Ethernet broken after update

jb1223 · 2019-05-27 09:20:24

Hi all,

First post, so please bear with me. I recently updated my kernel to 5.0.13. Since updating, my ethernet connection has been broken on most boots. I noticed this issue 3 days ago; after an unsuccessful day of trying to fix it, I took a day off, and it seemed to fix itself on the next boot. But since restarting after that, it no longer works.

I don't know much about internet stuff in general, but from what I can tell, it looks like the issue is connected to my routing table not being set up. Going off of the Network Configuration page on the wiki, it seems like the modules are loaded correctly:

uname -a
*************************************
Linux noisy-cricket 5.0.13-arch1-1-ARCH #1 SMP PREEMPT Sun May 5 18:05:41 UTC 2019 x86_64 GNU/Linux

lspci -v
*************************************
00:19.0 Ethernet controller: Intel Corporation Ethernet Connection I217-V (rev 05)
	DeviceName:  Onboard LAN
	Subsystem: ASUSTeK Computer Inc. Ethernet Connection I217-V
	Flags: bus master, fast devsel, latency 0, IRQ 33
	Memory at ef400000 (32-bit, non-prefetchable) [size=128K]
	Memory at ef439000 (32-bit, non-prefetchable) [size=4K]
	I/O ports at f040 [disabled] [size=32]
	Capabilities: [c8] Power Management version 2
	Capabilities: [d0] MSI: Enable+ Count=1/1 Maskable- 64bit+
	Capabilities: [e0] PCI Advanced Features
	Kernel driver in use: e1000e
	Kernel modules: e1000e

dmesg | grep e1000e
*************************************
[    2.498776] e1000e: Intel(R) PRO/1000 Network Driver - 3.2.6-k
[    2.498776] e1000e: Copyright(c) 1999 - 2015 Intel Corporation.
[    2.498932] e1000e 0000:00:19.0: Interrupt Throttling Rate (ints/sec) set to dynamic conservative mode
[    2.625155] e1000e 0000:00:19.0 0000:00:19.0 (uninitialized): registered PHC clock
[    2.714944] e1000e 0000:00:19.0 eth0: (PCI Express:2.5GT/s:Width x1) ac:22:0b:dc:27:a8
[    2.714948] e1000e 0000:00:19.0 eth0: Intel(R) PRO/1000 Network Connection
[    2.714994] e1000e 0000:00:19.0 eth0: MAC: 11, PHY: 12, PBA No: FFFFFF-0FF
[    2.717356] e1000e 0000:00:19.0 eno1: renamed from eth0
[    6.174640] e1000e: eno1 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None

The interface is up:

ip a
*************************************
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default qlen 1000
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
       valid_lft forever preferred_lft forever
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: eno1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc fq_codel state UP group default qlen 1000
    link/ether ac:22:0b:dc:27:a8 brd ff:ff:ff:ff:ff:ff
    inet6 fe80::b925:4759:b65a:a83b/64 scope link 
       valid_lft forever preferred_lft forever
3: wlp3s0: <NO-CARRIER,BROADCAST,MULTICAST,UP> mtu 1500 qdisc noqueue state DOWN group default qlen 1000
    link/ether 24:0a:64:1c:90:f0 brd ff:ff:ff:ff:ff:ff

The command "ip route" produces no output.

I can successfully ping localhost, but I am unable to ping 8.8.8.8:

ping 8.8.8.8
*************************************
connect: Network is unreachable

Also, I've double checked that I'm not running two network managers. It looks like I should be fine in that regard:

systemctl status netctl
*************************************
● netctl.service - (Re)store the netctl profile state
   Loaded: loaded (/usr/lib/systemd/system/netctl.service; disabled; vendor preset: disabled)
   Active: inactive (dead)
     Docs: man:netctl.special(7)

systemctl status dhcpcd
*************************************
● dhcpcd.service - dhcpcd on all interfaces
   Loaded: loaded (/usr/lib/systemd/system/dhcpcd.service; enabled; vendor preset: disabled)
   Active: active (running) since Sun 2019-05-26 18:52:57 PDT; 19min ago
  Process: 479 ExecStart=/usr/bin/dhcpcd -q -b (code=exited, status=0/SUCCESS)
 Main PID: 493 (dhcpcd)
    Tasks: 1 (limit: 4915)
   Memory: 2.5M
   CGroup: /system.slice/dhcpcd.service
           └─493 /usr/bin/dhcpcd -q -b

May 26 18:52:57 noisy-cricket dhcpcd[493]: wlp3s0: adding address fe80::a736:9b18:64cb:f395
May 26 18:52:57 noisy-cricket dhcpcd[493]: wlp3s0: carrier lost
May 26 18:52:57 noisy-cricket dhcpcd[493]: wlp3s0: deleting address fe80::a736:9b18:64cb:f395
May 26 18:53:00 noisy-cricket dhcpcd[493]: eno1: carrier acquired
May 26 18:53:00 noisy-cricket dhcpcd[493]: eno1: IAID 0b:dc:27:a8
May 26 18:53:00 noisy-cricket dhcpcd[493]: eno1: adding address fe80::b925:4759:b65a:a83b
May 26 18:53:00 noisy-cricket dhcpcd[493]: control command: dhcpcd -4 -q -t 30 -L eno1
May 26 18:53:00 noisy-cricket dhcpcd[493]: eno1: soliciting a DHCP lease
May 26 18:53:00 noisy-cricket dhcpcd[493]: eno1: soliciting an IPv6 router
May 26 18:53:13 noisy-cricket dhcpcd[493]: eno1: no IPv6 Routers available

In case this is relevant, I have a system with a Windows 10/Ubuntu triple boot, although each OS is on a separate drive.

Thanks for your help! Please let me know if any other outputs would be useful. I'm not sure if the bug is in my system, or if it's an issue with my school's internet. I've had issues with it in the past, but this is the first time it has only affected Arch.

Last edited by jb1223 (2019-05-27 09:24:37)

Head_on_a_Stick · 2019-05-27 09:51:22

jb1223 wrote:

Please let me know if any other outputs would be useful

Use this to check for conflicting networking services:

systemctl list-unit-files --state=enabled

And try booting with linux-lts to eliminate problems with the new kernel version.

graysky · 2019-05-27 09:57:28

jb1223 wrote:

First post, so please bear with me. I recently updated my kernel to 5.0.13.
...
uname -a
*************************************
Linux noisy-cricket 5.0.13-arch1-1-ARCH #1 SMP PREEMPT Sun May 5 18:05:41 UTC 2019 x86_64 GNU/Linux

For 1, that kernel package is over 2 weeks ago. Is your mirrorlist pointing to a stale mirror?

jb1223 · 2019-05-27 10:11:28

Head_on_a_Stick wrote:

jb1223 wrote:
Please let me know if any other outputs would be useful
Use this to check for conflicting networking services:
systemctl list-unit-files --state=enabled

This is what I got:

UNIT FILE        STATE  
autovt@.service  enabled
dhcpcd.service   enabled
getty@.service   enabled
remote-fs.target enabled

4 unit files listed.

Head_on_a_Stick wrote:

And try booting with linux-lts to eliminate problems with the new kernel version.

I'm unfamiliar with this, will get back after reading the page and trying it out. Thank you!

jb1223 · 2019-05-27 10:18:29

graysky wrote:

jb1223 wrote:
First post, so please bear with me. I recently updated my kernel to 5.0.13.
...
uname -a
*************************************
Linux noisy-cricket 5.0.13-arch1-1-ARCH #1 SMP PREEMPT Sun May 5 18:05:41 UTC 2019 x86_64 GNU/Linux
For 1, that kernel package is over 2 weeks ago. Is your mirrorlist pointing to a stale mirror?

I think my mirror list is fine, I updated on the 12th. I realized that I'm a bit behind as I was writing this, I'm hoping I can fix it as is and update from there.

jb1223 · 2019-05-27 11:42:29

Head_on_a_Stick wrote:

And try booting with linux-lts to eliminate problems with the new kernel version.

So I downloaded it onto a USB, started setting it up, realized I didn't have bc, tried to chroot from Ubuntu to download it, and noticed I'm getting a similar issue on that OS as well - a missing routing table, but everything else looks fine. That one should definitely be working, so I'm starting to wonder if it's another issue on the school's end. I think I'll contact them tomorrow, until then I'll probably just hope it starts working again like last time. Thanks again for the suggestions!

seth · 2019-05-27 14:37:29

You're not getting a lease. Stop the service and run dhcpcd directly. You may also try "dhclient" as alternative (w/ slightly different default protocol handling)
Also check the router logs (if you can)

jb1223 · 2019-05-27 20:58:07

seth wrote:

You're not getting a lease. Stop the service and run dhcpcd directly. You may also try "dhclient" as alternative (w/ slightly different default protocol handling)
Also check the router logs (if you can)

I tried stopping and starting. It looks like I just end up in the same spot, but here's the output anyway:

After booting:

systemctl status dhcpcd
***************************
● dhcpcd.service - dhcpcd on all interfaces
   Loaded: loaded (/usr/lib/systemd/system/dhcpcd.service; enabled; vendor preset: disabled)
   Active: active (running) since Mon 2019-05-27 06:42:44 PDT; 3min 7s ago
  Process: 479 ExecStart=/usr/bin/dhcpcd -q -b (code=exited, status=0/SUCCESS)
 Main PID: 488 (dhcpcd)
    Tasks: 1 (limit: 4915)
   Memory: 2.5M
   CGroup: /system.slice/dhcpcd.service
           └─488 /usr/bin/dhcpcd -q -b

May 27 06:42:44 noisy-cricket dhcpcd[488]: wlp3s0: soliciting a DHCP lease
May 27 06:42:44 noisy-cricket dhcpcd[488]: wlp3s0: carrier lost
May 27 06:42:44 noisy-cricket dhcpcd[488]: wlp3s0: deleting address fe80::a736:9b18:64cb:f395
May 27 06:42:47 noisy-cricket dhcpcd[488]: eno1: carrier acquired
May 27 06:42:47 noisy-cricket dhcpcd[488]: eno1: IAID 0b:dc:27:a8
May 27 06:42:47 noisy-cricket dhcpcd[488]: eno1: adding address fe80::b925:4759:b65a:a83b
May 27 06:42:47 noisy-cricket dhcpcd[488]: control command: dhcpcd -4 -q -t 30 -L eno1
May 27 06:42:47 noisy-cricket dhcpcd[488]: eno1: soliciting a DHCP lease
May 27 06:42:47 noisy-cricket dhcpcd[488]: eno1: soliciting an IPv6 router
May 27 06:43:00 noisy-cricket dhcpcd[488]: eno1: no IPv6 Routers available

After stopping dhcpcd:

systemctl status dhcpcd
***************************
● dhcpcd.service - dhcpcd on all interfaces
   Loaded: loaded (/usr/lib/systemd/system/dhcpcd.service; enabled; vendor preset: disabled)
   Active: inactive (dead) since Mon 2019-05-27 06:46:22 PDT; 21s ago
  Process: 479 ExecStart=/usr/bin/dhcpcd -q -b (code=exited, status=0/SUCCESS)
  Process: 822 ExecStop=/usr/bin/dhcpcd -x (code=exited, status=0/SUCCESS)
 Main PID: 488 (code=exited, status=0/SUCCESS)

May 27 06:46:22 noisy-cricket systemd[1]: Stopping dhcpcd on all interfaces...
May 27 06:46:22 noisy-cricket dhcpcd[822]: sending signal TERM to pid 488
May 27 06:46:22 noisy-cricket dhcpcd[822]: waiting for pid 488 to exit
May 27 06:46:22 noisy-cricket dhcpcd[822]: sending signal TERM to pid 488
May 27 06:46:22 noisy-cricket dhcpcd[822]: waiting for pid 488 to exit
May 27 06:46:22 noisy-cricket dhcpcd[488]: received SIGTERM, stopping
May 27 06:46:22 noisy-cricket dhcpcd[488]: wlp3s0: removing interface
May 27 06:46:22 noisy-cricket dhcpcd[488]: eno1: removing interface
May 27 06:46:22 noisy-cricket systemd[1]: dhcpcd.service: Succeeded.
May 27 06:46:22 noisy-cricket systemd[1]: Stopped dhcpcd on all interfaces.

After starting dhcpcd again:

systemctl status dhcpcd
***************************
● dhcpcd.service - dhcpcd on all interfaces
   Loaded: loaded (/usr/lib/systemd/system/dhcpcd.service; enabled; vendor preset: disabled)
   Active: active (running) since Mon 2019-05-27 06:46:53 PDT; 5min ago
  Process: 842 ExecStart=/usr/bin/dhcpcd -q -b (code=exited, status=0/SUCCESS)
 Main PID: 843 (dhcpcd)
    Tasks: 1 (limit: 4915)
   Memory: 1.4M
   CGroup: /system.slice/dhcpcd.service
           └─843 /usr/bin/dhcpcd -q -b

May 27 06:46:53 noisy-cricket systemd[1]: Starting dhcpcd on all interfaces...
May 27 06:46:53 noisy-cricket dhcpcd[842]: dev: loaded udev
May 27 06:46:53 noisy-cricket dhcpcd[842]: forked to background, child pid 843
May 27 06:46:53 noisy-cricket systemd[1]: Started dhcpcd on all interfaces.
May 27 06:46:53 noisy-cricket dhcpcd[843]: DUID 00:04:2e:33:f5:80:d7:da:11:dd:93:af:ac:22:0b:dc:27:a8
May 27 06:46:53 noisy-cricket dhcpcd[843]: eno1: IAID 0b:dc:27:a8
May 27 06:46:53 noisy-cricket dhcpcd[843]: wlp3s0: waiting for carrier
May 27 06:46:53 noisy-cricket dhcpcd[843]: eno1: soliciting a DHCP lease
May 27 06:46:54 noisy-cricket dhcpcd[843]: eno1: soliciting an IPv6 router
May 27 06:47:06 noisy-cricket dhcpcd[843]: eno1: no IPv6 Routers available

Unfortunately, I don't think I have access to the router logs. That would be something for the network administrators, correct?

Thanks for your suggestions!

seth · 2019-05-27 21:03:21

That would be something for the network administrators, correct?

Yes.

The idea however was to skip the service, ie. stop it to run dhcpcd *directly*, not by restarting a service.

dhcpcd -w 4

(the "-w" will make it wait as foreground process until it gets an IPv4, also we're omitting the quit parameter to have it talk to us)

jb1223 · 2019-05-27 22:22:43

seth wrote:

The idea however was to skip the service, ie. stop it to run dhcpcd *directly*, not by restarting a service.
dhcpcd -w 4

Sorry for the mixup! After stopping dhcpcd and starting it with that, this is my output:

dhcpcd --waitip=4
******************
dev: loaded udev
DUID: 00:04:2e:33:f5:80:d7:da:11:dd:93:af:ac:22:0b:dc:27:a8
eno1: IAID 0b:dc:27:a8
wlp3s0: waiting for carrier
eno1: soliciting a DHCP lease
eno1: soliciting an IPv6 router
eno1: no IPv6 Routers available
timed out
forked to background, child pid 895

seth · 2019-05-27 22:29:47

Kill the process, edit /etc/dhcpcd.conf, commend "duid" and uncomment "clientid". Try again.

jb1223 · 2019-05-27 22:42:16

seth wrote:

Kill the process, edit /etc/dhcpcd.conf, commend "duid" and uncomment "clientid". Try again.

$ kill -- 895
$ vi /etc/dhcpcd.conf
$ dhcpcd --waitip=4

This produces the same output as earlier (with a different child pid, of course).

Last edited by jb1223 (2019-05-27 22:49:50)

seth · 2019-05-28 06:01:31

You can poke around more by spoofing the MAC or lie about the vendorclass ID ("-v", see the dhcpcd manpage) or set the timeout to "-t 0" (if your dhcp server is really slow)
Also try dhclient and run "sudo nmap --script broadcast-dhcp-discover" to look for rogue dhcp servers (or actually any available) - but contacting the router admin to ask for logs on their side is a more direct approach to the problem.

Head_on_a_Stick · 2019-05-28 16:39:34

@OP: can you assign static IP addresses?

https://wiki.archlinux.org/index.php/Ne … IP_address

jb1223 · 2019-05-31 11:13:19

Sorry for dropping out, I've been super busy with classes ending. But now I'm free to go back to the land of debugging!

seth wrote:

...
... but contacting the router admin to ask for logs on their side is a more direct approach to the problem.

I've contacted them, still waiting on a reply. I've noticed that when it does work, it's when I boot it up during the workday (reasonable people hours). I'm now really starting to get suspicious it's on their end, because my system is working and then not working without me changing anything.

Head_on_a_Stick wrote:

@OP: can you assign static IP addresses?

I really don't know where to begin with that, as I don't know what IP address I should be assigning. I think I'm going to wait for a response from the network admins before I start delving into that. Also, I feel like I should be able to get dynamic addresses working, given that they have worked in the past. Thanks for the suggestion!

Head_on_a_Stick · 2019-06-01 15:16:39

jb1223 wrote:

I don't know what IP address I should be assigning

Wait until you manage to connect successfully then use

ip r

to find the correct addresses.

seth · 2019-06-01 15:22:13

when it does work, it's when I boot it up during the workday (reasonable people hours)

Run the nmap script I posted in #13, but it should be possible to figure whether anybody can get any lease on any OS at night or on weekends.

jb1223 · 2019-06-20 13:40:11

Hi all,

Sorry for dropping out again, I've been pretty busy and the weeks kinda just flew by.

In the meantime, I got a response from the network admin - he says he doesn't really know much about linux, but he did mention that the university network doesn't support IPv6. I've tried running dhcpcd with the -4 option, and it still gets stuck at 'Soliciting a DHCP lease'.

Additionally, just tonight, I downloaded the current arch image, and I did a live boot from a USB. I was still unable to obtain a connection, and it looks like the issue is the same (empty routing table).

Since the clean install isn't working, I'm starting to wonder if this has something to do with booting alongside Windows. I moved rooms on campus, and again, the ethernet works fine for Windows, but not for Arch or Ubuntu.

Should I start another thread, maybe with a different description in the title? I also don't know if this is old enough to be considered necromancy.

seth wrote:

Run the nmap script I posted in #13, but it should be possible to figure whether anybody can get any lease on any OS at night or on weekends.

Starting Nmap 7.70 ( https://nmap.org ) at 2019-06-20 05:17 PDT
Nmap done: 0 IP addresses (0 hosts up) scanned in 0.15 seconds

Head_on_a_Stick wrote:

Wait until you manage to connect successfully and then use
ip r
to find the correct addresses.

Unfortunately, I don't think that will work, for two reasons. First, I haven't had a successful connection since then. Second, after doing some digging through my university's networking FAQs, it looks like I'm not supposed to be using static IP addresses.

UPDATE: tried a few more things just before posting this, ended up bricking my desktop after a failed boot into ubuntu. Welp. Guess I'll be back after fixing that :') guess I was due for a mobo replacement anyway, I had to do some ungodly things just to boot arch from my desired drive.

ewaller · 2019-06-20 14:25:50

Hello, neighbor (I am a couple miles north and a little east of where I think you are). If you are going to that school on California Blvd, I'm pretty sure that someone there is using Linux. In fact, I know so.
When you get your system back together, check what happens if you boot from the arch install media. See if you get an address with it. If so, check the modules that are pulled into the kernel.

jb1223 · 2019-06-20 14:50:24

ewaller wrote:

If you are going to that school on California Blvd, I'm pretty sure that someone there is using Linux. In fact, I know so.

Hey neighbor! Yup, you got me

Definitely going to be trying again once I get the mobo replaced. I'm hoping that it was just a hardware issue, and by replacing it, things should work again. People here are definitely using Linux, so I guess that does narrow down my possible issues.

seth · 2019-06-20 14:54:31

As for the dual boot condition please check https://wiki.archlinux.org/index.php/Du … t_Start-Up *especially* if you're saying that this is a random works/works not situation AND at the same time of day when it failed on linux, it works on windows.

jb1223 · 2019-06-20 15:04:22

seth wrote:

As for the dual boot condition please check https://wiki.archlinux.org/index.php/Du … t_Start-Up *especially* if you're saying that this is a random works/works not situation AND at the same time of day when it failed on linux, it works on windows.

I've definitely had fast (and secure) boot off since the beginning, and I double checked the system times earlier today (prior to the mobo fire). I'll be sure to double check that once I get the system up again though, thanks! I guess my boot setup was a bit strange, I had to manually mod the BIOS to get it to boot the drive containing Arch. I'm planning on getting a mobo that already has support for that drive, so hopefully with a cleaner setup I'll have fewer issues.

Thanks for all your help! I'm pretty sure I ran through one of your old threads while digging, didn't notice the name until now.

Last edited by jb1223 (2019-06-20 15:05:10)

jb1223 · 2019-07-05 20:58:41

I finally replaced all the parts and reinstalled arch, and it looks like it's working without issue! Thanks again for all the help, I'm still not sure what the issue was but at least it's fixed now. Time to get back to reinstalling everything else...

Arch Linux

#1 2019-05-27 09:20:24

Ethernet broken after update

#2 2019-05-27 09:51:22

Re: Ethernet broken after update

#3 2019-05-27 09:57:28

Re: Ethernet broken after update

#4 2019-05-27 10:11:28

Re: Ethernet broken after update

#5 2019-05-27 10:18:29

Re: Ethernet broken after update

#6 2019-05-27 11:42:29

Re: Ethernet broken after update

#7 2019-05-27 14:37:29

Re: Ethernet broken after update

#8 2019-05-27 20:58:07

Re: Ethernet broken after update

#9 2019-05-27 21:03:21

Re: Ethernet broken after update

#10 2019-05-27 22:22:43

Re: Ethernet broken after update

#11 2019-05-27 22:29:47

Re: Ethernet broken after update

#12 2019-05-27 22:42:16

Re: Ethernet broken after update

#13 2019-05-28 06:01:31

Re: Ethernet broken after update

#14 2019-05-28 16:39:34

Re: Ethernet broken after update

#15 2019-05-31 11:13:19

Re: Ethernet broken after update

#16 2019-06-01 15:16:39

Re: Ethernet broken after update

#17 2019-06-01 15:22:13

Re: Ethernet broken after update

#18 2019-06-20 13:40:11

Re: Ethernet broken after update

#19 2019-06-20 14:25:50

Re: Ethernet broken after update

#20 2019-06-20 14:50:24

Re: Ethernet broken after update

#21 2019-06-20 14:54:31

Re: Ethernet broken after update

#22 2019-06-20 15:04:22

Re: Ethernet broken after update

#23 2019-07-05 20:58:41

Re: Ethernet broken after update

Board footer