Skip to content

Latest commit

 

History

History
791 lines (456 loc) · 17.7 KB

conf_options.md

File metadata and controls

791 lines (456 loc) · 17.7 KB

Configuring TheRouter

There are three groups of configuration options and commands:

  • command line options
  • configuration file commands
  • rcli commands

Command line options

Command line options can be modified by editing the run script /usr/local/sbin/router_run.sh. Most of them are DPDK EAL command line options, therefore you can refer to DPDK documentation EAL Command-line Options for detailed description of them.

Before running TheRouter you must check the following options and use your own values depending on the hardware you use:

  • -c

Set the hexadecimal bitmask of the cores to run on.

  • --lcores

! Note: the_router.a0.01.tar.gz version has supported only a 4 cores cpu. You must use at least a 4 cores cpu and then configure TheRouter to use 4 lcores using --lcores='0@0,1@1,2@2,3@3' parameter

Map lcore set to physical cpu set

  • -n

Set the number of memory channels to use.

  • -w

Add a PCI device in white list.

Example of the startup script cmd options:

the_router --proc-type=primary -c 0xF --lcores='0@0,1@1,2@2,3@3' --syslog='daemon' -n2 -w 0000:01:00.0 -w 0000:01:00.1 -- -c $1 -d

Note: Lcore 0 will be used for TheRouter's control plane task and can be shared with any linux tasks. The other cores will be used in TheRouter's data plane process. You should isolate them during the linux starup process by using linux kernel command line parameters isolcpus. Otherwise, performance of TheRouter's working threads could be very low due the context switching.

Configuration file options

This options are stored in the /etc/router.conf file.

Configuration file consists of the sections:

  • startup
  • runtime

Each section contains commands. Everything on the same line is considered as a single command. Symbol # is used to comment a whole line.

startup {
	startup_command_1
	startup_command_2
	...
	startup_command_n
}


runtime {
	runtime_command_1
	runtime_command_2
	...
	runtime_command_n
}

Configuration file example

startup {
  # total number of mbufs
  sysctl set mbuf 8192

  port 0 mtu 1500 tpid 0x8100 state enabled
  port 1 mtu 1500 tpid 0x8100 state enabled

  rx_queue port 0 queue 0 lcore 1
  rx_queue port 0 queue 1 lcore 2
  rx_queue port 0 queue 2 lcore 3

  rx_queue port 1 queue 0 lcore 3
  rx_queue port 1 queue 1 lcore 2
  rx_queue port 1 queue 2 lcore 1

  sysctl set global_packet_counters 1

#  sysctl set arp_cache_timeout 300
}

runtime {
  vif add name p0 port 1 type untagged
  ip addr add 10.0.0.1/24 dev p0

  vif add name p1 port 0 type untagged
  ip addr add 10.0.1.1/24 dev p1

  ip route add 0.0.0.0/0 via 10.0.1.2 src 10.0.1.1

  npf load "/etc/npf.conf"
}

NPF configaration file example

group default {
  pass final on p0 all
  pass final on p1 all
}

Startup commands

Startup commands are the commands that are used to initilize TheRouter's susbystem and properties that can't be modified once TheRouter have started. This commands can only be used in the startup section of the configuration file and can't be used by the rcli interface.

  • port

     port <dpdk_port_number> mtu <mtu_size> tpid <tpid_value> state enabled

  • rx_queue

     rx_queue port <dpdk_port_number> queue <queue_number> lcore <lcore>

  • sysctl

     sysctl set <name> <value>

  • npf load

     npf load "<path_to_npf_configuration_file>"

Note: that you should enclose path to file with '"'.

Runtime commands

Runtime commands are the commands that can be either executed via rcli interface or used in the runtime section of the configuration file.

ip addr

  • ip addr add

     rcli ip addr add <net>/<mask> dev <vif_name>

  • ip addr del

     rcli ip addr del <net>/<mask> dev <vif_name>

  • sh ip addr

     rcli sh ip addr

ip route tables

  • ip route table add

     rcli ip route table add <route_table_name>

  • ip route table del

     rcli ip route table del <route_table_name>

  • rcli sh ip route tables

     rcli rcli sh ip route tables

U32 sets

  • u32set create

     rcli u32set create <u32set_name> size <size> bucket_size <bucket_size>

  • u32set destroy

     rcli u32set destroy <u32set_name>

  • ipset add

     rcli ipset add <u32set_name> <ipv4>

  • ipset del

     rcli ipset del <u32set_name> <ipv4>

  • ipset test

     rcli ipset test <u32set_name> <ipv4>

  • l2set add

     rcli l2set add <u32set_name> port <port_number> svid <svid> cvid <cvid>

  • l2set del

     rcli l2set del <u32set_name> port <port_number> svid <svid> cvid <cvid>

  • l2set test

     rcli l2set test <u32set_name> port <port_number> svid <svid> cvid <cvid>

PBR rules

  • sh ip pbr rules

     rcli sh ip pbr rules

  • ip pbr rule add

     ip pbr rule add prio <prio_num> u32set <u32set_name> type "ip" table <route_table_name>

 ip pbr rule add prio <prio_num> u32set <u32set_name> type "l2" table <route_table_name>
 
 rcli ip pbr rule add prio <prio_num> from <net/mask> <route_table_name>

  • ip pbr rule del

     rcli ip pbr rule del prio <prio_num>

  • ip pbr flush

     rcli ip pbr flush

ip route

  • ip route add

     rcli ip route add <net>/<mask> dev <vif_name> src <src_ip> [table <table_name>]

    or

     ip route add <net>/<mask> via <gw_ip> src <src_ip> [table <table_name>]

    or

     ip route add <net>/<mask> unreachable [table <table_name>]

  • ip route del

     ip route del <net/mask> [table <table_name>]

  • sh ip route

     rcli sh ip route

vif

  • vif add

     rcli vif add name <name> port <port_num> type <type> [svid <svid>] [cvid <cvid>] [flags <flag1,flag2...>]

    Type parameter can take one of the following values:

    • untagged
    • dot1q
    • qinq

    Flags:

    • kni
    • proxy_arp

  • vif del

     rcli vif del <name>

  • sh vif

     rcli sh vif

  • sh vif counters

     rcli sh vif counters

  • clear vif counters

     rcli clear vif counters

arp

  • arp add

     rcli arp add <ip> <mac> dev <vif_name> [static]

  • arp del

     rcli arp del <ip> dev <vif_name>

  • sh arp cache

     rcli sh arp cache

sysctl

  • sysctl set

     rcli sysctl set <name> <value>

  • sysctl set

     rcli sysctl get <name> <value>

ping

  • ping

     rcli ping --help
 Usage: ping [-c,--count count] [-i,--interval interval_in_ms] [-s icmp_payload_size]
 [-f,--dont_frag] [-a,--source_address ip_source_address] [-w,--nowait]
 [-h,--help] destination

NPF

  • sh npf conndb size

     rcli sh npf conndb size

  • sh npf conndb summary

     rcli sh npf conndb summary

  • sh npf stat

     rcli sh npf stat

  • npf clear stat

     rcli npf clear stat

NPF sysctl variables controlling connection tracking state timeouts

* NPF_TCPS_CLOSED
* NPF_TCPS_SYN_SENT
* NPF_TCPS_SIMSYN_SENT
* NPF_TCPS_SYN_RECEIVED
* NPF_TCPS_ESTABLISHED
* NPF_TCPS_FIN_SENT
* NPF_TCPS_FIN_RECEIVED
* NPF_TCPS_CLOSE_WAIT
* NPF_TCPS_FIN_WAIT
* NPF_TCPS_CLOSING
* NPF_TCPS_LAST_ACK
* NPF_TCPS_TIME_WAIT
* NPF_ANY_CONN_CLOSED
* NPF_ANY_CONN_NEW
* NPF_ANY_CONN_ESTABLISHED

Example:

startup {
	...

	# any protocol timeouts (UDP)
	sysctl set NPF_ANY_CONN_CLOSED 2
	sysctl set NPF_ANY_CONN_NEW 30
	sysctl set NPF_ANY_CONN_ESTABLISHED 60
	
	# TCP timeouts
	sysctl set NPF_TCPS_CLOSED 10
	sysctl set NPF_TCPS_SYN_SENT 30
	sysctl set NPF_TCPS_SIMSYN_SENT 30
	sysctl set NPF_TCPS_SYN_RECEIVED 60
	sysctl set NPF_TCPS_ESTABLISHED 600
	sysctl set NPF_TCPS_FIN_SENT 240
	sysctl set NPF_TCPS_FIN_RECEIVED 240
	sysctl set NPF_TCPS_CLOSE_WAIT 45
	sysctl set NPF_TCPS_FIN_WAIT 60
	sysctl set NPF_TCPS_CLOSING 30
	sysctl set NPF_TCPS_LAST_ACK 30
	sysctl set NPF_TCPS_TIME_WAIT 120

	...
}

NAT events logging via IPFIX

  • Enable nat events

     sysctl set ipfix_nat_events 1

  • Setup ipfix collector

     ipfix_collector addr 192.168.20.2

Other commands

  • shutdown

     rcli shutdown

Router statistic commands

  • sh port ext stat

     rcli sh port ext stat

  • sh port stat

     rcli sh port stat

  • sh cmbuf stats

     rcli sh cmbuf stats

  • sh mbuf stats

     rcli sh mbuf stats

  • sh stats

     rcli sh stats

  • clear stats

     rcli clear stats

IPv6

  • ipv6 enable

     ipv6 enable dev <vif_name>

Enables IPv6 protocol on an interface, create link-local address using eui-64 scheme

  • ipv6 disable

     ipv6 disable dev <vif_name>

Disables IPv6 protocol on the interface. Delete all addresses and routes depending on an interface

  • sh ipv6 addr

      sh ipv6 addr

    Outputs ipv6 addresses assignes to interfaces

  • ipv6 addr add eui-64

     ipv6 addr add eui-64 <prefix>/<length> eui-64 dev <vif_name>

Generates an address using prefix and interface id (EUI64 scheme) and assign the address to an interface

  • ipv6 addr add

     ipv6 addr add <address>/<length> dev <vif_name>

Assigns the given ipv6 address to an interface

  • ipv6 addr link-local

     ipv6 addr link-local <address> dev <vif_name>

Assigns the given ipv6 link-local address to an interface

  • ipv6 addr link-local eui-64

     ipv6 addr link-local eui-64 dev <vif_name>

Generates a link-local address using the eui-64 scheme and assigns it to an interface

  • ipv6 addr auto

     ipv6 addr auto dev <vif_name> enable|disable

Enables SLAAC client for an interface. Once enabled the router starts assigning dynamic ipv6 addresses at an interface based on the information received in Router Advertisement messages

  • ipv6 addr del

     ipv6 addr del <address or prefix>/<length> dev <vif_name>

Deletes an address from an interface

  • sh ipv6 route

     sh ipv6 route

Outputs ipv6 routing table

  • ipv6 route add

     ipv6 route add <prefix/prefix-length> dev <vif_name>

Adds a connected route to a prefix into the main routing table

  • ipv6 route add

     ipv6 route add <prefix/length> dev <vif_name> via <ipv6-address> [table <ipv6_routing_table>]

Adds a route to a prefix via a gateway with given address into a routing table

  • ipv6 route add

     ipv6 route add ::/0 via <ipv6-address> [table <ipv6_routing_table>]

Adds the default route into a routing table

  • ipv6 route add

     ipv6 route add <prefix/length> unreachable [table <ipv6_routing_table>]

Adds an unrechable route into a routing table

  • ipv6 route del

     ipv6 route del <prefix/length> [table <ipv6_routing_table>]

Deletes a route from a routing table

  • ipv6 route default auto

     ipv6 route default auto dev <vif_name> enable|disable

Enables intallation of default routes based on the information in Router Advertisement messages. Once enabled the router will create the default route based on the first RA message received on the interface and will associate a timer with that route. The timer is set to RA.lifetime value. When it expires the default route will be deleted and the router will install a new default route based on another RA message received at the interface.

  • ipv6 nd ra

     ipv6 nd ra enable|disable dev <vif_name>

Enables or disables Router Advertisements at an interface. If disabled router will not transmit Router Advertisement messages at an interface and will not answer to Router Solicitation messages

  • sh ipv6 arp

     sh ipv6 arp

Outputs ipv6 neighbor cache entries

  • ipv6 arp add

     ipv6 arp add <ipv6-address> <mac-address> dev <vif_name> [static]

Creates or alters an ipv6 neighbor cache entry

  • ipv6 arp del

     ipv6 arp del <ipv6-address> dev <vif_name>

Deletes an ipv6 neighbor cache entry

  • icmp6 error msg

     icmp6 error msg type <number> code <number> enable|disable

Enables or disables generation of an icmp messages with the given type and code

  • sh icmp6 error msg

     sh icmp6 error msg type <number> code <number>

Outputs state of icmp error message

  • ipv6 nd ra lifetime

     ipv6 nd ra lifetime <number> dev <vif_name>

Configures lifetime field value of Router Advertisement messages sent from an interface

  • ipv6 nd ra interval

     ipv6 nd ra interval <min_number> <max_number> dev <vif_name>

Configures the MinRtrAdvInterval and MaxRtrAdvInterval values (seconds) See https://tools.ietf.org/html/rfc4861#page-40 6.2.1. Router Configuration Variables

  • ipv6 nd ra reachable

     ipv6 nd ra reachable <number> dev <vif_name>

Configures the value to be placed in the Reachable Time field in the Router Advertisement messages sent by the router. The value zero means unspecified (by this router).
MUST be no greater than 3,600,000 milliseconds (1 hour).

  • ipv6 nd ra retrans_timer

     ipv6 nd ra retrans_timer <number> dev <vif_name>

Configures the value to be placed in the Retrans Timer field in the Router Advertisement messages sent by the router.
The value zero means unspecified (by this router).

  • ipv6 nd ra hop_limit

     ipv6 nd ra hop_limit <number> dev <vif_name>

Configures the default value to be placed in the Cur Hop Limit field in the Router Advertisement messages sent by the router. The value should be set to the current diameter of the Internet. The value zero means unspecified (by this router).

  • ipv6 nd ra prefix add|update

     ipv6 nd ra prefix add|update <prefix/length> [valid_lt <number>] 
   [preferred_lt <number>] [flags O,A] dev <vif_name>

Adds or updates a prefix to/in Router Advertisement messages sent from an interface

  • ipv6 nd ra prefix del

     ipv6 nd ra prefix add <prefix/length> dev <vif_name>

Deletes a prefix from Router Advertisement messages sent from an interface

IPv6 sysctl variables

  • nd_retrans_timer

The time in milliseconds between retransmissions of Neighbor Solicitation messages to a neighbor when resolving the address or when probing the reachability of a neighbor.

  • fib6_max_route_tables

Max number of ipv6 routing tables

  • fib6_max_routes

Max number of ipv6 route entries

  • fib6_max_next_hops

Max number of ipv6 next hop entries

  • fib6_max_lpm_tbl8

Max number of lpm6 tbl8. See https://doc.dpdk.org/guides/prog_guide/lpm6_lib.html

  • fib6_max_num_local_ip_addrs

Max number if ipv6 addresses assigned at interfaces

  • max_num_solicited_node_addrs

Max number of solicited node addresses

  • nd_neighbor_cache_size

ipv6 neighbor cache size

  • nd_neighbor_cache_entry_ttl

ipv6 neigbour cache entry time to live. Seconds.

  • icmp6_packet_rate

Icmpv6 error transmission rate in packets per seconds.

  • icmp6_transmission_burst

Max number of icmpv6 error messages that could be send at once

  • icmp6_num_buckets

Num icmp6 buckets

  • max_rtr_solicitation_delay

number of seconds to delay the transmission of router solicitation messages

  • dad_attempts

Number of attempts for Duplicate address detection algorithm

IPV6 VRRP version 3

  • vrrp create group

     vrrp create group <vrrp_id> dev <vif_name> address-family af_ipv6 version 3

Creates vrrp group

  • vrrp group ipv6 add

     vrrp group <vrrp_id> dev <vif_name> ipv6 add <link-local ipv6 address>

Setup or change ipv6 link-local address of a vrrp3 ipv6 group

  • vrrp group ipv6 add

     vrrp group <vrrp_id> dev <vif_name> ipv6 add <ipv6_address> secondary

Add secondary ipv6 global address to a vrrp3 ipv6 group

  • vrrp group ipv6 del

     vrrp group <vrrp_id> dev <vif_name> ipv6 del <ipv6_address> secondary

Del secondary ipv6 global address from a vrrp3 ipv6 group

  • vrrp group prio

     vrrp group <vrrp_id> dev <vif_name> prio <value>

Change priority of a vrrp group

  • vrrp group advert_int

     vrrp group <vrrp_id> dev <vif_name> advert_int <value>

Change advertisement transmission interval (cetiseconds, 100 centiseconds == 1 sec) of a vrrp group

  • vrrp group accept_mode

     vrrp group <vrrp_id> dev <vif_name> accept_mode <on|off>

Change accept_mode of a vrrp group

  • vrrp group preempt_mode

     vrrp group <vrrp_id> dev <vif_name> preempt_mode <on|off>

Change preempt_mode of a vrrp group

  • sh vrrp

     sh vrrp

Shows vrrp group information

  • vrrp group del

     vrrp del group <vrrp_id> dev <vif_name>

Deletes vrrp group

  • vrrp group nd ra enable/disable

     vrrp group <vrrp_id> dev <vif_name> nd ra enable|disable

Enables or disables transmission of ND Router Advertisement messages for a VRRP IPV6 group

  • vrrp group nd ra lifetime

     vrrp group <vrrp_id> dev <vif_name> nd ra lifetime <value>

Configures lifetime field value of Router Advertisement messages sent for a VRRP IPV6 group

  • vrrp group nd ra interval

     vrrp group <vrrp_id> dev <vif_name> nd ra interval <min_number> <max_number>

Configures the MinRtrAdvInterval and MaxRtrAdvInterval values (seconds) See https://tools.ietf.org/html/rfc4861#page-40 6.2.1. Router Configuration Variables

  • vrrp group nd ra reachable

     vrrp group <vrrp_id> dev <vif_name> nd ra reachable <number>

Configures the value to be placed in the Reachable Time field in the Router Advertisement messages sent by the router. The value zero means unspecified (by this router).
MUST be no greater than 3,600,000 milliseconds (1 hour).

  • vrrp group nd ra retrans_timer

     vrrp group <vrrp_id> dev <vif_name> nd ra retrans_timer <number>

Configures the value to be placed in the Retrans Timer field in the Router Advertisement messages sent by the router.
The value zero means unspecified (by this router).

  • vrrp group nd ra hop_limit

     vrrp group <vrrp_id> dev <vif_name> nd ra hop_limit <number>

Configures the default value to be placed in the Cur Hop Limit field in the Router Advertisement messages sent by the router. The value should be set to the current diameter of the Internet. The value zero means unspecified (by this router).

  • vrrp group nd ra prefix add|update

     vrrp group <vrrp_id> dev <vif_name> nd ra prefix add|update <prefix/length> [valid_lt <number>] 
   [preferred_lt <number>] [flags O,A] dev <vif_name>

Adds or updates a prefix to/in Router Advertisement messages sent for a VRRP IPV6 group

  • vrrp group nd ra prefix del

     vrrp group <vrrp_id> dev <vif_name>  nd ra prefix add <prefix/length> dev <vif_name>

Deletes a prefix from Router Advertisement messages sent for a VRRP IPV6 group