Manual:CRS1xx/2xx series switches

From MikroTik Wiki
Revision as of 08:21, 26 January 2018 by Becs (talk | contribs) (→‎QoS Group)
Jump to navigation Jump to search
Version.png

Applies to RouterOS: v6.12 +


Summary

The Cloud Router Switch series are highly integrated switches with high performance MIPS CPU and feature-rich packet processor. The CRS switches can be designed into various Ethernet applications including unmanaged switch, Layer 2 managed switch, carrier switch and wireless/wired unified packet processing.

Icon-warn.png

Warning: This article applies to CRS1xx and CRS2xx series switches and not to CRS3xx series switches. For CRS3xx series devices read the CRS3xx series switches manual.


FeaturesDescription
Forwarding
  • Configurable ports for switching or routing
  • Full non-blocking wirespeed switching
  • Up to 16k MAC entries in Unicast FDB for Layer 2 unicast forwarding
  • Up to 1k MAC entries in Multicast FDB for multicast forwarding
  • Up to 256 MAC entries in Reserved FDB for control and management purposes
  • All Forwarding Databases support IVL and SVL
  • Configurable Port based MAC learning limit
  • Jumbo frame support (CRS1xx: 4064 Bytes; CRS2xx: 9204 Bytes)
Mirroring
  • Various types of mirroring:
    • Port based mirroring
    • VLAN based mirroring
    • MAC based mirroring
  • 2 independent mirroring analyzer ports
VLAN
  • Fully compatible with IEEE802.1Q and IEEE802.1ad VLAN
  • 4k active VLANs
  • Flexible VLAN assignment:
    • Port based VLAN
    • Protocol based VLAN
    • MAC based VLAN
  • From any to any VLAN translation and swapping
  • 1:1 VLAN switching - VLAN to port mapping
  • VLAN filtering
Port Isolation and Leakage
  • Applicable for Private VLAN implementation
  • 3 port profile types: Promiscuous, Isolated and Community
  • Up to 28 Community profiles
  • Leakage profiles allow bypassing egress VLAN filtering
Trunking
  • Supports static link aggregation groups
  • Up to 8 Port Trunk groups
  • Up to 8 member ports per Port Trunk group
  • Hardware automatic failover and load balancing
Quality of Service (QoS)
  • Flexible QoS classification and assignment:
    • Port based
    • MAC based
    • VLAN based
    • Protocol based
    • PCP/DEI based
    • DSCP based
    • ACL based
  • QoS remarking and remapping for QoS domain translation between service provider and client networks
  • Overriding of each QoS assignment according to the configured priority
Shaping and Scheduling
  • 8 queues on each physical port
  • Shaping per port, per queue, per queue group
Access Control List
  • Ingress and Egress ACL tables
  • Up to 128 ACL rules (limited by RouterOS)
  • Classification based on ports, L2, L3, L4 protocol header fields
  • ACL actions include filtering, forwarding and modifying of the protocol header fields

Cloud Router Switch models

This table clarifies main differences between Cloud Router Switch models.

Model Switch Chip CPU Wireless SFP+ port Access Control List Jumbo Frame (Bytes)
CRS105-5S-FB QCA-8511 400MHz - - + 9204
CRS106-1C-5S QCA-8511 400MHz - - + 9204
CRS112-8G-4S QCA-8511 400MHz - - + 9204
CRS210-8G-2S+ QCA-8519 400MHz - + + 9204
CRS212-1G-10S-1S+ QCA-8519 400MHz - + + 9204
CRS226-24G-2S+ QCA-8519 400MHz - + + 9204
CRS125-24G-1S QCA-8513L 600MHz - - - 4064
CRS125-24G-1S-2HnD QCA-8513L 600MHz + - - 4064
CRS109-8G-1S-2HnD QCA-8513L 600MHz + - - 4064

Cloud Router Switch configuration examples

Abbreviations and Explanations

CVID - Customer VLAN id: inner VLAN tag id of the IEEE 802.1ad frame

SVID - Service VLAN id: outer VLAN tag id of the IEEE 802.1ad frame

IVL - Independent VLAN learning - learning/lookup is based on both MAC addresses and VLAN IDs.

SVL - Shared VLAN learning - learning/lookup is based on MAC addresses - not on VLAN IDs.

TPID - Tag Protocol Identifier

PCP - Priority Code Point: a 3-bit field which refers to the IEEE 802.1p priority

DEI - Drop Eligible Indicator

DSCP - Differentiated services Code Point

Drop precedence - internal CRS switch QoS attribute used for packet enqueuing or dropping.

Port Switching

Similarly to other RouterBoards, port switching on CRS allows wire-speed traffic forwarding among a group of ports, like the ports were a regular Ethernet switch. This feature is configurable by setting a "master-port" property to one or more ports in /interface ethernet menu. The "master-port" will be the port through which the RouterOS will communicate to all ports in the group. Interfaces which have the "master-port" specified become isolated - no traffic can be received and no traffic can be sent out directly from RouterOS.

Here is a general diagram of RouterBoard with a five port switch chip:

Switch-png.png

A packet that is received by one of the ports always passes through the switch logic first. Switch logic decides to which ports the packet should be going to. Passing packet "up" or giving it to RouterOS is also called sending it to switch chip's “CPU” port. It means at that point switch forwards the packet to CPU port the packet starts to get processed by RouterOS as incoming packet of the “master-port”. If the packet does not have to go to “CPU” port, it is handled entirely by switch logic, does not require any CPU resources and happens at wire-speed.

Additionally, CRS series switches support multiple “master-port” configurations and have no port selection limitations for a port group which makes possible many various switched port combinations with all CRS switch interfaces. But no port can be in more than one switch group.

For example, consider a CRS125 switch with 24 Ethernet interfaces and 1 SFP interface:

[admin@MikroTik] > interface ethernet print 
Flags: X - disabled, R - running, S - slave 
 #    NAME            MTU MAC-ADDRESS       ARP        MASTER-PORT          SWITCH         
 0 R  ether1         1500 D4:CA:6D:F9:FE:2F enabled    none                 switch1        
 1    ether2         1500 D4:CA:6D:F9:FE:30 enabled    none                 switch1        
 2    ether3         1500 D4:CA:6D:F9:FE:31 enabled    none                 switch1        
 3    ether4         1500 D4:CA:6D:F9:FE:32 enabled    none                 switch1        
 4 R  ether5         1500 D4:CA:6D:F9:FE:33 enabled    none                 switch1        
 5 R  ether6         1500 D4:CA:6D:F9:FE:34 enabled    none                 switch1        
 6    ether7         1500 D4:CA:6D:F9:FE:35 enabled    none                 switch1        
 7    ether8         1500 D4:CA:6D:F9:FE:36 enabled    none                 switch1         
...     
22    ether23        1500 D4:CA:6D:F9:FE:45 enabled    none                 switch1        
23 R  ether24        1500 D4:CA:6D:F9:FE:46 enabled    none                 switch1        
24    sfp1           1500 D4:CA:6D:F9:FE:47 enabled    none                 switch1        

And there are configured 3 switch groups: 1) ether2, ether3, ether4, ether5, ether6; 2) ether13, ether14, ether15, ether16, ether17, ether18, ether19, ether20; 3) ether21, ether22, ether23, ether24, sfp1.

Ports ether1, ether7-ether12 are not switched in this example, they remain as independent router ports.

[admin@MikroTik] /interface ethernet>
set ether3,ether4,ether5,ether6 master-port=ether2
[admin@MikroTik] /interface ethernet>
set ether14,ether15,ether16,ether17,ether18,ether19,ether20 master-port=ether13
[admin@MikroTik] /interface ethernet>
set ether22,ether23,ether24,sfp1 master-port=ether21

[admin@MikroTik] /interface ethernet> print 
Flags: X - disabled, R - running, S - slave 
 #    NAME            MTU MAC-ADDRESS       ARP        MASTER-PORT          SWITCH         
 0 R  ether1         1500 D4:CA:6D:F9:FE:2F enabled    none                 switch1        
 1 R  ether2         1500 D4:CA:6D:F9:FE:30 enabled    none                 switch1        
 2  S ether3         1500 D4:CA:6D:F9:FE:31 enabled    ether2               switch1        
 3  S ether4         1500 D4:CA:6D:F9:FE:32 enabled    ether2               switch1        
 4 RS ether5         1500 D4:CA:6D:F9:FE:33 enabled    ether2               switch1        
 5 RS ether6         1500 D4:CA:6D:F9:FE:34 enabled    ether2               switch1        
 6    ether7         1500 D4:CA:6D:F9:FE:35 enabled    none                 switch1        
 7    ether8         1500 D4:CA:6D:F9:FE:36 enabled    none                 switch1        
 8    ether9         1500 D4:CA:6D:F9:FE:37 enabled    none                 switch1        
 9    ether10        1500 D4:CA:6D:F9:FE:38 enabled    none                 switch1        
10    ether11        1500 D4:CA:6D:F9:FE:39 enabled    none                 switch1        
11    ether12        1500 D4:CA:6D:F9:FE:3A enabled    none                 switch1        
12 R  ether13        1500 D4:CA:6D:F9:FE:3B enabled    none                 switch1        
13  S ether14        1500 D4:CA:6D:F9:FE:3C enabled    ether13              switch1        
14  S ether15        1500 D4:CA:6D:F9:FE:3D enabled    ether13              switch1        
15 RS ether16        1500 D4:CA:6D:F9:FE:3E enabled    ether13              switch1        
16  S ether17        1500 D4:CA:6D:F9:FE:3F enabled    ether13              switch1        
17  S ether18        1500 D4:CA:6D:F9:FE:40 enabled    ether13              switch1        
18  S ether19        1500 D4:CA:6D:F9:FE:41 enabled    ether13              switch1        
19  S ether20        1500 D4:CA:6D:F9:FE:42 enabled    ether13              switch1        
20 R  ether21        1500 D4:CA:6D:F9:FE:43 enabled    none                 switch1        
21  S ether22        1500 D4:CA:6D:F9:FE:44 enabled    ether21              switch1        
22  S ether23        1500 D4:CA:6D:F9:FE:45 enabled    ether21              switch1        
23 RS ether24        1500 D4:CA:6D:F9:FE:46 enabled    ether21              switch1        
24  S sfp1           1500 D4:CA:6D:F9:FE:47 enabled    ether21              switch1        

Now ether2 is the “master-port” of the group 1, ether13 – of the group 2 and ether21 – of the group 3.

Note: Previously a link was detected only on interfaces with a physical connection, but now since the ether2, ether13 and ether21 have connection to CPU, the running flag is propagated to them, as well.

Alt text
CRS Port Switching Example

In essence this configuration is the same as if you had a RouterBoard with 10 Ethernet interfaces and 3 switches:

Alt text
CRS Port Switching Logic
Icon-note.png

Note: Dynamic reserved VLAN entries (VLAN4091; VLAN4090; VLAN4089; etc.) are created in CRS switch when switched port groups are added by setting new master-ports. These VLANs are necessary for internal operation and have lower precedence than user configured VLANs.


Icon-note.png

Note: Multiple master-port configuration is designed as fast and simple port isolation solution, but it limits a part of VLAN functionality supported by CRS switch-chip. For advanced configurations use one master-port within CRS switch chip for all ports, configure VLANs and isolate port groups with port isolation profile configuration.



Bridge Hardware Offloading

Since RouterOS v6.40rc29 there are user interface changes which convert RouterBoard master-port configuration into a bridge with hardware offloading. From now on bridges will handle all Layer2 forwarding and the use of switch chip (hw-offload) will automatically turn on if appropriate conditions are met. The rest of RouterOS Switch features remain untouched in usual menus. By default all newly created bridge ports have hw=yes option and it allows enabling of hw-offload when possible. If such functionality is not required, it can be disabled by hw=no on bridge port to have completely software operated bridging.

Icon-note.png

Note: Downgrading to previous RouterOS versions will not restore master-port configuration. The bridge with no hw-offload will appear instead and master-port configuration will have to be redone from the beginning.


Following table states what features currently in v6.40rc keep bridge hardware offloading enabled on certain RouterBoard and switch chip models.

Notes:

  • Enabling this feature maintains hw-offload: +
  • Enabling this feature turns off hw-offload: -
RouterBoard/[Switch Chip] Model Features in Switch menu Bridge STP/RSTP Bridge MSTP Bridge IGMP Snooping Bridge VLAN Filtering Bonding
CRS3xx series + + + + + -
CRS1xx/CRS2xx series + + - + - -
[QCA8337] + + - - - -
[AR8327] + + - - - -
[AR8227] + + - - - -
[AR8316] + + - - - -
[AR7240] + + - - - -
RB750Gr3 [MT7621] + - - - - -
RB1100AHx4 [RTL8367] + - - - - -
[ICPlus175D] + - - - - -
  • Port switching with master-port configuration before v6.40rc29
[admin@MikroTik] > interface ethernet export 
/interface ethernet
set [ find default-name=ether3 ] master-port=ether2
set [ find default-name=ether4 ] master-port=ether2
set [ find default-name=ether5 ] master-port=ether2
[admin@MikroTik] > 

[admin@MikroTik] > interface ethernet print 
Flags: X - disabled, R - running, S - slave 
 #    NAME             MTU MAC-ADDRESS       ARP             MASTER-PORT          SWITCH         
 0 R  ether1          1500 D4:CA:6D:E2:64:64 enabled         none                 switch1        
 1 R  ether2          1500 D4:CA:6D:E2:64:65 enabled         none                 switch1        
 2 RS ether3          1500 D4:CA:6D:E2:64:66 enabled         ether2               switch1        
 3 RS ether4          1500 D4:CA:6D:E2:64:67 enabled         ether2               switch1        
 4 RS ether5          1500 D4:CA:6D:E2:64:68 enabled         ether2               switch1                     
[admin@MikroTik] > 
  • Port switching with bridge configuration and enabled hw-offload since v6.40rc29
[admin@MikroTik] > interface bridge export 
/interface bridge
add name=bridge1 igmp-snooping=no  protocol-mode=none
/interface bridge port
add bridge=bridge1 interface=ether2
add bridge=bridge1 interface=ether3
add bridge=bridge1 interface=ether4
add bridge=bridge1 interface=ether5
[admin@MikroTik] > 

[admin@MikroTik] > interface bridge port print 
Flags: X - disabled, I - inactive, D - dynamic, H - hw-offload 
 #     INTERFACE              BRIDGE              HW  PVID PRIORITY  PATH-COST INTERNAL-PATH-COST    HORIZON
 0   H ether2                 bridge1             yes    1     0x80         10                 10       none
 1   H ether3                 bridge1             yes    1     0x80         10                 10       none
 2   H ether4                 bridge1             yes    1     0x80         10                 10       none
 3   H ether5                 bridge1             yes    1     0x80         10                 10       none
[admin@MikroTik] > 

Global Settings

Sub-menu: /interface ethernet switch


CRS switch chip is configurable from the /interface ethernet switch console menu.

Property Description
name (string value; Default: switch1) Name of the switch.
bridge-type (customer-vid-used-as-lookup-vid | service-vid-used-as-lookup-vid; Default: customer-vid-used-as-lookup-vid) Bridge type defines which VLAN tag is used as Lookup-VID. Lookup-VID serves as the VLAN key for all VLAN-based lookup.
mac-level-isolation (yes | no; Default: yes) Enables or disables MAC level isolation.
use-svid-in-one2one-vlan-lookup (yes | no; Default: no) Whether to use service VLAN id for 1:1 VLAN switching lookup.
use-cvid-in-one2one-vlan-lookup (yes | no; Default: yes) Whether to use customer VLAN id for 1:1 VLAN switching lookup.
multicast-lookup-mode

(dst-ip-and-vid-for-ipv4 | dst-mac-and-vid-always;

Default:dst-ip-and-vid-for-ipv4)
Lookup mode for IPv4 multicast bridging.
  • dst-mac-and-vid-always - For all packet types lookup key is destination MAC and VLAN id.
  • dst-ip-and-vid-for-ipv4 - For IPv4 packets lookup key is destination IP and VLAN id. For other packet types lookup key is destination MAC and VLAN id.
unicast-fdb-timeout (time interval; Default: 5m) Timeout for Unicast FDB entries.
override-existing-when-ufdb-full (yes | no; Default: no) Enable or disable to override existing entry which has the lowest aging value when UFDB is full.


Property Description
drop-if-no-vlan-assignment-on-ports (ports; Default: none) Ports which drop frames if no MAC-based, Protocol-based VLAN assignment or Ingress VLAN Translation is applied.
drop-if-invalid-or-src-port-
-not-member-of-vlan-on-ports

(ports; Default: none)
Ports which drop invalid and other port VLAN id frames.
unknown-vlan-lookup-mode (ivl | svl; Default: svl) Lookup and learning mode for packets with invalid VLAN.
forward-unknown-vlan (yes | no; Default: yes) Whether to allow forwarding VLANs which are not members of VLAN table.


Property Description
bypass-vlan-ingress-filter-for (protocols; Default: none) Protocols which are excluded from Ingress VLAN filtering. These

protocols are not dropped if they have invalid VLAN. (arp, dhcpv4, dhcpv6,

eapol, igmp, mld, nd, pppoe-discovery, ripv1)
bypass-ingress-port-policing-for (protocols; Default: none) Protocols which are excluded from Ingress Port Policing. (arp, dhcpv4, dhcpv6, eapol, igmp, mld, nd, pppoe-discovery, ripv1)
bypass-l2-security-check-filter-for (protocols; Default: none) Protocols which are excluded from Policy rule security check. (arp, dhcpv4, dhcpv6, eapol, igmp, mld, nd, pppoe-discovery, ripv1)


Property Description
ingress-mirror0 (port | trunk,format; Default: none,modified) The first ingress mirroring analyzer port or trunk and mirroring format:
  • analyzer-configured - The packet is same as the packet to destination. VLAN format is modified based on the VLAN configurations of the analyzer port.
  • modified - The packet is same as the packet to destination. VLAN format is modified based on the VLAN configurations of the egress port.
  • original - Traffic is mirrored without any change to the original incoming packet format. But service VLAN tag is stripped in edge port.
ingress-mirror1 (port | trunk,format; Default: none,modified) The second ingress mirroring analyzer port or trunk and mirroring format:
  • analyzer-configured - The packet is same as the packet to destination. VLAN format is modified based on the VLAN configurations of the analyzer port.
  • modified - The packet is same as the packet to destination. VLAN format is modified based on the VLAN configurations of the egress port.
  • original - Traffic is mirrored without any change to the original incoming packet format. But service VLAN tag is stripped in edge port.
ingress-mirror-ratio (1/32768..1/1; Default: 1/1) Proportion of ingress mirrored packets compared to all packets.
egress-mirror0 (port | trunk,format; Default: none,modified) The first egress mirroring analyzer port or trunk and mirroring format:
  • analyzer-configured - The packet is same as the packet to destination. VLAN format is modified based on the VLAN configurations of the analyzer port.
  • modified - The packet is same as the packet to destination. VLAN format is modified based on the VLAN configurations of the egress port.
  • original - Traffic is mirrored without any change to the original incoming packet format. But service VLAN tag is stripped in edge port.
egress-mirror1 (port | trunk,format; Default: none,modified) The second egress mirroring analyzer port or trunk and mirroring format:
  • analyzer-configured - The packet is same as the packet to destination. VLAN format is modified based on the VLAN configurations of the analyzer port.
  • modified - The packet is same as the packet to destination. VLAN format is modified based on the VLAN configurations of the egress port.
  • original - Traffic is mirrored without any change to the original incoming packet format. But service VLAN tag is stripped in edge port.
egress-mirror-ratio (1/32768..1/1; Default: 1/1) Proportion of egress mirrored packets compared to all packets.
mirror-egress-if-ingress-mirrored (yes | no; Default: no) When packet is applied to both ingress and egress mirroring, if this

setting is disabled, only ingress mirroring is performed on the packet; if this

setting is enabled both mirroring types are applied.
mirror-tx-on-mirror-port (yes | no; Default: no)
mirrored-packet-qos-priority (0..7; Default: 0) Remarked priority in mirrored packets.
mirrored-packet-drop-precedence (drop | green | red | yellow; Default: green) Remarked drop precedence in mirrored packets. This QoS attribute is used for mirrored packet enqueuing or dropping.
fdb-uses (mirror0 | mirror1; Default: mirror0) Analyzer port used for FDB-based mirroring.
vlan-uses (mirror0 | mirror1; Default: mirror0) Analyzer port used for VLAN-based mirroring.

Port Settings

Sub-menu: /interface ethernet switch port


Property Description
vlan-type (edge-port | network-port; Default: network-port) Port VLAN type specifies whether VLAN id is used in UFDB learning. Network port learns VLAN id in UFDB, edge port does not - VLAN 0. It can be observed only in IVL learning mode.
isolation-leakage-profile-override (yes | no; Default:

!isolation-leakage-profile-override)

isolation-leakage-profile (0..31;)
Custom port profile for port isolation/leakage configurations.
  • Port-level isolation profile 0. Uplink port - allows the port to communicate with all ports in the device.
  • Port-level isolation profile 1. Isolated port - allows the port to communicate only with uplink ports.
  • Port-level isolation profile 2 - 31. Community port - allows communication among the same community ports and uplink ports.
learn-override (yes | no; Default: !learn-override)
learn-limit (1..1023; Default: !learn-limit)
Enable or disable MAC address learning and set MAC limit on the port. MAC learning limit is disabled by default when !learn-override and !learn-limit
drop-when-ufdb-entry-src-drop (yes | no; Default: yes) Enable or disable to drop packets when UFDB entry has action src-drop.
allow-unicast-loopback (yes | no; Default: no) Unicast loopback on port. When enabled, it permits sending back when

source port and destination port are the same one for known unicast

packets.
allow-multicast-loopback (yes | no; Default: no) Multicast loopback on port. When enabled, it permits sending back when

source port and destination port are the same for registered multicast or

broadcast packets.
action-on-static-station-move (copy-to-cpu | drop | forward | redirect-to-cpu; Default: forward) Action for packets when UFDB already contains static entry with such MAC but with a different port.
drop-dynamic-mac-move (yes | no; Default: no) Prevents MAC relearning until UFDB timeout if MAC is already learned on other port.


Property Description
allow-fdb-based-vlan-translate (yes | no; Default: no) Enable or disable MAC-based VLAN translation on the port.
allow-mac-based-service-vlan-assignment-for (all-frames | none |

tagged-frame-only | untagged-and-priority-tagged-frame-only; Default:

none)
Frame type for which applies MAC-based service VLAN translation.
allow-mac-based-customer-vlan-assignment-for (all-frames | none |

tagged-frame-only | untagged-and-priority-tagged-frame-only; Default:

none)
Frame type for which applies MAC-based customer VLAN translation.
default-customer-pcp (0..7; Default: 0) Default customer PCP of the port.
default-service-pcp (0..7; Default: 0) Default service PCP of the port.
pcp-propagation-for-initial-pcp (yes | no; Default: no) Enables or disables PCP propagation for initial PCP assignment on ingress.
  • If the port vlan-type is Edge port, the service PCP is copied from the customer PCP.
  • If the port vlan-type is Network port, the customer PCP is copied from the service PCP.
filter-untagged-frame (yes | no; Default: no) Whether to filter untagged frames on the port.
filter-priority-tagged-frame (yes | no; Default: no) Whether to filter tagged frames with priority on the port.
filter-tagged-frame (yes | no; Default: no) Whether to filter tagged frames on the port.


Property Description
egress-vlan-tag-table-lookup-key (according-to-bridge-type | egress-vid; Default: egress-vid) Egress VLAN table (VLAN Tagging) lookup:
  • egress-vid - Lookup VLAN id is CVID when Edge port is configured, SVID when Network port is configured.
  • according-to-bridge-type - Lookup VLAN id is CVID when customer VLAN bridge is configured, SVID when service VLAN bridge is configured. Customer tag is unmodified for Edge port in service VLAN bridge.
egress-vlan-mode (tagged | unmodified | untagged; Default: unmodified) Egress VLAN tagging action on the port.
egress-pcp-propagation (yes | no; Default: no) Enables or disables egress PCP propagation.
  • If the port vlan-type is Edge port, the service PCP is copied from the customer PCP.
  • If the port vlan-type is Network port, the customer PCP is copied from the service PCP.


Property Description
ingress-mirror-to (mirror0 | mirror1 | none; Default: none) Analyzer port for port-based ingress mirroring.
ingress-mirroring-according-to-vlan (yes | no; Default: no)
egress-mirror-to (mirror0 | mirror1 | none; Default: none) Analyzer port for port-based egress mirroring.


Property Description
qos-scheme-precedence (da-based | dscp-based | ingress-acl-based | pcp-based | protocol-based | sa-based | vlan-based; Default: pcp-based, sa-based, da-based, dscp-based, protocol-based, vlan-based) Specifies applied QoS assignment schemes on ingress of the port.
  • da-based
  • dscp-based
  • ingress-acl-based
  • pcp-based
  • protocol-based
  • sa-based
  • vlan-based
pcp-or-dscp-based-qos-change-dei (yes | no; Default: no) Enable or disable PCP or DSCP based DEI change on port.
pcp-or-dscp-based-qos-change-pcp (yes | no; Default: no) Enable or disable PCP or DSCP based PCP change on port.
pcp-or-dscp-based-qos-change-dscp (yes | no; Default: no) Enable or disable PCP or DSCP based DSCP change on port.
dscp-based-qos-dscp-to-dscp-mapping (yes | no; Default: yes) Enable or disable DSCP to internal DSCP mapping on port.
pcp-based-qos-drop-precedence-mapping (PCP/DEI-range:drop-precedence; Default: 0-15:green) The new value of drop precedence for the PCP/DEI to drop precedence (drop | green | red | yellow) mapping. Multiple mappings allowed separated by comma e.g. "0-7:yellow,8-15:red".
pcp-based-qos-dscp-mapping (PCP/DEI-range:DEI; Default: 0-15:0) The new value of DSCP for the PCP/DEI to DSCP (0..63) mapping. Multiple mappings allowed separated by comma e.g. "0-7:25,8-15:50".
pcp-based-qos-dei-mapping (PCP/DEI-range:DEI; Default: 0-15:0) The new value of DEI for the PCP/DEI to DEI (0..1) mapping. Multiple mappings allowed separated by comma e.g. "0-7:0,8-15:1".
pcp-based-qos-pcp-mapping (PCP/DEI-range:DEI; Default: 0-15:0) The new value of PCP for the PCP/DEI to PCP (0..7) mapping. Multiple mappings allowed separated by comma e.g. "0-7:3,8-15:4".
pcp-based-qos-priority-mapping (PCP/DEI-range:DEI; Default: 0-15:0) The new value of internal priority for the PCP/DEI to priority (0..15) mapping. Multiple mappings allowed separated by comma e.g. "0-7:5,8-15:15".


Property Description
priority-to-queue (priority-range:queue; Default: 0-15:0,1:1,2:2,3:3) Internal priority (0..15) mapping to queue (0..7) per port.
per-queue-scheduling (Scheduling-type:Weight;

Default: wrr-group0:1,wrr-group0:2,wrr-group0:4,wrr-group0:8,wrr-group0:16,wrr-group0:32,

wrr-group0:64,wrr-group0:128)


Property Description
ingress-customer-tpid-override (yes | no;

Default:!ingress-customer-tpid-override)

ingress-customer-tpid (0..10000; Default: 0x8100)
Ingress customer TPID override allows accepting specific frames with a custom customer tag TPID. Default value is for tag of 802.1Q frames.
egress-customer-tpid-override (yes | no; Default:

!egress-customer-tpid-override)
egress-customer-tpid (0..10000; Default:

0x8100)
Egress customer TPID override allows custom identification for egress frames with a customer tag. Default value is for tag of 802.1Q frames.
ingress-service-tpid-override (yes | no; Default:

!ingress-service-tpid-override)

ingress-service-tpid (0..10000; Default: 0x88A8)
Ingress service TPID override allows accepting specific frames with a custom service tag TPID. Default value is for service tag of 802.1AD frames.
egress-service-tpid-override (yes | no; Default:

!egress-service-tpid-override)
egress-service-tpid (0..10000; Default:

0x88A8)
Egress service TPID override allows custom identification for egress frames with a service tag. Default value is for service tag of 802.1AD frames.


Property Description
custom-drop-counter-includes (counters; Default: none) Custom include to count dropped packets for switch port custom-drop-packet counter.
  • device-loopback
  • fdb-hash-violation
  • exceeded-port-learn-limitation
  • dynamic-station-move
  • static-station-move
  • ufdb-source-drop
  • host-source-drop
  • unknown-host
  • ingress-vlan-filtered
queue-custom-drop-counter0-includes (counters; Default: none) Custom include to count dropped packets for switch port tx-queue-custom0-drop-packet

and bytes for tx-queue-custom0-drop-byte counters.

  • red
  • yellow
  • green
  • queue0
  • ...
  • queue7
queue-custom-drop-counter1-includes (counters; Default: none) Custom include to count dropped packets for switch port tx-queue-custom1-drop-packet

and bytes for tx-queue-custom1-drop-byte counters.

  • red
  • yellow
  • green
  • queue0
  • ...
  • queue7
policy-drop-counter-includes (counters; Default: none) Custom include to count dropped packets for switch port policy-drop-packet counter.
  • ingress-policing
  • ingress-acl
  • egress-policing
  • egress-acl

Forwarding Databases

Unicast FDB

Sub-menu: /interface ethernet switch unicast-fdb


The unicast forwarding database supports up to 16318 MAC entries.

Property Description
action (action; Default: forward) Action for UFDB entry:
  • dst-drop - Packets are dropped when their destination MAC match the entry.
  • dst-redirect-to-cpu - Packets are redirected to CPU when their destination MAC match the entry.
  • forward - Packets are forwarded.
  • src-and-dst-drop - Packets are dropped when their source MAC or destination MAC match the entry.
  • src-and-dst-redirect-to-cpu - Packets are redirected to CPU when their source MAC or destination MAC match the entry.
  • src-drop - Packets are dropped when their source MAC match the entry.
  • src-redirect-to-cpu - Packets are redirected to CPU when their source MAC match the entry.
disabled (yes | no; Default: no) Enables or disables Unicast FDB entry.
isolation-profile (community1 | community2 | isolated | promiscuous; Default: promiscuous) MAC level isolation profile.
mac-address (MAC address) The action command applies to the packet when the destination MAC or source MAC matches the entry.
mirror (yes | no; Default: no) Enables or disables mirroring based on source MAC or destination MAC.
port (port) Matching port for the Unicast FDB entry.
qos-group (none; Default: none) Defined QoS group from QoS group menu.
svl (yes | no; Default: no) Unicast FDB learning mode:
  • Shared VLAN Learning (svl) - learning/lookup is based on MAC addresses - not on VLAN IDs.
  • Independent VLAN Learning (ivl) - learning/lookup is based on both MAC addresses and VLAN IDs.
vlan-id (0..4095) Unicast FDB lookup/learning VLAN id.

Multicast FDB

Sub-menu: /interface ethernet switch multicast-fdb


CRS125 switch-chip supports up to 1024 entries in MFDB for multicast forwarding. For each multicast packet, destination MAC or destination IP lookup is performed in MFDB. MFDB entries are not automatically learnt and can only be configured.

Property Description
address (X.X.X.X | XX:XX:XX:XX:XX:XX) Matching IP address or MAC address for multicast packets.
bypass-vlan-filter (yes | no; Default: no) Allow to bypass VLAN filtering for matching multicast packets.
disabled (yes | no; Default: no) Enables or disables Multicast FDB entry.
ports (ports) Member ports for multicast traffic.
qos-group (none; Default: none) Defined QoS group from QoS group menu.
svl (yes | no; Default: no) Multicast FDB learning mode:
  • Shared VLAN Learning (svl) - learning/lookup is based on MAC addresses - not on VLAN IDs.
  • Independent VLAN Learning (ivl) - learning/lookup is based on both MAC addresses and VLAN IDs.
vlan-id (0..4095; Default: 0) Multicast FDB lookup VLAN id. If VLAN learning mode is IVL, VLAN id is lookup id, otherwise VLAN id = 0.

Reserved FDB

Sub-menu: /interface ethernet switch reserved-fdb


Cloud Router Switch supports 256 RFDB entries. Each RFDB entry can store either Layer2 unicast or multicast MAC address with specific commands.

Property Description
action (copy-to-cpu | drop | forward | redirect-to-cpu; Default: forward) Action for RFDB entry:
  • copy-to-cpu - Packets are copied to CPU when their destination MAC match the entry.
  • drop - Packets are dropped when their destination MAC match the entry.
  • forward - Packets are forwarded when their destination MAC match the entry.
  • redirect-to-cpu - Packets are redirected to CPU when their destination MAC match the entry.
bypass-ingress-port-policing (yes | no; Default: no) Allow to bypass Ingress Port Policer for matching packets.
bypass-ingress-vlan-filter (yes | no; Default: no) Allow to bypass VLAN filtering for matching packets.
disabled (yes | no; Default: no) Enables or disables Reserved FDB entry.
mac-address (MAC address; Default: 00:00:00:00:00:00) Matching MAC address for Reserved FDB entry.
qos-group (none; Default: none) Defined QoS group from QoS group menu.

VLAN

VLAN Table

Sub-menu: /interface ethernet switch vlan


The VLAN table supports 4096 VLAN entries for storing VLAN member information as well as other VLAN information such as QoS, isolation, forced VLAN, learning, and mirroring.

Property Description
disabled (yes | no; Default: no) Indicate whether the VLAN entry is disabled. Only enabled entry is applied to lookup process and forwarding decision.
flood (yes | no; Default: no) Enables or disables forced VLAN flooding per VLAN. If the feature is

enabled, the result of destination MAC lookup in the UFDB or MFDB is ignored,

and the packet is forced to flood in the VLAN.
ingress-mirror (yes | no; Default: no) Enable the ingress mirror per VLAN to support the VLAN-based mirror function.
learn (yes | no; Default: yes) Enables or disables source MAC learning for VLAN.
ports (ports) Member ports of the VLAN.
qos-group (none; Default: none) Defined QoS group from QoS group menu.
svl (yes | no; Default: no) FDB lookup mode for lookup in UFDB and MFDB.
  • Shared VLAN Learning (svl) - learning/lookup is based on MAC addresses - not on VLAN IDs.
  • Independent VLAN Learning (ivl) - learning/lookup is based on both MAC addresses and VLAN IDs.
vlan-id (0..4095) VLAN id of the VLAN member entry.

Egress VLAN Tag

Sub-menu: /interface ethernet switch egress-vlan-tag


Egress packets can be assigned different VLAN tag format. The VLAN tags can be removed, added, or remained as is when the packet is sent to the egress port (destination port). Each port has dedicated control on the egress VLAN tag format. The tag formats include:

  • Untagged
  • Tagged
  • Unmodified

The Egress VLAN Tag table includes 4096 entries for VLAN tagging selection.

Property Description
disabled (yes | no; Default: no) Enables or disables Egress VLAN Tag table entry.
tagged-ports (ports) Ports which are tagged in egress.
vlan-id (0..4095) VLAN id which is tagged in egress.

Ingress/Egress VLAN Translation

The Ingress VLAN Translation table allows for up to 16 entries for each port. One or multiple fields can be selected from packet header for lookup in the Ingress VLAN Translation table. The S-VLAN or C-VLAN or both configured in the first matched entry is assigned to the packet.

Sub-menu: /interface ethernet switch ingress-vlan-translation


Sub-menu: /interface ethernet switch egress-vlan-translation


Property Description
customer-dei (0..1; Default: none) Matching DEI of the customer tag.
customer-pcp (0..7; Default: none) Matching PCP of the customer tag.
customer-vid (0..4095; Default: none) Matching VLAN id of the customer tag.
customer-vlan-format (any | priority-tagged-or-tagged | tagged | untagged-or-tagged; Default:any) Type of frames with customer tag for which VLAN translation rule is valid.
disabled (yes | no; Default: no) Enables or disables VLAN translation entry.
new-customer-vid (0..4095; Default: none) The new customer VLAN id which replaces matching customer VLAN id.
new-service-vid (0..4095; Default: none) The new service VLAN id which replaces matching service VLAN id.
pcp-propagation (yes | no; Default: no) Enables or disables PCP propagation.
  • If the port type is Edge, the customer PCP is copied from the service PCP.
  • If the port type is Network, the service PCP is copied from the customer PCP.
ports (ports) Matching switch ports for VLAN translation rule.
protocol (protocols; Default: none) Matching Ethernet protocol. (only for Ingress VLAN Translation)
sa-learning (yes | no; Default: no) Enables or disables source MAC learning after VLAN translation. (only for Ingress VLAN Translation)
service-dei (0..1; Default: none) Matching DEI of the service tag.
service-pcp (0..7; Default: none) Matching PCP of the service tag.
service-vid (0..4095; Default: none) Matching VLAN id of the service tag.
service-vlan-format (any | priority-tagged-or-tagged | tagged | untagged-or-tagged; Default:any) Type of frames with service tag for which VLAN translation rule is valid.

Protocol Based VLAN

Sub-menu: /interface ethernet switch protocol-based-vlan


Protocol Based VLAN table is used to assign VID and QoS attributes to related protocol packet per port.

Property Description
disabled (yes | no; Default: no) Enables or disables Protocol Based VLAN entry.
frame-type (ethernet | llc | rfc-1042; Default: ethernet) Encapsulation type of the matching frames.
new-customer-vid (0..4095; Default: 0) The new customer VLAN id which replaces original customer VLAN id for specified protocol.
new-service-vid (0..4095; Default: 0) The new service VLAN id which replaces original service VLAN id for specified protocol.
ports (ports) Matching switch ports for Protocol based VLAN rule.
protocol (protocol; Default: 0) Matching protocol for Protocol based VLAN rule.
qos-group (none; Default: none) Defined QoS group from QoS group menu.
set-customer-vid-for (all | none | tagged | untagged-or-priority-tagged; Default: all) Customer VLAN id assignment command for different packet type.
set-qos-for (all | none | tagged | untagged-or-priority-tagged; Default: none) Frame type for which QoS assignment command applies.
set-service-vid-for (all | none | tagged | untagged-or-priority-tagged; Default: all) Service VLAN id assignment command for different packet type.

MAC Based VLAN

Sub-menu: /interface ethernet switch mac-based-vlan


MAC Based VLAN table is used to assign VLAN based on source MAC.

Property Description
disabled (yes | no; Default: no) Enables or disables MAC Based VLAN entry.
new-customer-vid (0..4095; Default: 0) The new customer VLAN id which replaces original service VLAN id for matched packets.
new-service-vid (0..4095; Default: 0) The new service VLAN id which replaces original service VLAN id for matched packets.
src-mac-address (MAC address) Matching source MAC address for MAC based VLAN rule.
Icon-note.png

Note: All CRS1xx/2xx series switches support up to 1024 MAC Based VLAN table entries.


1:1 VLAN Switching

Sub-menu: /interface ethernet switch one2one-vlan-switching


1:1 VLAN switching can be used to replace the regular L2 bridging for matched packets. When a packet hits an 1:1 VLAN switching table entry, the destination port information in the entry is assigned to the packet. The matched destination information in UFDB and MFDB entry no longer applies to the packet.

Property Description
customer-vid (0..4095; Default: 0) Matching customer VLAN id for 1:1 VLAN switching.
disabled (yes | no; Default: no) Enables or disables 1:1 VLAN switching table entry.
dst-port (port) Destination port for matched 1:1 VLAN switching packets.
service-vid (0..4095; Default: 0) Matching customer VLAN id for 1:1 VLAN switching.

Port Isolation/Leakage

Sub-menu: /interface ethernet switch port-isolation


Sub-menu: /interface ethernet switch port-leakage


The CRS switches support flexible multi-level isolation features, which can be used for user access control, traffic engineering and advanced security and network management. The isolation features provide an organized fabric structure allowing user to easily program and control the access by port, MAC address, VLAN, protocol, flow and frame type. The following isolation and leakage features are supported:

  • Port-level isolation
  • MAC-level isolation
  • VLAN-level isolation
  • Protocol-level isolation
  • Flow-level isolation
  • Free combination of the above

Port-level isolation supports different control schemes on source port and destination port. Each entry can be programmed with access control for either source port or destination port.

  • When the entry is programmed with source port access control, the entry is

applied to the ingress packets.

  • When the entry is programmed with destination port access control, the entry

is applied to the egress packets.

Port leakage allows bypassing egress VLAN filtering on the port. Leaky port is allowed to access other ports for various applications such as security, network control and management. Note: When both isolation and leakage is applied to the same port, the port is isolated.

Property Description
disabled (yes | no; Default: no) Enables or disables port isolation/leakage entry.
flow-id (0..63; Default: none)
forwarding-type (bridged; routed; Default: bridged,routed) Matching traffic forwarding type on Cloud Router Switch.
mac-profile (community1 | community2 | isolated | promiscuous; Default: none) Matching MAC isolation/leakage profile.
port-profile (0..31; Default: none) Matching Port isolation/leakage profile.
ports (ports; Default: none) Isolated/leaked ports.
protocol-type (arp; nd; dhcpv4; dhcpv6; ripv1; Default: arp,nd,dhcpv4,dhcpv6,ripv1) Included protocols for isolation/leakage.
registration-status (known; unknown; Default: known,unknown) Registration status for matching packets. Known are present in UFDB and MFDB, unknown are not.
traffic-type (unicast; multicast; broadcast; Default: unicast,multicast,broadcast) Matching traffic type.
type (dst | src; Default: src) Lookup type of the isolation/leakage entry:
  • src - Entry applies to ingress packets of the ports.
  • dst - Entry applies to egress packets of the ports.
vlan-profile (community1 | community2 | isolated | promiscuous; Default: none) Matching VLAN isolation/leakage profile.

Trunking

Sub-menu: /interface ethernet switch trunk


The Trunking in the Cloud Router Switches provides static link aggregation groups with hardware automatic failover and load balancing. IEEE802.3ad and IEEE802.1ax compatible Link Aggregation Control Protocol is not supported yet. Up to 8 Trunk groups are supported with up to 8 Trunk member ports per Trunk group. CRS Port Trunking calculates transmit-hash based on all following parameters: L2 src-dst MAC + L3 src-dst IP + L4 src-dst Port.

Property Description
disabled (yes | no; Default: no) Enables or disables port trunking entry.
member-ports (ports) Member ports of the Trunk group.
name (string value; Default: trunkX) Name of the Trunk group.

Quality of Service

Shaper

Sub-menu: /interface ethernet switch shaper


Traffic shaping restricts the rate and burst size of the flow which is transmitted out from the interface. The shaper is implemented by a token bucket. If the packet exceeds the maximum rate or the burst size, which means no enough token for the packet, the packet is stored to buffer until there is enough token to transmit it.


Property Description
burst (integer; Default: 100k) Maximum data rate which can be transmitted while the burst is allowed.
disabled (yes | no; Default: no) Enables or disables traffic shaper entry.
meter-unit (bit | packet; Default: bit) Measuring units for traffic shaper rate.
port (port) Physical port for traffic shaper.
rate (integer; Default: 1M) Maximum data rate limit.
target (port | queueX | wrr-groupX; Default: port) Three levels of shapers are supported on each port (including CPU port):
  • Port level - Entry applies to port of the switch-chip.
  • WRR group level - Entry applies to one of the 2 Weighted Round Robin queue groups (wrr-group0, wrr-group1) on port.
  • Queue level - Entry applies to one of the 8 queues (queue0 - queue7) on port.

Ingress Port Policer

Sub-menu: /interface ethernet switch ingress-port-policer


Property Description
burst (integer; Default: 100k) Maximum data rate which can be transmitted while the burst is allowed.
disabled (yes | no; Default: no) Enables or disables ingress port policer entry.
meter-len (layer-1 | layer-2 | layer-3; Default: layer-1) Packet classification which sets the packet byte length for metering.
  • layer-1 - includes entire layer-2 frame + FCS + inter-