Switch Port Attributes

Cumulus Linux exposes network interfaces for several types of physical and logical devices:

• lo is the network loopback device
• ethN are switch management ports (for out of band management only)
• swpN are switch front panel ports
• (optional) brN are bridges (IEEE 802.1Q VLANs)
• (optional) bondN are bonds (IEEE 802.3ad link aggregation trunks, or port channels)

Each physical network interface (port) has a number of configurable settings:

• Auto-negotiation
• Duplex Mode
• Link speed
• MTU (maximum transmission unit)
• FEC (forward error correction)

Most of these settings are configured automatically for you, depending upon your switch ASIC; however, you must always set MTU manually.

For NVIDIA Spectrum ASICs, the firmware configures FEC, link speed, duplex mode and auto-negotiation automatically, following a predefined list of parameter settings until the link comes up. You can disable FEC if necessary, which forces the firmware to not try any FEC options.

For Broadcom-based switches, enable auto-negotiation on each port. When enabled, Cumulus Linux automatically configures the best link parameter settings based on the module type (speed, duplex, auto-negotiation, and FEC, where supported).

This topic describes the auto-negotiation, link speed, duplex mode, MTU, and FEC settings and provides a table showing the default configuration for various port and cable types. Breakout port configuration, logical switch port limitations, and troubleshooting is also provided.

Auto-negotiation

By default on a Broadcom-based switch, auto-negotiation is disabled - except on 10G and 1000BASE-T fixed copper switch ports, where it is required for links to work. For RJ-45 SFP adapters, you need to manually configure the desired link speed and auto-negotiation as described in the default settings table below.

If you disable auto-negotiation later or never enable it, then you have to configure any settings that deviate from the port default - such as duplex mode, FEC, and link speed settings.

Depending upon the connector used for a port, enabling auto-negotiation also enables forward error correction (FEC), if the cable requires it (see the table below). The correct FEC mode is set based on the speed of the cable when auto-negotiation is enabled.

To configure auto-negotiation for a switch:

cumulus@switch:~$ net add interface swp1 link autoneg on
cumulus@switch:~$ net pending
cumulus@switch:~$ net commit

Port Speed and Duplex Mode

Cumulus Linux supports both half- and full-duplex configurations. Half-duplex is supported only with speeds of less than 1G.

Supported port speeds include 100M, 1G, 10G, 25G, 40G, 50G and 100G. In Cumulus Linux, you set the speed on Broadcom-based switch in Mbps, where the setting for 1G is 1000, 40G is 40000, and 100G is 100000.

You can configure ports to the following speeds (unless there are restrictions in the /etc/cumulus/ports.conf file of a particular platform).

To configure the port speed and duplex mode:

cumulus@switch:~$ net add interface swp1 link speed 10000
cumulus@switch:~$ net pending
cumulus@switch:~$ net commit

auto swp1
iface swp1
    link-speed 10000

cumulus@switch:~$ net add interface swp1 link speed 100 
cumulus@switch:~$ net add interface swp1 link duplex half
cumulus@switch:~$ net pending
cumulus@switch:~$ net commit

auto swp1
iface swp1
    link-speed 100
    link-duplex half

```
cumulus@switch:~$ sudo nano /etc/network/interfaces

auto swp1
iface swp1
    address 10.1.1.1/24
    link-speed 10000
    link-duplex full
```

MTU

Interface MTU applies to traffic traversing the management port, front panel or switch ports, bridge, VLAN subinterfaces, and bonds (both physical and logical interfaces). MTU is the only interface setting that you must set manually.

In Cumulus Linux, ifupdown2 assigns 1500 as the default MTU setting. On a Mellanox switch, the initial MTU value set by the driver is 9238. After you configure the interface, the default MTU setting is 1500.

To change the MTU setting, run the following commands:

cumulus@switch:~$ net add interface swp1 mtu 9000
cumulus@switch:~$ net pending
cumulus@switch:~$ net commit

auto swp1
iface swp1
    mtu 9000

Set a Policy for Global System MTU

For a global policy to set MTU, create a policy document (called mtu.json). For example:

cumulus@switch:~$ sudo cat /etc/network/ifupdown2/policy.d/mtu.json
{
  "address": {"defaults": { "mtu": "9216" }
            }
}

MTU for a Bridge

The MTU setting is the lowest MTU of any interface that is a member of the bridge (every interface specified in bridge-ports in the bridge configuration of the /etc/network/interfaces file). There is no need to specify an MTU on the bridge. Consider this bridge configuration:

auto bridge
iface bridge
    bridge-ports bond1 bond2 bond3 bond4 peer5
    bridge-vids 100-110
    bridge-vlan-aware yes

For a bridge to have an MTU of 9000, set the MTU for each of the member interfaces (bond1 to bond 4, and peer5), to 9000 at minimum.

When configuring MTU for a bond, configure the MTU value directly under the bond interface; the configured value is inherited by member links/slave interfaces. If you need a different MTU on the bond, set it on the bond interface, as this ensures the slave interfaces pick it up. There is no need to specify MTU on the slave interfaces.

VLAN interfaces inherit their MTU settings from their physical devices or their lower interface; for example, swp1.100 inherits its MTU setting from swp1. Therefore, specifying an MTU on swp1 ensures that swp1.100 inherits the MTU setting for swp1.

If you are working with VXLANs, the MTU for a virtual network interface (VNI must be 50 bytes smaller than the MTU of the physical interfaces on the switch, as those 50 bytes are required for various headers and other data. Also, consider setting the MTU much higher than 1500.

To show the MTU setting for an interface:

cumulus@switch:~$ net show interface swp1
    Name    MAC                Speed      MTU  Mode
--  ------  -----------------  -------  -----  ---------
UP  swp1    44:38:39:00:00:04  1G        1500  Access/L2

cumulus@switch:~$ ip link show dev swp1
3: swp1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP mode DEFAULT qlen 500
   link/ether 44:38:39:00:03:c1 brd ff:ff:ff:ff:ff:ff

FEC

Forward Error Correction (FEC) is an encoding and decoding layer that enables the switch to detect and correct bit errors introduced over the cable between two interfaces. Because 25G transmission speeds can introduce a higher than acceptable bit error rate (BER) on a link, FEC is required or recommended for 25G, 4x25G, and 100G link speeds.

For the link to come up, the two interfaces on each end must use the same FEC setting.

There are two FEC types:

• Reed Solomon (RS), IEEE 802.3 Clause 108 (CL108) on individual 25G channels and Clause 91 on 100G (4channels). This is the highest FEC algorithm, providing the best bit-error correction.
• Base-R (BaseR), Fire Code (FC), IEEE 802.3 Clause 74 (CL74). Base-R provides less protection from bit errors than RS FEC but adds less latency.

Cumulus Linux includes additional FEC options:

• Auto FEC instructs the hardware to select the best FEC. For copper DAC, FEC can be negotiated with the remote end. However, optical modules do not have auto-negotiation capability; if the device chooses a preferred mode, it might not match the remote end. This is the current default on a Spectrum switch.
• No FEC (no error correction is done). This is the current default on a Broadcom switch.

For 25G DAC, 4x25G Breakouts DAC and 100G DAC cables, the IEEE 802.3by specification creates 3 classes:

• CA-25G-L (long cables - achievable cable length of at least 5m) dB loss less or equal to 22.48. Requires RS FEC and expects BER of 10-5 or better with RS FEC enabled.
• CA-25G-S (short cables - achievable cable length of at least 3m) dB loss less or equal to 16.48. Requires Base-R FEC and expects BER of 10-8 or better with Base-R FEC enabled.
• CA-25G-N (no FEC - achievable cable length of at least 3m) dB loss less or equal to 12.98. Does not require FEC. Expects BER 10-12 or better with no FEC.

The IEEE classification is based on various dB loss measurements and minimum achievable cable length. You can build longer and shorter cables if they comply to the dB loss and BER requirements.

If a cable is manufactured to CA-25G-S classification and FEC is not enabled, the BER might be unacceptable in a production network. It is important to set the FEC according to the cable class (or better) to have acceptable bit error rates. See Determining Cable Class below.

You can check bit errors using cl-netstat (RX_ERR column) or ethtool -S (HwIfInErrors counter) after a large amount of traffic has passed through the link. A non-zero value indicates bit errors. Expect error packets to be zero or extremely low compared to good packets. If a cable has an unacceptable rate of errors with FEC enabled, replace the cable.

For 25G, 4x25G Breakout, and 100G Fiber modules and AOCs, there is no classification of 25G cable types for dB loss, BER or length. FEC is recommended but might not be required if the BER is low enough.

Determine Cable Class of 100G and 25G DACs

You can determine the cable class for 100G and 25G DACs from the Extended Specification Compliance Code field (SFP28: 0Ah, byte 35, QSFP28: Page 0, byte 192) in the cable EEPROM programming.

For 100G DACs, most manufacturers use the 0x0Bh 100GBASE-CR4 or 25GBASE-CR CA-L value (the 100G DAC specification predates the IEEE 802.3by 25G DAC specification). RS FEC is the expected setting for 100G DAC but might not be required with shorter or better cables.

You can determine for which grade the manufacturer has designated the cable as follows.

For the SFP28 DAC, run the following command:

cumulus@switch:~$ sudo ethtool -m swp35 hex on | grep 0020 | awk '{ print $6}'
0c

The values at location 0x0024 are:

• 0x0b : CA-L (long cable - RS FEC required)
• 0x0c : CA-S (short cable - Base-R or better FEC required)
• 0x0d : CA-N (no FEC required)

For the QSFP28 DAC, run the following command:

cumulus@switch:~$ sudo ethtool -m swp51s0 hex on | grep 00c0 | awk '{print $2}'
0b

The values at 0x00c0 are:

• 0x0b : CA-L (long cable - RS FEC required) or 100G CR4
• 0x0c : CA-S (short cable - Base-R or better FEC required)
• 0x0d : CA-N (no FEC required)

In each example below, the Compliance field is derived using the method described above and is not visible in the ethool -m output.

3meter cable that does not require FEC 
(CA-N)  
Cost: More expensive  
Cable size: 26AWG (Note that AWG does not necessarily correspond to overall dB loss or BER performance)  
Compliance Code: 25GBASE-CR CA-N

3meter cable that requires Base-R FEC 
(CA-S)  
Cost: Less expensive  
Cable size: 26AWG  
Compliance Code: 25GBASE-CR CA-S

When in doubt, consult the manufacturer directly to determine the cable classification.

Spectrum ASIC FEC Behavior

The firmware in a Spectrum ASIC applies FEC configuration to 25G and 100G cables based on the cable type and whether the peer switch also has a Spectrum ASIC.

When the link is between two switches with Spectrum ASICs:

• For 25G optical modules, the Spectrum ASIC firmware chooses Base-R/FC-FEC.
• For 25G DAC cables with attenuation less or equal to 16db, the firmware chooses Base-R/FC-FEC.
• For 25G DAC cables with attenuation higher than 16db, the firmware chooses RS-FEC.
• For 100G cables/modules, the firmware chooses RS-FEC.

When linking to a non-Spectrum peer, the firmware lets the peer decide. The Spectrum ASIC supports RS-FEC (for both 100G and 25G), Base-R/FC-FEC (25G only), or no-FEC (for both 100G and 25G).

How Does Cumulus Linux use FEC?

This depends upon the make of the switch you are using.

A Spectrum switch enables FEC automatically when it powers up; that is, the setting is fec auto. The port firmware tests and determines the correct FEC mode to bring the link up with the neighbor. It is possible to get a link up to a Spectrum switch without enabling FEC on the remote device as the switch eventually finds a working combination to the neighbor without FEC.

On a Broadcom switch, Cumulus Linux does not enable FEC by default; that is, the setting is fec off. Configure FEC explicitly to match the configured FEC on the link neighbor. On 100G DACs, you can configure link-autoneg so that the port attempts to negotiate FEC settings with the remote peer.

The following sections describe how to show the current FEC mode, and to enable and disable FEC.

Show the Current FEC Mode

Cumulus Linux returns different output for the ethtool --show-fec command, depending upon whether you are using a Broadcom or Mellanox Spectrum switch.

On a Broadcom switch, the --show-fec output tells you exactly what you configured, even if the link is down due to a FEC mismatch with the neighbor.

On a Spectrum switch, the --show-fec output tells you the current active state of FEC only if the link is up; that is, if the FEC modes matches that of the neighbor. If the link is not up, the value displays None, which is not valid.

To show the FEC mode currently enabled on a given switch port, run the ethtool --show-fec <interface> command.

cumulus@switch:~$ sudo ethtool --show-fec swp1
FEC parameters for swp1:
Configured FEC encodings: Auto
Active FEC encoding: Off

Enable or Disable FEC

To enable Reed Solomon (RS) FEC on a link:

cumulus@switch:~$ sudo net add interface swp1 link fec rs
cumulus@switch:~$ sudo net pending
cumulus@switch:~$ sudo net commit

To enable Base-R/FireCode FEC on a link:

cumulus@switch:~$ sudo net add interface swp1 link fec baser
cumulus@switch:~$ sudo net pending
cumulus@switch:~$ sudo net commit

To enable FEC with Auto-negotiation:

cumulus@switch:~$ sudo net add interface swp1 link autoneg on
cumulus@switch:~$ sudo net pending
cumulus@switch:~$ sudo net commit

To show the FEC and auto-negotiation settings for an interface, run the following command:

cumulus@switch:~$ sudo ethtool swp1 | egrep 'FEC|auto'
Supports auto-negotiation: Yes
Supported FEC modes: RS
Advertised auto-negotiation: Yes
Advertised FEC modes: RS
Link partner advertised auto-negotiation: Yes
Link partner advertised FEC modes: Not reported

To disable FEC on a link:

cumulus@switch:~$ sudo net add interface swp1 link fec off
cumulus@switch:~$ sudo net pending
cumulus@switch:~$ sudo net commit

Interface Configuration Recommendations for Broadcom Platforms

The recommended configuration for each type of interface is described in the following table. These are the link settings that are applied to the port hardware when auto-negotiation is enabled on a Broadcom-based switch. If further troubleshooting is required to bring a link up, use the table below as a guide to set the link parameters.

Except as noted below, the settings for both sides of the link are expected to be the same.

Speed	Auto-negotiation	FEC Setting	Manual Configuration Examples	Notes
100BASE-T (RJ-45 SFP adapter)	Off	N/A	NCLU commands $ net add interface swp1 link speed 100 $ net add interface swp1 link autoneg off Configuration in /etc/network/interfaces auto swp1 iface swp1    link-autoneg off    link-speed 100	The module has two sets of electronics: the port side, which communicates with the switch ASIC and the RJ-45 adapter side. Auto-negotiation is always used on the RJ-45 adapter side of the link by the PHY built into the module. This is independent of the switch setting. Set auto-negotiation to off. Auto-negotiation must be enabled on the server side in this scenario.
100BASE-T on a 1G fixed copper port	On	N/A	NCLU commands $ net add interface swp1 link speed 100 $net add interface swp1 link autoneg on Configuration in /etc/network/interfaces auto swp1 iface swp1    ink-autoneg on    link-speed 100	10M or 100M speeds are possible with auto-negotiation off on both sides. Testing on an Edgecore AS4610-54P showed the ASIC reporting auto-negotiation as on. Power over Ethernet might require auto-negotiation to be on.
1000BASE-T (RJ-45 SFP adapter)	Off	N/A	NCLU commands $ net add interface swp1 link speed 1000 $ net add interface swp1 link autoneg off Configuration in /etc/network/interfaces auto swp1 iface swp1    link-autoneg off    link-speed 1000	The module has two sets of electronics: the port side, which communicates with the switch ASIC and the RJ-45 side. Auto-negotiation is always used on the RJ-45 side of the link by the PHY built into the module. This is independent of the switch setting. Set auto-negotiation to off. Auto-negotiation must be enabled on the server side.
1000BASE-T on a 1G fixed copper port	On	N/A	NCLU commands $ net add interface swp1 link speed 1000 $ net add interface swp1 link autoneg on Configuration in /etc/network/interfaces auto swp1 iface swp1    link-autoneg on    link-speed 1000
1000BASE-T on a 10G fixed copper port	On	N/A	NCLU commands $ net add interface swp1 link speed 1000 $ net add interface swp1 link autoneg on Configuration in /etc/network/interfaces auto swp1 iface swp1    link-autoneg on    link-speed 1000
1000BASE-SX 1000BASE-LX (1G Fiber)	Recommended On	N/A	NCLU commands $ net add interface swp1 link speed 1000 $ net add interface swp1 link autoneg on Configuration in /etc/network/interfaces auto swp1 iface swp1    link-autoneg on    link-speed 1000	Without auto-negotiation, the link stays up when there is a single fiber break. See the limitation discussed in 10G and 1G SFPs Inserted in a 25G Port below.
10GBASE-T (RJ-45 SFP Module)	Off	N/A	NCLU commands $ net add interface swp1 link speed 10000 $ net add interface swp1 link autoneg off Configuration in /etc/network/interfaces auto swp1 iface swp1    link-autoneg off    link-speed 10000	The module has two sets of electronics - the port side, which communicates to the switch ASIC and the RJ-45 side. Auto-negotiation is always used on the RJ-45 side of the link by the PHY built into the module. This is independent of the switch setting. Set link-autoneg to off. Auto-negotiation needs to be enabled on the server side.
10GBASE-T fixed copper port	On	N/A	NCLU commands $ net add interface swp1 link speed 10000 $ net add interface swp1 link autoneg on Configuration in /etc/network/interfaces auto swp1 iface swp1    link-autoneg on    link-speed 10000
10GBASE-CR 10GBASE-LR 10GBASE-SR 10G AOC	Off	N/A	NCLU commands $ net add interface swp1 link speed 10000 $ net add interface swp1 link autoneg off Configuration in /etc/network/interfaces auto swp1 iface swp1    link-autoneg off    link-speed 10000
40GBASE-CR4	Recommended On	Disable	NCLU commands $ net add interface swp1 link speed 40000 $ net add interface swp1 link autoneg on Configuration in /etc/network/interfaces auto swp1 iface swp1    link-autoneg on    link-speed 40000	40G standards mandate auto-negotiation be enabled for DAC connections.
40GBASE-SR4 40GBASE-LR4 40G AOC	Off	Disable	NCLU commands $ net add interface swp1 link speed 40000 $ net add interface swp1 link autoneg off Configuration in /etc/network/interfaces auto swp1 iface swp1    link-autoneg off    link-speed 40000
100GBASE-CR4	On	auto-negotiated	NCLU commands $ net add interface swp1 link speed 100000 $ net add interface swp1 link autoneg on Configuration in /etc/network/interfaces auto swp1 iface swp1    link-autoneg on    link-speed 100000
100GBASE-SR4 100G AOC	Off	RS	NCLU commands $ net add interface swp1 link speed 100000 $ net add interface swp1 link autoneg off $ net add interface swp1 link fec rs Configuration in /etc/network/interfaces auto swp1 iface swp1    link-autoneg off    link-speed 100000    link-fec rs
100GBASE-LR4	Off	None	NCLU commands $ net add interface swp1 link speed 100000 $ net add interface swp1 link autoneg off $ net add interface swp1 link fec off Configuration in /etc/network/interfaces auto swp1 iface swp1    link-autoneg off    link-speed 100000    link-fec off
25GBASE-CR	On	auto-negotiated	NCLU commands $ net add interface swp1 link speed 25000 $ net add interface swp1 link autoneg on Configuration in /etc/network/interfaces auto swp1 iface swp1    link-autoneg on    link-speed 25000	Tomahawk predates 802.3by. It does not support RS FEC or auto-negotiation of RS FEC on a 25G port or subport. It does support Base-R FEC.
25GBASE-SR	Off	RS	NCLU commands $ net add interface swp1 link speed 25000 $ net add interface swp1 link autoneg off $ net add interface swp1 link fec rs Configuration in /etc/network/interfaces auto swp1 iface swp1    link-autoneg off    link-speed 25000    link-fec rs	Tomahawk predates 802.3by and does not support RS FEC on a 25G port or subport; however it does support Base-R FEC. The configuration for Base-R FEC is as follows: NCLU commands $ net add interface swp1 link speed 25000 $ net add interface swp1 link autoneg off $ net add interface swp1 link fec baser Configuration in /etc/network/interfaces auto swp1 iface swp1    link-autoneg off    link-speed 25000    link-fec baser Configure FEC to the setting that the cable requires.
25GBASE-LR	Off	None	NCLU commands $ net add interface swp1 link speed 25000 $ net add interface swp1 link autoneg off $ net add interface swp1 link fec off Configuration in /etc/network/interfaces auto swp1 iface swp1    link-autoneg off     link-speed 25000    link-fec off

Default Policies for Interface Settings

Instead of configuring settings for each individual interface, you can specify a policy for all interfaces on a switch or tailor custom settings for each interface. Create a file in /etc/network/ifupdown2/policy.d/ and populate the settings accordingly. The following example shows a file called address.json.

cumulus@switch:~$ cat /etc/network/ifupdown2/policy.d/address.json
{
    "ethtool": {
        "defaults": {
            "link-duplex": "full"
        },
        "iface_defaults": {
            "swp1": {
                "link-autoneg": "on",
                "link-speed": "1000"
          },
            "swp16": {
                "link-autoneg": "off",
                "link-speed": "10000"
            },
            "swp50": {
                "link-autoneg": "off",
                "link-speed": "100000",
                "link-fec": "rs"
            }
        }
    },
    "address": {
        "defaults": { "mtu": "9000" },
        "iface_defaults": {
            "eth0": {"mtu": "1500"}
        }
    }
}

Breakout Ports

Cumulus Linux lets you:

• Break out 100G switch ports into 2x50G, 4x25G, or 4x10G with breakout cables.
• Break out 40G switch ports into four separate 10G ports for use with breakout cables.
• Combine (aggregate or gang) four 10G switch ports into one 40G port for use with a breakout cable (not to be confused with a bond).

Configure a Breakout Port

To configure a breakout port:

cumulus@switch:~$ net add interface swp1 breakout 4x25G
cumulus@switch:~$ net pending
cumulus@switch:~$ net commit

cumulus@switch:~$ net add interface swp2 breakout disabled
cumulus@switch:~$ net pending
cumulus@switch:~$ net commit

cumulus@switch:~$ cat /etc/network/interfaces
...
auto swp1s0
iface swp1s0

auto swp1s1
iface swp1s1

auto swp1s2
iface swp1s2

auto swp1s3
iface swp1s3
...

cumulus@switch:~$ sudo cat /etc/network/interfaces
...
auto swp1s0
iface swp1s0

auto swp1s1
iface swp1s1

auto swp1s2
iface swp1s2

auto swp1s3
iface swp1s3
...

Remove a Breakout Port

To remove a breakout port:

Combine Four 10G Ports into One 40G Port

You can gang (combine) four 10G ports into one 40G port for use with a breakout cable, provided you follow these requirements:

• You must gang four 10G ports in sequential order. For example, you cannot gang swp1, swp10, swp20 and swp40 together.
• The ports must be in increments of four, with the starting port being swp1 (or swp5, swp9, or so forth); so you cannot gang swp2, swp3, swp4 and swp5 together.

cumulus@switch:~$ net add int swp1-4 breakout /4
cumulus@switch:~$ net pending
cumulus@switch:~$ net commit

# SFP+ ports#
# <port label 1-48> = [10G|40G/4]
1=40G/4
2=40G/4
3=40G/4
4=40G/4
5=10G

# SFP+ ports#
# <port label 1-48> = [10G|40G/4]
1=40G/4
2=40G/4
3=40G/4
4=40G/4
5=10G

cumulus@switch:~$ sudo systemctl restart switchd.service

Logical Switch Port Limitations

100G and 40G switches can support a certain number of logical ports, depending on the manufacturer; these include:

• Mellanox SN2700, SN2700B, SN2410, and SN2410B switches
• Switches with Broadcom Tomahawk, Trident II, Trident II+, and Trident3 chipsets

Before you configure any logical/unganged ports on a switch, check the limitations listed in /etc/cumulus/ports.conf; this file is specific to each manufacturer.

The following example shows the logical port limitation provided in the Dell Z9254F-ON ports.conf file. The maximum number of ports for this switch is 128.

# ports.conf --
#
#   configure port speed, aggregation, and subdivision.
#
# The Dell Z9264F has:
#      64 QSFP28 ports numbered 1-64
#         These ports are configurable as 100G, 50G, 40G, or split into
#         2x50G, 4x25G, or 4x10G ports.
#
# NOTE:  You must restart switchd for any changes to take effect.
# Only "odd-numbered " port can be split into 4 interfaces and if an odd-numbered
# port is split in a 4X configuration, the port adjacent to it (even-numbered port)
# has to be set to "disabled " in this file. When splitting a port into two
# interfaces, like 2x50G, it is NOT required that the adjacent port be
# disabled. For example, when splitting port 11 into 4 10G interfaces, port
# 12 must be configured as "disabled" like this:
#
#   11=4x10G
#   12=disabled

# QSFP28 ports
#
# <port label> = [100G|50G|40G|2x50G|4x25G|4x10G|disabled]

Mellanox SN2700 and SN2700B switches have a limit of 64 logical ports in total. However, the logical ports must be configured in a specific way. See the note above.

Verification and Troubleshooting Commands

Statistics

To show high-level interface statistics, run the net show interface command:

cumulus@switch:~$ net show interface swp1

    Name    MAC                Speed      MTU  Mode
--  ------  -----------------  -------  -----  ---------
UP  swp1    44:38:39:00:00:04  1G        1500  Access/L2

Vlans in disabled State
-------------------------
br0

Counters      TX    RX
----------  ----  ----
errors         0     0
unicast        0     0
broadcast      0     0
multicast      0     0

LLDP
------  ----  ---------------------------
swp1    ====  44:38:39:00:00:03(server01)

To show low-level interface statistics, run the following ethtool command:

cumulus@switch:~$ sudo ethtool -S swp1
NIC statistics:
      HwIfInOctets: 21870
      HwIfInUcastPkts: 0
      HwIfInBcastPkts: 0
      HwIfInMcastPkts: 243
      HwIfOutOctets: 1148217
      HwIfOutUcastPkts: 0
      HwIfOutMcastPkts: 11353
      HwIfOutBcastPkts: 0
      HwIfInDiscards: 0
      HwIfInL3Drops: 0
      HwIfInBufferDrops: 0
      HwIfInAclDrops: 0
      HwIfInBlackholeDrops: 0
      HwIfInDot3LengthErrors: 0
      HwIfInErrors: 0
      SoftInErrors: 0
      HwIfOutErrors: 0
      HwIfOutQDrops: 0
      HwIfOutNonQDrops: 0
      SoftOutErrors: 0
      SoftOutDrops: 0
      SoftOutTxFifoFull: 0
      HwIfOutQLen: 0

Query SFP Port Information

To verify SFP settings, run the ethtool -m command. The following example shows the vendor, type and power output for the swp4 interface.

cumulus@switch:~$ sudo ethtool -m swp4 | egrep 'Vendor|type|power\s+:'
        Transceiver type                          : 10G Ethernet: 10G Base-LR
        Vendor name                               : FINISAR CORP.
        Vendor OUI                                : 00:90:65
        Vendor PN                                 : FTLX2071D327
        Vendor rev                                : A
        Vendor SN                                 : UY30DTX
        Laser output power                        : 0.5230 mW / -2.81 dBm
        Receiver signal average optical power     : 0.7285 mW / -1.38 dBm

Caveats and Errata

Port Speed and the ifreload -a Command

When configuring port speed or break outs in the /etc/cumulus/ports.conf file, you need to run the ifreload -a command to reload the configuration after restarting switchd in the following cases:

• If you configure, or configure then remove, the port speed in the /etc/cumulus/ports.conf file and you also set or remove the speed on the same physical port or breakouts of that port in the /etc/network/interfaces file since the last time you restarted switchd.
• If you break out a switch port or remove a break out port and the port speed is set in both the /etc/cumulus/ports.conf file and the /etc/network/interfaces file.

Port Speed Configuration

If you change the port speed in the /etc/cumulus/ports.conf file but the speed is also configured for that port in the /etc/network/interfaces file, after you edit the /etc/cumulus/ports.conf file and restart switchd, you must also run the ifreload -a command so that the /etc/network/interfaces file is also updated with your change.

10G and 1G SFPs Inserted in a 25G Port

For 10G and 1G SFPs inserted in a 25G port on a Broadcom platform, you must configure the four ports in the same core to be 10G. Each set of four 25G ports are controlled by a single core; therefore, each core must run at the same clock speed. The four ports must be in sequential order; for example, swp1, swp2, swp3, and swp4, unless a particular core grouping is specified in the /etc/cumulus/ports.conf file.

1. Edit the /etc/cumulus/ports.conf file and configure the four ports to be 10G. 1G SFPs are clocked at 10G speeds; therefore, for 1G SFPs, the /etc/cumulus/ports.conf file entry must also specify 10G. Currently you cannot use NCLU commands for this step.

   ...
   # SFP28 ports
   #
   # <port label 1-48> = [25G|10G|100G/4|40G/4]
   1=25G
   2=25G
   3=25G
   4=25G
   5=10G
   6=10G
   7=10G
   8=10G
   9=25G
   ...
   

   You cannot use ethtool -s speed XX (or ifreload -a after setting the speed in the /etc/network/interfaces file) to change the port speed unless the four ports in a core group are already configured to 10G and switchd has been restarted. If the ports are still in 25G mode, using ethtool or ifreload to change the speed to 10G or 1G returns an error (and a return code of 255).

   If you change the speed with ethtool to a setting already in use in the /etc/cumulus/ports.conf file, ethtool (and ifreload -a) do not return an error and no changes are made.

2. Restart switchd.

3. If you want to set the speed of any SFPs to 1G, set the port speed to 1000 Mbps using NCLU commands; this is not necessary for 10G SFPs. You don’t need to set the port speed to 1G for all four ports. For example, if you intend only for swp5 and swp6 to use 1G SFPs, do the following:

   cumulus@switch:~$ net add interface swp5-swp6 link speed 1000
   cumulus@switch:~$ net pending
   cumulus@switch:~$ net commit
   

Timeout Error on Quanta LY8 and LY9 Switches

On Quanta T5048-LY8 and T3048-LY9 switches, an Operation timed out error occurs when you remove and reinsert a QSFP module.

You cannot remove the QSFPx2 module while the switch is powered on; it is not hot-swappable. However, if an Operation timed out error occurs, restart switchd to bring the link up. Be aware that this disrupts your network.

On the T3048-LY9, run the following commands:

cumulus@switch:~$ sudo echo 0 > qsfpd_power_enable/value
cumulus@switch:~$ sudo rmmod quanta_ly9_rangeley_platform
cumulus@switch:~$ sudo modprobe quanta_ly9_rangeley_platform
cumulus@switch:~$ sudo systemctl restart switchd.service

On the T5048-LY8, run the following commands:

cumulus@switch:~$ sudo echo 0 > qsfpd_power_enable/value
cumulus@switch:~$ sudo systemctl restart switchd.service

swp33 and swp34 Disabled on Some Switches

The front SFP+ ports (swp33 and swp34) are disabled in Cumulus Linux on the following switches:

• Dell Z9100-ON
• Penguin Arctica 3200-series switches (the 3200C, 3200XL and 3200XLP)
• Supermicro SSE-C3632S

These ports appear as disabled in the /etc/cumulus/ports.conf file.

200G Interfaces on the Dell S5248F Switch

On the Dell S5248F switch, the 2x200G QSFP-DD interfaces labeled 49/50 and 51/52 are not supported natively at 200G speeds. The interfaces are supported with 100G cables; however, you can only use one 100G from each QSFP-DD port. The upper QSFP-DD port is named swp49 and the lower QSFP-DD port is named swp52.

QSFP+ Ports on the Dell S5232F Switch

Cumulus Linux does not support the 2x10G QSFP+ ports on the Dell S5232F switch.

QSFP+ Ports on the Dell S4148T Switch

On the Dell S4148T switch, the two QSFP+ ports are set to disabled by default and the four QSFP28 ports are configured for 100G. The following example shows the default settings in the /etc/cumulus/ports.conf file for this switch:

cumulus@switch:~$ sudo cat /etc/cumulus/ports.conf
...
# QSFP+ ports
#
# <port label 27-28> = [4x10G|40G]
27=disabled
28=disabled
# QSFP28 ports
#
# <port label 25-26, 29-30> = [4x10G|4x25G|2x50G|40G|50G|100G]
25=100G
26=100G
29=100G
30=100G

To enable the two QSFP+ ports, you must configure all four QSFP28 ports for either 40G or 4x10G. You cannot use either of the QSFP+ ports if any of the QSFP28 ports are configured for 100G.

The following example shows the /etc/cumulus/ports.conf file with all four QSFP28 ports configured for 40G and both QSFP+ ports enabled:

cumulus@switch:~$ sudo cat /etc/cumulus/ports.conf
...
# QSFP+ ports
#
# <port label 27-28> = [4x10G|40G]
27=40G
28=40G
# QSFP28 ports
#
# <port label 25-26, 29-30> = [4x10G|4x25G|2x50G|40G|50G|100G]
25=40G
26=40G
29=40G
30=40G

1000BASE-T Modules Not Supported on Certain Edgecore and Cumulus Express Switches

1000BASE-T modules are not supported on the following switches:

• Cumulus Express CX-5148-S
• Cumulus Express CX-8132-S
• Edgecore AS7326-56X
• Edgecore AS7726-32X

Mellanox SN2100 Switch and eth0 Link Speed

After rebooting the Melllanox SN2100 switch, eth0 always has a speed of 100Mb/s. If you bring the interface down and then back up again, the interface negotiates 1000Mb. This only occurs the first time the interface comes up.

To work around this issue, add the following commands to the /etc/rc.local file to flap the interface automatically when the switch boots:

modprobe -r igb
sleep 20
modprobe igb

Link Speed on the EdgeCore AS7326-56X Switch

On the EdgeCore AS7326-56X switch, all four switch ports in each port group must be set to the same link speed; otherwise, the links do not come up. These ports are set to 25G by default, but can also be set to 10G. The port groups on this switch are as follows, where each row is a port group:

• 1 2 3 6*
• 4 5 7* 9
• 8 10 11* 12
• 13 14 15 18*
• 16 17 19* 21
• 20 22 23* 24
• 25 26 27 30*
• 28 29 31* 33
• 32 34 35* 36
• 37 38 39 42*
• 40* 41 43 45
• 44* 46 47 48

For example, if you configure port 19 for 10G, you must also configure ports 16, 17 and 21 for 10G.

Additionally, you can gang each port group together as a 100G or 40G port. When ganged together, one port (based on the arrangement of the ports) is designated as the gang leader. This port’s number is used to configure the ganged ports and is marked with an asterisk (*) above.

Link Speed on the Lenovo NE2572O Switch

The Lenovo NE2572O switch has external retimers on swp1 through swp8. Currently, these ports only support a speed of 25G.

Link Speed and Auto-negotiation on Switches with SOL

The following switches that use Serial over LAN technology (SOL) do not support eth0 speed or auto-negotiation changes:

• EdgeCore AS7816-64X
• Penguin Arctica 4804ip
• Penguin Arctica NX3200c
• Penguin Arctica NX4808xxv

Delay in Reporting Interface as Operational Down

When you remove two transceivers simultaneously from a switch, both interfaces show the carrier down status immediately. However, it takes one second for the second interface to show the operational down status. In addition, the services on this interface also take an extra second to come down.

Mellanox Spectrum-2 and Tomahawk-based Switches Support Different FEC Modes

The Mellanox Spectrum-2 (25G) switch only supports RS FEC. The Tomahawk-based switch only supports BASE-R FEC. These two switches do not share compatible FEC modes and do not interoperate reliably.

Maverick Switches with Modules that Don’t Support Auto-negotiation

On a Maverick switch, if auto-negotiation is configured on a 10G interface and the installed module does not support auto-negotiation (for example, 10G DAC, 10G Optical, 1G RJ45 SFP), the link breaks. To work around this issue, disable auto-negotiation on interfaces where it is not supported.

Dell Z9264F-ON 10G Interfaces are Unsupported

The Dell Z9264F-ON has 64x100G + 2x 10G SFP+ ports. The 2x 10G SFP+ ports are not supported in Cumulus Linux.

Related Information

• Debian - Network Configuration