You are not logged in.
System in question has 3 drives with 3 separate installs/setups of Archlinux....
1 ( my primary everyday boot too ) has i3 and KDE setup.... I have noticed in the past couple of days that my internet speed ( looking at results from speedtest-cli ) has been limited to 25 Mbit/s so at first I thought I had a router or service issue, however all other systems on my network report the expected 200+ Mbit/s range.
So when I boot into any of the other two instances of Arch on my System in question.... speedtest results are good ... they are in the expected 200 Mbit/s range...
But for some reason when I boot into my everyday drive/instance .... it will not report past 25 Mbit/s ..... My local DHCP server is assigning a reserved address based on the NIC and I get the same IP, DHCP settings etc... between any drive or instance that I boot to on this system.
All three drives in terms of network setup are using NetworkManager and.... all three separate instances have the same updated version etc...
I am stumped currently trying to figure out what in the world could be limiting or restricting my internet speed when booting to this instance.
Between all instances .... nmcli active/profile etc... output is identical ... so configuration is the same... IP is the same, the routes are the same... Yet this one instance will not get past 25 Mbit/s on speedtest
Local network performance between this system and another node appears to be normal... but I don't know off hand if there is a point to point test like speedtest for a local system.. file transfer and rsync are normal and expected.
Thanks in advance for any suggestions in areas to look?
Offline
Wired or Wireless?
What are the chipsets used in the NIC?
What is the output of ip link ??
Nothing is too wonderful to be true, if it be consistent with the laws of nature -- Michael Faraday
Sometimes it is the people no one can imagine anything of who do the things no one can imagine. -- Alan Turing
---
How to Ask Questions the Smart Way
Offline
Wired
03:00.0 Ethernet controller: Qualcomm Atheros Killer E2400 Gigabit Ethernet Controller (rev 10)
Subsystem: Micro-Star International Co., Ltd. [MSI] Killer E2400 Gigabit Ethernet Controller
Flags: bus master, fast devsel, latency 0, IRQ 18
Memory at de100000 (64-bit, non-prefetchable) [size=256K]
I/O ports at d000 [size=128]
Capabilities: <access denied>
Kernel driver in use: alx
Kernel modules: alx
ip link
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: enp3s0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc mq state UP mode DEFAULT group default qlen 1000
link/ether d8:cb:8a:9f:46:7a brd ff:ff:ff:ff:ff:ff
Offline
Try using ethtool to list the settings of enp3s0: ethtool enp3s0
Check the values of Speed and Auto-negotiation. In particular, ensure the later is on (one can use ethtool to set that)
Last edited by ewaller (2019-02-20 15:25:02)
Nothing is too wonderful to be true, if it be consistent with the laws of nature -- Michael Faraday
Sometimes it is the people no one can imagine anything of who do the things no one can imagine. -- Alan Turing
---
How to Ask Questions the Smart Way
Offline
Ohh... snap... so something is triggering the interface to 100... but checking the NetworkManager configs... 803-E settings... they are the same between the instances..
/dev/sda (issue instance)
~$ ethtool enp3s0
Settings for enp3s0:
Supported ports: [ TP ]
Supported link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Full
Supported pause frame use: Symmetric Receive-only
Supports auto-negotiation: Yes
Supported FEC modes: Not reported
Advertised link modes: Not reported
Advertised pause frame use: No
Advertised auto-negotiation: No
Advertised FEC modes: Not reported
Speed: 100Mb/s
Duplex: Half
Port: Twisted Pair
PHYAD: 0
Transceiver: internal
Auto-negotiation: off
MDI-X: Unknown
Cannot get wake-on-lan settings: Operation not permitted
Current message level: 0x000060e4 (24804)
link ifup rx_err tx_err hw wol
Link detected: yes
/dev/sdb & /dev/sdc ....
Settings for enp3s0:
Supported ports: [ TP ]
Supported link modes: 10baseT/Half 10baseT/Full
100baseT/Half 100baseT/Full
1000baseT/Full
Supported pause frame use: Symmetric Receive-only
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: 1000Mb/s
Duplex: Full
Port: Twisted Pair
PHYAD: 0
Transceiver: internal
Auto-negotiation: on
MDI-X: Unknown
Current message level: 0x000060e4 (24804)
link ifup rx_err tx_err hw wol
Link detected: yes
Offline
Look again. auto negotiation is off on sda. on for the other 2
Edit: and duplex is half.
Last edited by ewaller (2019-02-20 15:41:14)
Nothing is too wonderful to be true, if it be consistent with the laws of nature -- Michael Faraday
Sometimes it is the people no one can imagine anything of who do the things no one can imagine. -- Alan Turing
---
How to Ask Questions the Smart Way
Offline
Right... but I don't understand how.... because when I look at the GUI applet... I see this
So I need to google and learn some more details on how to figure out why it's setting to 100 even though the settings say to auto negotiate... I didn't even think to look at the link speed... it's just always been correct until now.
Offline
Well... I think something is loading on boot that is preventing the autoneg and advert link modes etc...
From NM config perspective... there is no custom or edited confs .. the /etc/NetworkManager files are identical between my other instances.. however just quickly trying stuff with ethtool on sda ... I get the following...
$ sudo ethtool --change enp3s0 autoneg on
Cannot set new settings: Invalid argument
not setting autoneg
$sudo ethtool --change enp3s0 speed 1000 duplex full
Cannot set new settings: Invalid argument
not setting speed
not setting duplex
So now I believe the question is a little more specific....
What could be interfering with the autoneg and advertised link speeds to prevent proper Auto-Negotiation ?
Edit: additional info from dmesg.... something somewhere is setting it to 100 and I currently can't find what....
fresh boot into /dev/sdb
$ dmesg | grep enp3s0
[ 2.237661] alx 0000:03:00.0 enp3s0: renamed from eth0
[ 3.015117] IPv6: ADDRCONF(NETDEV_UP): enp3s0: link is not ready
[ 3.015932] IPv6: ADDRCONF(NETDEV_UP): enp3s0: link is not ready
[ 4.690685] alx 0000:03:00.0 enp3s0: NIC Up: 1 Gbps Full
[ 4.690902] IPv6: ADDRCONF(NETDEV_CHANGE): enp3s0: link becomes ready
fresh boot into /dev/sda (issue instance)
$ dmesg | grep enp3s0
[ 2.298284] alx 0000:03:00.0 enp3s0: renamed from eth0
[ 3.321927] IPv6: ADDRCONF(NETDEV_UP): enp3s0: link is not ready
[ 3.322585] IPv6: ADDRCONF(NETDEV_UP): enp3s0: link is not ready
[ 5.455434] alx 0000:03:00.0 enp3s0: NIC Up: 1 Gbps Full
[ 5.455655] IPv6: ADDRCONF(NETDEV_CHANGE): enp3s0: link becomes ready
[ 5.455721] alx 0000:03:00.0 enp3s0: NIC Up: 100 Mbps Half
Last edited by grininmonkey (2019-02-20 16:44:24)
Offline
I wasn't getting anywhere in determining what was causing the NIC to reset to 100 Mbps after initially setting to 1 Gbps.
I had "Zero" files in /etc/NetworkManager/system-connections (no saved profiles via applet/GUI)
I had no modifications to the default /etc/NetworkManager/NetworkManager.conf file
All other folders within /etc/NetworkManager contained no files
So I tried reinstalling NetworkManager which didn't seem to work until I rebooted.... I rebooted twice and both times it seems to be correctly setting to 1 Gbps...
So I hope it sticks... and I guess this is "Solved" ... but not sure what was causing all this...
Offline