It is a well known fact that VPN links have smaller packet size due to incapsulation overhead. A large packet with MSS that exceeds the MSS of the VPN link should be fragmented prior to sending it via that kind of connection. However, if the packet has DF flag set, it cannot be fragmented and should be discarded. On links that have broken path MTU discovery (PMTUD) it may lead to a number of problems, including problems with FTP and HTTP data transfer and e-mail services.
It is a well known fact that VPN links have smaller packet size due to encapsulation overhead. A large packet with MSS that exceeds the MSS of the VPN link should be fragmented prior to sending it via that kind of connection. However, if the packet has DF flag set, it cannot be fragmented and should be discarded. On links that have broken path MTU discovery (PMTUD) it may lead to a number of problems, including problems with FTP and HTTP data transfer and e-mail services.
Mangle is a kind of 'marker' that marks packets for future processing with special marks. Many other facilities in RouterOS make use of these marks, e.g. queue trees, NAT, routing. They identify a packet based on its mark and process it accordingly. The mangle marks exist only within the router, they are not transmitted across the network.
Additionally, the mangle facility is used to modify some fields in the IP header, like TOS (DSCP) and TTL fields.
Properties
Property
Description
action (action name; Default: accept)
Action to take if packet is matched by the rule:
accept - accept the packet. Packet is not passed to next firewall rule.
add-dst-to-address-list - add destination address to Address list specified by address-list parameter
add-src-to-address-list - add source address to Address list specified by address-list parameter
change-dscp - change Differentiated Services Code Point (DSCP) field value specified by the new-dscp parameter
change-mss - change Maximum Segment Size field value of the packet to a value specified by the new-mss parameter
change-ttl - change Time to Live field value of the packet to a value specified by the new-ttl parameter
clear-df - clear 'Do Not Fragment' Flag
jump - jump to the user defined chain specified by the value of jump-target parameter
log - add a message to the system log containing following data: in-interface, out-interface, src-mac, protocol, src-ip:port->dst-ip:port and length of the packet. After packet is matched it is passed to next rule in the list, similar as passthrough
mark-connection - place a mark specified by the new-connection-mark parameter on the entire connection that matches the rule
mark-packet - place a mark specified by the new-packet-mark parameter on a packet that matches the rule
mark-routing - place a mark specified by the new-routing-mark parameter on a packet. This kind of marks is used for policy routing purposes only
passthrough - if packet is matched by the rule, increase counter and go to next rule (useful for statistics).
return - pass control back to the chain from where the jump took place
route - forces packets to a specific gateway IP by ignoring normal routing decision (prerouting chain only)
set-priority - set priority specified by the new-priority parameter on the packets sent out through a link that is capable of transporting priority (VLAN or WMM-enabled wireless interface). Read more>
sniff-pc
sniff-tzsp - send packet to a remote TZSP compatible system (such as Wireshark). Set remote target with sniff-target and sniff-target-port parameters (Wireshark recommends port 37008)
strip-ipv4-options - strip IPv4 option fields from IP header.
address-list (string; Default: )
Name of the address list to be used. Applicable if action is add-dst-to-address-list or add-src-to-address-list
address-list-timeout (time; Default: 00:00:00)
Time interval after which the address will be removed from the address list specified by address-list parameter. Used in conjunction with add-dst-to-address-list or add-src-to-address-list actions
Value of 00:00:00 will leave the address in the address list forever
chain (name; Default: )
Specifies to which chain the rule will be added. If the input does not match the name of an already defined chain, a new chain will be created.
comment (string; Default: )
Descriptive comment for the rule.
connection-bytes (integer-integer; Default: )
Matches packets only if a given amount of bytes has been transfered through the particular connection. 0 - means infinity, for example connection-bytes=2000000-0 means that the rule matches if more than 2MB has been transfered through the relevant connection
connection-limit (integer,netmask; Default: )
Restrict connection limit per address or address block
connection-mark (no-mark | string; Default: )
Matches packets marked via mangle facility with particular connection mark. If no-mark is set, rule will match any unmarked connection.
connection-nat-state (srcnat | dstnat; Default: )
Can match connections that are srcnatted, dstnatted or both. Note that connection-state=related connections connection-nat-state is determined by direction of the first packet. and if connection tracking needs to use dst-nat to deliver this connection to same hosts as main connection it will be in connection-nat-state=dstnat even if there are no dst-nat rules at all.
Connection Rate is a firewall matcher that allows the capture of traffic based on the present speed of the connection. Read more >>
connection-state (estabilished | invalid | new | related; Default: )
Interprets the connection tracking analysis data for a particular packet:
established - a packet which belongs to an existing connection
invalid - a packet that does not have determined state in connection tracking (ussualy - sevear out-of-order packets, packets with wrong sequence/ack number, or in case of resource overusage on router), for this reason invalid packet will not participate in NAT (as only connection-state=new packets do), and will still contain original source IP address when routed. We strongly suggest to drop all connection-state=invalid packets in firewall filter forward and input chains
new - the packet has started a new connection, or otherwise
associated with a connection which has not seen packets in both directions
related - a packet which is related to, but not part of an existing connection, such as ICMP errors or a packet which begins FTP data connection
Matches packets from related connections based on information from their connection tracking helpers. A relevant connection helper must be enabled under /ip firewall service-port
content (string; Default: )
Match packets that contain specified text
dscp (integer: 0..63; Default: )
Matches DSCP IP header field.
dst-address (IP/netmask | IP range; Default: )
Matches packets where destination is equal to specified IP or falls into specified IP range.
dst-address-list (name; Default: )
Matches destination address of a packet against user-defined address list
dst-address-type (unicast | local | broadcast | multicast; Default: )
Matches destination address type:
unicast - IP address used for point to point transmission
local - if dst-address is assigned to one of router's interfaces
broadcast - packet is sent to all devices in subnet
multicast - packet is forwarded to defined group of devices
Matches packets until a given pps limit is exceeded. As opposed to the limit matcher, every destination IP address / destination port has it's own limit. Parameters are written in following format: count[/time],burst,mode[/expire].
count - maximum average packet rate measured in packets per time interval
time - specifies the time interval in which the packet rate is measured (optional)
burst - number of packets which are not counted by packet rate
mode - the classifier for packet rate limiting
expire - specifies interval after which recored ip address /port will be deleted (optional)
dst-port (integer[-integer]: 0..65535; Default: )
List of destination port numbers or port number ranges
fragment (yes|no; Default: )
Matches fragmented packets. First (starting) fragment does not count. If connection tracking is enabled there will be no fragments as system automatically assembles every packet
Actual interface the packet has entered the router, if incoming interface is bridge
in-interface (name; Default: )
Interface the packet has entered the router
ingress-priority (integer: 0..63; Default: )
Matches ingress priority of the packet. Priority may be derived from VLAN, WMM or MPLS EXP bit. Read more >>
ipsec-policy (in | out, ipsec | none; Default: )
Matches the policy used by IpSec. Value is written in following format: direction, policy. Direction is Used to select whether to match the policy used for decapsulation or the policy that will be used for encapsulation.
in - valid in the PREROUTING, INPUT and FORWARD chains
out - valid in the POSTROUTING, OUTPUT and FORWARD chains
ipsec - matches if the packet is subject to IpSec processing;
none - matches packet that is not subject to IpSec processing (for example, IpSec transport packet).
For example, if router receives Ipsec encapsulated Gre packet, then rule ipsec-policy=in,ipsec will match Gre packet, but rule ipsec-policy=in,none will match ESP packet.
any - match packet with at least one of the ipv4 options
loose-source-routing - match packets with loose source routing option. This option is used to route the internet datagram based on information supplied by the source
no-record-route - match packets with no record route option. This option is used to route the internet datagram based on information supplied by the source
no-router-alert - match packets with no router alter option
no-source-routing - match packets with no source routing option
no-timestamp - match packets with no timestamp option
record-route - match packets with record route option
router-alert - match packets with router alter option
strict-source-routing - match packets with strict source routing option
timestamp - match packets with timestamp
jump-target (name; Default: )
Name of the target chain to jump to. Applicable only if action=jump
Matches packets of specified size or size range in bytes.
passthrough (yes|no; Default: )
whether to let the packet to pass further (like action passthrough) after marking it with a given mark (property only valid if action is mark packet, connection or routing mark).
PCC matcher allows division of traffic into equal streams with ability to keep packets with specific set of options in one particular stream. Read more >>
port (integer[-integer]: 0..65535; Default: )
Matches if any (source or destination) port matches the specified list of ports or port ranges. Applicable only if protocol is TCP or UDP
protocol (name or protocol ID; Default: tcp)
Matches particular IP protocol specified by protocol name or number
psd (integer,time,integer,integer; Default: )
Attempts to detect TCP and UDP scans. Parameters are in following format WeightThreshold, DelayThreshold, LopPortWeight, HighPortWeight
WeightThreshold - total weight of the latest TCP/UDP packets with different destination ports coming from the same host to be treated as port scan sequence
DelayThreshold - delay for the packets with different destination ports coming from the same host to be treated as possible port scan subsequence
LowPortWeight - weight of the packets with privileged (<=1024) destination port
HighPortWeight - weight of the packet with non-priviliged destination port
random (integer: 1..99; Default: )
Matches packets randomly with given probability.
routing-mark (string; Default: )
Matches packets marked by mangle facility with particular routing mark
src-address (IP/Netmask, IP range; Default: )
Matches packets where source is equal to specified IP or falls into specified IP range.
src-address-list (name; Default: )
Matches source address of a packet against user-defined address list
src-address-type (unicast | local | broadcast | multicast; Default: )
Matches source address type:
unicast - IP address used for point to point transmission
local - if address is assigned to one of router's interfaces
broadcast - packet is sent to all devices in subnet
multicast - packet is forwarded to defined group of devices
src-port (integer[-integer]: 0..65535; Default: )
List of source ports and ranges of source ports. Applicable only if protocol is TCP or UDP.
src-mac-address (MAC address; Default: )
Matches source MAC address of the packet
tcp-flags (ack | cwr | ece | fin | psh | rst | syn | urg; Default: )
Matches specified TCP flags
ack - acknowledging data
cwr - congestion window reduced
ece - ECN-echo flag (explicit congestion notification)
It is a well known fact that VPN links have smaller packet size due to encapsulation overhead. A large packet with MSS that exceeds the MSS of the VPN link should be fragmented prior to sending it via that kind of connection. However, if the packet has DF flag set, it cannot be fragmented and should be discarded. On links that have broken path MTU discovery (PMTUD) it may lead to a number of problems, including problems with FTP and HTTP data transfer and e-mail services.
In case of link with broken PMTUD, a decrease of the MSS of the packets coming through the VPN link solves the problem. The following example demonstrates how to decrease the MSS value via mangle:
Marking each packet is quite resource expensive especially if rule has to match against many parameters from IP header or address list containing hundreds of entries.
Lets say we want to
mark all tcp packets except tcp/80 and match these packets against first address list
mark all udp packets and match them against second address list.
Setup looks quite simple and probably will work without problems in small networks. Now multiply count of rules by 10, add few hundred entries in address list, run 100Mbit of traffic over this router and you will see how rapidly CPU usage is increasing. The reason for such behavior is that each rule reads IP header of every packet and tries to match collected data against parameters specified in firewall rule.
Fortunately if connection tracking is enabled, we can use connection marks to optimize our setup.
Now first rule will try to match data from IP header only from first packet of new connection and add connection mark. Next rule will no longer check IP header for each packet, it will just compare connection marks resulting in lower CPU consumption. Additionally passthrough=no was added that helps to reduce CPU consumption even more.
