: This test is generally needed if you want to match anything beyond\nthe IP header. The last 6 bits of byte 6 and all of byte 7 are 0 iff this is a com‐\nplete packet (not a fragment). Alternatively, you can allow first fragments by only\ntesting the last 5 bits of byte 6.\n\n... 4 & 0x3FFF = 0 && ...\n\nLast test: the first byte past the IP header (the type) is 0. This is where we have to\nuse the @syntax. The length of the IP header (IHL) in 32 bit words is stored in the\nright half of byte 0 of the IP header itself.\n\n... 0 >> 22 & 0x3C @ 0 >> 24 = 0\"\n\nThe first 0 means read bytes 0-3, >>22 means shift that 22 bits to the right. Shifting\n24 bits would give the first byte, so only 22 bits is four times that plus a few more\nbits. &3C then eliminates the two extra bits on the right and the first four bits of\nthe first byte. For instance, if IHL=5, then the IP header is 20 (4 x 5) bytes long.\nIn this case, bytes 0-1 are (in binary) xxxx0101 yyzzzzzz, >>22 gives the 10 bit value\nxxxx0101yy and &3C gives 010100. @ means to use this number as a new offset into the\npacket, and read four bytes starting from there. This is the first 4 bytes of the ICMP\npayload, of which byte 0 is the ICMP type. Therefore, we simply shift the value 24 to\nthe right to throw out all but the first byte and compare the result with 0.\n\nExample:\n\nTCP payload bytes 8-12 is any of 1, 2, 5 or 8\n\nFirst we test that the packet is a tcp packet (similar to ICMP).\n\n--u32 \"6 & 0xFF = 6 && ...\n\nNext, test that it is not a fragment (same as above).\n\n... 0 >> 22 & 0x3C @ 12 >> 26 & 0x3C @ 8 = 1,2,5,8\"\n\n0>>22&3C as above computes the number of bytes in the IP header. @ makes this the new\noffset into the packet, which is the start of the TCP header. The length of the TCP\nheader (again in 32 bit words) is the left half of byte 12 of the TCP header. The\n12>>26&3C computes this length in bytes (similar to the IP header before). \"@\" makes\nthis the new offset, which is the start of the TCP payload. Finally, 8 reads bytes\n8-12 of the payload and = checks whether the result is any of 1, 2, 5 or 8.\n" }, { "name": "udp", "content": "These extensions can be used if `--protocol udp' is specified. It provides the following op‐\ntions:\n\n[!] --source-port,--sport port[:port]\nSource port or port range specification. See the description of the --source-port op‐\ntion of the TCP extension for details.\n\n[!] --destination-port,--dport port[:port]\nDestination port or port range specification. See the description of the --destina‐‐\ntion-port option of the TCP extension for details.\n" } ] }, "TARGET EXTENSIONS": { "content": "iptables can use extended target modules: the following are included in the standard distri‐\nbution.\n\nAUDIT\nThis target creates audit records for packets hitting the target. It can be used to record\naccepted, dropped, and rejected packets. See auditd(8) for additional details.\n\n--type {accept|drop|reject}\nSet type of audit record. Starting with linux-4.12, this option has no effect on gen‐\nerated audit messages anymore. It is still accepted by iptables for compatibility rea‐\nsons, but ignored.\n\nExample:\n\niptables -N AUDITDROP\n\niptables -A AUDITDROP -j AUDIT\n\niptables -A AUDITDROP -j DROP\n\nCHECKSUM\nThis target selectively works around broken/old applications. It can only be used in the\nmangle table.\n", "subsections": [ { "name": "--checksum-fill", "content": "Compute and fill in the checksum in a packet that lacks a checksum. This is particu‐\nlarly useful, if you need to work around old applications such as dhcp clients, that\ndo not work well with checksum offloads, but don't want to disable checksum offload in\nyour device.\n\nCLASSIFY\nThis module allows you to set the skb->priority value (and thus classify the packet into a\nspecific CBQ class).\n\n--set-class major:minor\nSet the major and minor class value. The values are always interpreted as hexadecimal\neven if no 0x prefix is given.\n", "long": "--checksum-fill" }, { "name": "CLUSTERIP (IPv4-specific)", "content": "This module allows you to configure a simple cluster of nodes that share a certain IP and MAC\naddress without an explicit load balancer in front of them. Connections are statically dis‐\ntributed between the nodes in this cluster.\n\nPlease note that CLUSTERIP target is considered deprecated in favour of cluster match which\nis more flexible and not limited to IPv4.\n\n--new Create a new ClusterIP. You always have to set this on the first rule for a given\nClusterIP.\n\n--hashmode mode\nSpecify the hashing mode. Has to be one of sourceip, sourceip-sourceport, sour‐‐\nceip-sourceport-destport.\n\n--clustermac mac\nSpecify the ClusterIP MAC address. Has to be a link-layer multicast address\n\n--total-nodes num\nNumber of total nodes within this cluster.\n\n--local-node num\nLocal node number within this cluster.\n\n--hash-init rnd\nSpecify the random seed used for hash initialization.\n\nCONNMARK\nThis module sets the netfilter mark value associated with a connection. The mark is 32 bits\nwide.\n\n--set-xmark value[/mask]\nZero out the bits given by mask and XOR value into the ctmark.\n\n--save-mark [--nfmask nfmask] [--ctmask ctmask]\nCopy the packet mark (nfmark) to the connection mark (ctmark) using the given masks.\nThe new nfmark value is determined as follows:\n\nctmark = (ctmark & ~ctmask) ^ (nfmark & nfmask)\n\ni.e. ctmask defines what bits to clear and nfmask what bits of the nfmark to XOR into\nthe ctmark. ctmask and nfmask default to 0xFFFFFFFF.\n\n--restore-mark [--nfmask nfmask] [--ctmask ctmask]\nCopy the connection mark (ctmark) to the packet mark (nfmark) using the given masks.\nThe new ctmark value is determined as follows:\n\nnfmark = (nfmark & ~nfmask) ^ (ctmark & ctmask);\n\ni.e. nfmask defines what bits to clear and ctmask what bits of the ctmark to XOR into\nthe nfmark. ctmask and nfmask default to 0xFFFFFFFF.\n\n--restore-mark is only valid in the mangle table.\n\nThe following mnemonics are available for --set-xmark:\n\n--and-mark bits\nBinary AND the ctmark with bits. (Mnemonic for --set-xmark 0/invbits, where invbits is\nthe binary negation of bits.)\n\n--or-mark bits\nBinary OR the ctmark with bits. (Mnemonic for --set-xmark bits/bits.)\n\n--xor-mark bits\nBinary XOR the ctmark with bits. (Mnemonic for --set-xmark bits/0.)\n\n--set-mark value[/mask]\nSet the connection mark. If a mask is specified then only those bits set in the mask\nare modified.\n\n--save-mark [--mask mask]\nCopy the nfmark to the ctmark. If a mask is specified, only those bits are copied.\n\n--restore-mark [--mask mask]\nCopy the ctmark to the nfmark. If a mask is specified, only those bits are copied.\nThis is only valid in the mangle table.\n\nCONNSECMARK\nThis module copies security markings from packets to connections (if unlabeled), and from\nconnections back to packets (also only if unlabeled). Typically used in conjunction with\nSECMARK, it is valid in the security table (for backwards compatibility with older kernels,\nit is also valid in the mangle table).\n\n--save If the packet has a security marking, copy it to the connection if the connection is\nnot marked.\n" }, { "name": "--restore", "content": "If the packet does not have a security marking, and the connection does, copy the se‐\ncurity marking from the connection to the packet.\n\n\nCT\nThe CT target sets parameters for a packet or its associated connection. The target attaches\na \"template\" connection tracking entry to the packet, which is then used by the conntrack\ncore when initializing a new ct entry. This target is thus only valid in the \"raw\" table.\n", "long": "--restore" }, { "name": "--notrack", "content": "Disables connection tracking for this packet.\n\n--helper name\nUse the helper identified by name for the connection. This is more flexible than load‐\ning the conntrack helper modules with preset ports.\n\n--ctevents event[,...]\nOnly generate the specified conntrack events for this connection. Possible event types\nare: new, related, destroy, reply, assured, protoinfo, helper, mark (this refers to\nthe ctmark, not nfmark), natseqinfo, secmark (ctsecmark).\n\n--expevents event[,...]\nOnly generate the specified expectation events for this connection. Possible event\ntypes are: new.\n\n--zone-orig {id|mark}\nFor traffic coming from ORIGINAL direction, assign this packet to zone id and only\nhave lookups done in that zone. If mark is used instead of id, the zone is derived\nfrom the packet nfmark.\n\n--zone-reply {id|mark}\nFor traffic coming from REPLY direction, assign this packet to zone id and only have\nlookups done in that zone. If mark is used instead of id, the zone is derived from the\npacket nfmark.\n\n--zone {id|mark}\nAssign this packet to zone id and only have lookups done in that zone. If mark is\nused instead of id, the zone is derived from the packet nfmark. By default, packets\nhave zone 0. This option applies to both directions.\n\n--timeout name\nUse the timeout policy identified by name for the connection. This is provides more\nflexible timeout policy definition than global timeout values available at\n/proc/sys/net/netfilter/nfconntrack*timeout*.\n\nDNAT\nThis target is only valid in the nat table, in the PREROUTING and OUTPUT chains, and user-de‐\nfined chains which are only called from those chains. It specifies that the destination ad‐\ndress of the packet should be modified (and all future packets in this connection will also\nbe mangled), and rules should cease being examined. It takes the following options:\n\n--to-destination [ipaddr[-ipaddr]][:port[-port]]\nwhich can specify a single new destination IP address, an inclusive range of IP ad‐\ndresses. Optionally a port range, if the rule also specifies one of the following pro‐\ntocols: tcp, udp, dccp or sctp. If no port range is specified, then the destination\nport will never be modified. If no IP address is specified then only the destination\nport will be modified. In Kernels up to 2.6.10 you can add several --to-destination\noptions. For those kernels, if you specify more than one destination address, either\nvia an address range or multiple --to-destination options, a simple round-robin (one\nafter another in cycle) load balancing takes place between these addresses. Later\nKernels (>= 2.6.11-rc1) don't have the ability to NAT to multiple ranges anymore.\n", "long": "--notrack" }, { "name": "--random", "content": "If option --random is used then port mapping will be randomized (kernel >= 2.6.22).\n", "long": "--random" }, { "name": "--persistent", "content": "Gives a client the same source-/destination-address for each connection. This super‐\nsedes the SAME target. Support for persistent mappings is available from 2.6.29-rc2.\n\nIPv6 support available since Linux kernels >= 3.7.\n", "long": "--persistent" }, { "name": "DNPT (IPv6-specific)", "content": "Provides stateless destination IPv6-to-IPv6 Network Prefix Translation (as described by RFC\n6296).\n\nYou have to use this target in the mangle table, not in the nat table. It takes the following\noptions:\n\n--src-pfx [prefix/length]\nSet source prefix that you want to translate and length\n\n--dst-pfx [prefix/length]\nSet destination prefix that you want to use in the translation and length\n\nYou have to use the SNPT target to undo the translation. Example:\n\nip6tables -t mangle -I POSTROUTING -s fd00::/64 -o vboxnet0 -j SNPT --src-pfx\nfd00::/64 --dst-pfx 2001:e20:2000:40f::/64\n\nip6tables -t mangle -I PREROUTING -i wlan0 -d 2001:e20:2000:40f::/64 -j DNPT --src-pfx\n2001:e20:2000:40f::/64 --dst-pfx fd00::/64\n\nYou may need to enable IPv6 neighbor proxy:\n\nsysctl -w net.ipv6.conf.all.proxyndp=1\n\nYou also have to use the NOTRACK target to disable connection tracking for translated flows.\n\nDSCP\nThis target alters the value of the DSCP bits within the TOS header of the IPv4 packet. As\nthis manipulates a packet, it can only be used in the mangle table.\n\n--set-dscp value\nSet the DSCP field to a numerical value (can be decimal or hex)\n\n--set-dscp-class class\nSet the DSCP field to a DiffServ class.\n" }, { "name": "ECN (IPv4-specific)", "content": "This target selectively works around known ECN blackholes. It can only be used in the mangle\ntable.\n" }, { "name": "--ecn-tcp-remove", "content": "Remove all ECN bits from the TCP header. Of course, it can only be used in conjunc‐\ntion with -p tcp.\n", "long": "--ecn-tcp-remove" }, { "name": "HL (IPv6-specific)", "content": "This is used to modify the Hop Limit field in IPv6 header. The Hop Limit field is similar to\nwhat is known as TTL value in IPv4. Setting or incrementing the Hop Limit field can poten‐\ntially be very dangerous, so it should be avoided at any cost. This target is only valid in\nmangle table.\n" }, { "name": "Don't ever set or increment the value on packets that leave your local network!", "content": "--hl-set value\nSet the Hop Limit to `value'.\n\n--hl-dec value\nDecrement the Hop Limit `value' times.\n\n--hl-inc value\nIncrement the Hop Limit `value' times.\n\nHMARK\nLike MARK, i.e. set the fwmark, but the mark is calculated from hashing packet selector at\nchoice. You have also to specify the mark range and, optionally, the offset to start from.\nICMP error messages are inspected and used to calculate the hashing.\n\nExisting options are:\n\n--hmark-tuple tuple\nPossible tuple members are: src meaning source address (IPv4, IPv6 address), dst mean‐\ning destination address (IPv4, IPv6 address), sport meaning source port (TCP, UDP, UD‐\nPlite, SCTP, DCCP), dport meaning destination port (TCP, UDP, UDPlite, SCTP, DCCP),\nspi meaning Security Parameter Index (AH, ESP), and ct meaning the usage of the con‐\nntrack tuple instead of the packet selectors.\n\n--hmark-mod value (must be > 0)\nModulus for hash calculation (to limit the range of possible marks)\n\n--hmark-offset value\nOffset to start marks from.\n\nFor advanced usage, instead of using --hmark-tuple, you can specify custom\nprefixes and masks:\n\n--hmark-src-prefix cidr\nThe source address mask in CIDR notation.\n\n--hmark-dst-prefix cidr\nThe destination address mask in CIDR notation.\n\n--hmark-sport-mask value\nA 16 bit source port mask in hexadecimal.\n\n--hmark-dport-mask value\nA 16 bit destination port mask in hexadecimal.\n\n--hmark-spi-mask value\nA 32 bit field with spi mask.\n\n--hmark-proto-mask value\nAn 8 bit field with layer 4 protocol number.\n\n--hmark-rnd value\nA 32 bit random custom value to feed hash calculation.\n\nExamples:\n\niptables -t mangle -A PREROUTING -m conntrack --ctstate NEW\n-j HMARK --hmark-tuple ct,src,dst,proto --hmark-offset 10000 --hmark-mod 10 --hmark-rnd\n0xfeedcafe\n\niptables -t mangle -A PREROUTING -j HMARK --hmark-offset 10000 --hmark-tuple src,dst,proto\n--hmark-mod 10 --hmark-rnd 0xdeafbeef\n\nIDLETIMER\nThis target can be used to identify when interfaces have been idle for a certain period of\ntime. Timers are identified by labels and are created when a rule is set with a new label.\nThe rules also take a timeout value (in seconds) as an option. If more than one rule uses\nthe same timer label, the timer will be restarted whenever any of the rules get a hit. One\nentry for each timer is created in sysfs. This attribute contains the timer remaining for\nthe timer to expire. The attributes are located under the xtidletimer class:\n\n/sys/class/xtidletimer/timers/