0

I have two gigabit devices connected to a gigabit switch using cat 5e ethernet cables, but the wirings on the RJ45 connectors do not respect any of the established color standards, because they were made by a newcomer.

The connection between the devices is working, but only at 100Mb/s speed. When I check the NIC details using ethtool, the devices properly show up as gigabit capable.

Settings for eth0:
    Supported ports: [ TP ]
    Supported link modes:   10baseT/Half 10baseT/Full
                            100baseT/Half 100baseT/Full
                            1000baseT/Full
    Supported pause frame use: Symmetric
    Supports auto-negotiation: Yes
    Supported FEC modes: Not reported
    Advertised link modes:  10baseT/Half 10baseT/Full
                            100baseT/Half 100baseT/Full
                            1000baseT/Full
    Advertised pause frame use: Symmetric
    Advertised auto-negotiation: Yes
    Advertised FEC modes: Not reported
    Speed: 100Mb/s
    Duplex: Full
    Port: Twisted Pair
    PHYAD: 1
    Transceiver: internal
    Auto-negotiation: on
    MDI-X: on (auto)
    Supports Wake-on: pumbg
    Wake-on: g
    Current message level: 0x00000007 (7)
                   drv probe link
    Link detected: yes

If I unplug and replug the cables, I see the Speed being initially set to 1Gb/s, but then after a couple of seconds being downgraded to 100Mb/s.


[1469117.206807] igb 0000:01:00.0 eth0: igb: eth0 NIC Link is Down
[1469117.206899] br0: port 2(eth0) entered disabled state
[1469121.997530] igb 0000:01:00.0 eth0: igb: eth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: RX
[1469124.437297] igb 0000:01:00.0: exceed max 2 second
[1469124.437568] igb 0000:01:00.0 eth0: igb: eth0 NIC Link is Down
[1469125.877831] igb 0000:01:00.0 eth0: igb: eth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: RX
[1469128.307603] igb 0000:01:00.0: exceed max 2 second
[1469128.307875] igb 0000:01:00.0 eth0: igb: eth0 NIC Link is Down
[1469129.748145] igb 0000:01:00.0 eth0: igb: eth0 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: RX
[1469132.178032] igb 0000:01:00.0: exceed max 2 second
[1469132.178301] igb 0000:01:00.0 eth0: igb: eth0 NIC Link is Down
[1469132.558348] igb 0000:01:00.0 eth0: igb: eth0 NIC Link is Up 100 Mbps Full Duplex, Flow Control: RX
[1469132.558572] br0: port 2(eth0) entered blocking state
[1469132.558574] br0: port 2(eth0) entered forwarding state

The exact same problem is happening with identical wired cables and the same intel NICs, on other two devices connected to the very same switch.

# ip -d link show dev eth0
2: eth0: <BROADCAST,MULTICAST,PROMISC,UP,LOWER_UP> mtu 1500 qdisc mq master br0 state UP mode DEFAULT group default qlen 1000
    link/ether XX:XX:XX:XX:XX:XX brd ff:ff:ff:ff:ff:ff promiscuity 2 minmtu 68 maxmtu 9216
    bridge_slave state forwarding priority 32 cost 19 hairpin off guard off root_block off fastleave off learning on flood on port_id 0x8002 port_no 0x2 designated_port 32770 designated_cost 0 designated_bridge 0000.80:c6:ca:1:56:4e designated_root 0000.80:c6:ca:1:56:4e hold_timer    0.00 message_age_timer    0.00 forward_delay_timer    0.00 topology_change_ack 0 config_pending 0 proxy_arp off proxy_arp_wifi off mcast_router 1 mcast_fast_leave off mcast_flood on neigh_suppress off group_fwd_mask 0x0 group_fwd_mask_str 0x0 vlan_tunnel off isolated off addrgenmode none numtxqueues 8 numrxqueues 8 gso_max_size 65536 gso_max_segs 65535

If I replace the cable with another cat 5e of another brand, having proper color wiring, the speed is stable at 1Gb/s.

Can the wrong color wiring be causing this issue, maybe do to EM interferences in the cable itself?

Or is it more likely to be the first cable that is low quality or so?

2 Answers 2

2

The color coding scheme isn't technically required but it should be observed at all times for practical purposes - you wouldn't want to troubleshoot an installation where no reasonable scheme is used.

However, you do need to stick to correct pairing. The twisted pairs from the cable must be kept together, twist rates and directions must not be changed and you must connect the pairs 1-2,3-6,4-5 and 7-8, without reordering them between connectors.

When a four-paired cable causes gigabit ports to fall back to 100 Mbit/s there's usually (at least) one pair at fault. There's no mandatory fallback in that situation but a great number of ports use Broadcom's Ethernet@Wirespeed feature that is able to fall back in case of pair failures in PMA lane-level linking (on pairs 4-5 or 7-8).

they were made by a newcomer

Cabling deployment should never be done without proper training. Also, you should always certify an installation before production, according to category. Cat 6A installation is pretty demanding, even for an experienced installer.

Troubleshooting a faulty installation can cost a multiple of what you might have saved by doing it the cheap way. I'd have the termination redone properly and certified afterwards.

0

Depending on how it's wired, it could absolutely cause problems. The pairs (orange/orange-white, blue/blue-white, etc) are designed to work together as a balanced pair. These wires are twisted around each other in the cable. Different colors are twisted at different rates to reduce noise and crosstalk.

100base-t only uses 2 pairs (4 wires), 1GB uses all 4 pairs (2 channels TX, 2 channels RX)

If your wiring is, for example, using orange paired with AYTHING other than orange-white for a give channel, you will run into issues with noise, crosstalk, etc.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.