====== Сервис QOS ======

  * [[https://habr.com/post/420525/|Сети для самых матёрых. Часть пятнадцатая. QoS]]

===== Уилиты для экспериментов =====

  * [[Утилита iPerf]]

<code>
gate# apt install wondershaper 

gate# wondershaper eth1 10 10

gate# wondershaper eth1

gate# wondershaper clear eth1
</code>

===== Linux =====
===== Cisco =====

  * [[https://www.cisco.com/c/en/us/support/docs/switches/catalyst-3750-series-switches/91862-cat3750-qos-config.html|Cisco Catalyst 3750 QoS Configuration Examples]]
  * [[https://reggle.wordpress.com/2013/05/14/qos-part-v-hardware-queues-on-35603750-switch-platform/|QoS part V: hardware queues on 3560/3750 switch platform]]
  * [[http://cisco-conf.ru/qos/90-qos-catalyst-3750-.html|3750 QOS: Общие принципы]]

==== Включение ====

<code>
mls qos
</code>
==== Назначение QOS для трафика c определенным DSCP  ====

  * [[https://dreamforccie.wordpress.com/2010/04/23/convert-phb-af-to-dscp-decimal-and-vise-versa-its-so-easy/|Convert PHB AF to DSCP Decimal and Vise Versa: It’s So Easy!]]
  * [[https://supportforums.cisco.com/t5/wan-routing-and-switching/dscp-qos-value/td-p/920725|DSCP - QOS value]]

<code>
TOS: 0x000068b8

show mls qos maps dscp-output-q

interface GigabitEthernet1/0/1
 mls qos trust dscp
</code>

==== Назначение DSCP для трафика c определенным профилем ====

<code>
no ip access-list extended IP-Communicator
ip access-list extended IP-Communicator
! permit udp any any range 10000 20000
! permit ip any host 195.19.32.11
 permit ip host 195.19.32.11 any

class-map match-all Class-B
 match access-group name IP-Communicator

policy-map sample-policy2
 class Class-B
  set dscp ef

interface GigabitEthernet1/0/2
 service-policy input sample-policy2
</code>

==== Назначение QOS для трафика отдельного порта ====

<code>
show mls qos maps cos-output-q

interface GigabitEthernet1/0/11
 mls qos cos 5
 mls qos trust cos 
</code>

==== Выделение полосы в магистральном соединении для трафика с определенным QOS ====

<code>
sh mls qos queue-set

show mls qos interface g1/0/47 queueing

interface GigabitEthernet1/0/47
! Full priority for q1
 srr-queue bandwidth shape 0 0 0 0
 srr-queue bandwidth share 255 1 1 1

! 1/20 band for q1
 srr-queue bandwidth shape 20 0 0 0
 srr-queue bandwidth share 1 1 1 1

clear mls qos interface g1/0/47 statistics
show mls qos interface g1/0/47 statistics | inc queue
</code>

===== Равномерное ограничение полосы для всех с исключениями =====

==== FreeBSD (ipfw) ====
<code>
[gate:~] # cat /etc/ipfw.conf
</code><code>
pipe 1 config bw 100mbit/s
pipe 2 config bw 64kbit/s mask dst-ip 0xffffffff

add 10 pipe 1 ip from any to 192.168.100+X.2
add 20 pipe 2 ip from any to 192.168.100+X.0/24

add 100 allow ip from any to any
</code><code>
[gate:~] # cat /etc/rc.conf
</code><code>
...
dummynet_enable=yes
firewall_enable="YES"
firewall_type="/etc/ipfw.conf"
</code><code>
[gate:~] # /etc/rc.d/ipfw start

[gate:~] # ipfw pipe show
</code>
===== Балансировка нагрузки между провайдерами =====

==== FreeBSD (pf) ====

http://www.openbsd.org/faq/pf/pools.html
<code>
[gate:~] # cat /etc/pf.conf
</code><code>
lan_net = "192.168.1X.0/24"
int_if  = "tap0"
ext_if1 = "fxp0"
ext_if2 = "rl0"
ext_gw1 = "10.10.105.100"
ext_gw2 = "172.16.7.2"

nat on $ext_if1 from $lan_net to any -> ($ext_if1)
nat on $ext_if2 from $lan_net to any -> ($ext_if2)

pass in on $int_if route-to \
    { ($ext_if1 $ext_gw1), ($ext_if2 $ext_gw2) } round-robin \
    proto tcp from $lan_net to any flags S/SA modulate state
pass in on $int_if route-to \
    { ($ext_if1 $ext_gw1), ($ext_if2 $ext_gw2) } round-robin \
    proto { udp, icmp } from $lan_net to any keep state

pass out on $ext_if1 route-to ($ext_if2 $ext_gw2) from $ext_if2 \
   to any
pass out on $ext_if2 route-to ($ext_if1 $ext_gw1) from $ext_if1 \
   to any
</code>