"network" test can mis-detect speed of optical devices

[Rod Smith]

The "network" test can mis-detect the speed of optical devices. In particular, in 18.04.2 regression testing on jehan, which has two 10Gbps SFP+ devices, the network test detected the top speed of one of those devices as 1Gbps, since it has a 1Gbps optical-to-copper SFP. See:

https://certification.canonical.com/hardware/201802-26109/submission/139090/

[Jeff Lane  (bladernr)]

I don't believe this is a bug in Checkbox, checkbox can only use the info that the kernel and system tools provide. This could be a driver issue, or perhaps a cabling issue. Could you

A: verify the cables are plugged into the correct places and check via ethtool what the connect speeds are for each port.
B: swap the cables on the SUT, and re-check ethtool output to see if port 0 STILL shows 1Gbit, or see if the 1Gb speed follows the cable.
C: Redo step A but on the 4.15 kernel rather than 4.18, to see if this is a regression of some sort.
D: Happy Fun Bonus: Try 16.04 GA to see if this is present on 4.4 as well (maybe not necessary, but worth looking into for due diligence).

I also added the kernel in case this is something we need to look at from a driver perspective.

Here is lshw for that machine that shows the kernel thinks that the port is 1Gbit:

        *-network:0
             description: Ethernet interface
             product: 82599ES 10-Gigabit SFI/SFP+ Network Connection
             vendor: Intel Corporation
             physical id: 0
             bus info: pci@0000:3b:00.0
             logical name: enp59s0f0
             version: 01
             serial: a8:1e:84:f2:96:c5
             size: 1Gbit/s
             capacity: 1Gbit/s
             width: 64 bits
             clock: 33MHz
             capabilities: pm msi msix pciexpress vpd bus_master cap_list rom ethernet physical tp 1000bt-fd autonegotiation
             configuration: autonegotiation=on broadcast=yes driver=ixgbe driverversion=5.1.0-k duplex=full firmware=0x800004e0, 1.808.0 ip=10.20.31.214 latency=0 link=yes multicast=yes port=twisted pair speed=1Gbit/s
             resources: irq:37 memory:b8580000-b85fffff ioport:7020(size=32) memory:b8604000-b8607fff memory:b8500000-b857ffff memory:b8300000-b83fffff memory:b8200000-b82fffff
        *-network:1
             description: Ethernet interface
             product: 82599ES 10-Gigabit SFI/SFP+ Network Connection
             vendor: Intel Corporation
             physical id: 0.1
             bus info: pci@0000:3b:00.1
             logical name: enp59s0f1
             version: 01
             serial: a8:1e:84:f2:96:c6
             size: 10Gbit/s
             capacity: 10Gbit/s
             width: 64 bits
             clock: 33MHz
             capabilities: pm msi msix pciexpress vpd bus_master cap_list rom ethernet physical fibre 10000bt-fd
             configuration: autonegotiation=off broadcast=yes driver=ixgbe driverversion=5.1.0-k duplex=full firmware=0x800004e0, 1.808.0 ip=10.20.32.10 latency=0 link=yes multicast=yes speed=10Gbit/s
             resources: irq:103 memory:b8480000-b84fffff ioport:7000(size=32) memory:b8600000-b8603fff memory:b8400000-b847ffff memory:b8100000-b81fffff memory:b8000000-b80fffff

[Ubuntu Kernel Bot (ubuntu-kernel-bot)]

This bug is missing log files that will aid in diagnosing the problem. While running an Ubuntu kernel (not a mainline or third-party kernel) please enter the following command in a terminal window:

apport-collect 1817412

and then change the status of the bug to 'Confirmed'.

If, due to the nature of the issue you have encountered, you are unable to run this command, please add a comment stating that fact and change the bug status to 'Confirmed'.

This change has been made by an automated script, maintained by the Ubuntu Kernel Team.

[Launchpad Janitor (janitor)]

[Expired for linux (Ubuntu) because there has been no activity for 60 days.]