* Weird issue with r8169: shutdown on carrier comeback
@ 2026-02-02 17:00 Tanguy Ortolo
2026-02-02 17:25 ` Heiner Kallweit
0 siblings, 1 reply; 3+ messages in thread
From: Tanguy Ortolo @ 2026-02-02 17:00 UTC (permalink / raw)
To: Realtek linux nic maintainers, Francois Romieu; +Cc: netdev
Hello,
I am experiencing a weird computer issue that is at least partly linked
to the r8169 driver: the computer shuts down when the Ethernet cable is
pulled off and plugged again. Or when the switch goes down and up again.
To be more specific, the computer is a [Trigkey Key-N100][1] running
Debian Trixie. That computer has two Gigabit Ethernet controllers:
$ lspci | grep Ethernet
01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168/8211/8411 PCI Express Gigabit Ethernet Controller (rev 15)
02:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168/8211/8411 PCI Express Gigabit Ethernet Controller (rev 15)
[1]: https://devhome.corp.adobe.com/docs/default/component/ethos-flex/docs/devhome/prerequisites/
One of them is configured, and when I pull off its Ethernet cable and
plug it again after a few seconds (or if I reboot the switch it is
connected to), something simulates a power key press (log from
systemd-logind: “Power key pressed short”). Even though I configured
logind to ignore such events, one second after this, the computer shuts
itself down brutally, as if experiencing a power loss.
After a few experiments, I determined that:
* it happens on both Ethernet interfaces of that computer;
* it happens only on _configured_ interfaces: in particular, it does not
happen if I unplug the Ethernet cable, run `ip link set enp1s0 down`
and plug the cable again;
* it does not happen with the proprietary driver r8168.
I am not really sure the r8169 driver is at fault here; it sounds more
like a very weird hardware issue in specific conditions, which are met
when using it. Anyway, I though you deserved to know, in case it could
be useful. Please tell me if there is anything else I could do to
further diagnose that issue.
Regards,
--
Tanguy Ortolo
^ permalink raw reply [flat|nested] 3+ messages in thread* Re: Weird issue with r8169: shutdown on carrier comeback
2026-02-02 17:00 Weird issue with r8169: shutdown on carrier comeback Tanguy Ortolo
@ 2026-02-02 17:25 ` Heiner Kallweit
2026-02-03 10:01 ` Tanguy Ortolo
0 siblings, 1 reply; 3+ messages in thread
From: Heiner Kallweit @ 2026-02-02 17:25 UTC (permalink / raw)
To: Tanguy Ortolo, Realtek linux nic maintainers, Francois Romieu; +Cc: netdev
On 2/2/2026 6:00 PM, Tanguy Ortolo wrote:
> Hello,
>
> I am experiencing a weird computer issue that is at least partly linked to the r8169 driver: the computer shuts down when the Ethernet cable is pulled off and plugged again. Or when the switch goes down and up again.
>
> To be more specific, the computer is a [Trigkey Key-N100][1] running Debian Trixie. That computer has two Gigabit Ethernet controllers:
> $ lspci | grep Ethernet
> 01:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168/8211/8411 PCI Express Gigabit Ethernet Controller (rev 15)
> 02:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8111/8168/8211/8411 PCI Express Gigabit Ethernet Controller (rev 15)
>
> [1]: https://devhome.corp.adobe.com/docs/default/component/ethos-flex/docs/devhome/prerequisites/
>
> One of them is configured, and when I pull off its Ethernet cable and plug it again after a few seconds (or if I reboot the switch it is connected to), something simulates a power key press (log from systemd-logind: “Power key pressed short”). Even though I configured logind to ignore such events, one second after this, the computer shuts itself down brutally, as if experiencing a power loss.
>
> After a few experiments, I determined that:
> * it happens on both Ethernet interfaces of that computer;
> * it happens only on _configured_ interfaces: in particular, it does not
> happen if I unplug the Ethernet cable, run `ip link set enp1s0 down`
> and plug the cable again;
> * it does not happen with the proprietary driver r8168.
>
> I am not really sure the r8169 driver is at fault here; it sounds more like a very weird hardware issue in specific conditions, which are met when using it. Anyway, I though you deserved to know, in case it could be useful. Please tell me if there is anything else I could do to further diagnose that issue.
>
Thanks for the report. Interesting issue .. The network driver (r8169) has no link to the input system.
Therefore it's likely your network manager which is causing this behavior. Maybe r8168 is different in what
network notifications it sends. But hard to say w/o knowing what the network manager looks at to detect
link-down/-up events.
> Regards,
>
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: Weird issue with r8169: shutdown on carrier comeback
2026-02-02 17:25 ` Heiner Kallweit
@ 2026-02-03 10:01 ` Tanguy Ortolo
0 siblings, 0 replies; 3+ messages in thread
From: Tanguy Ortolo @ 2026-02-03 10:01 UTC (permalink / raw)
To: Heiner Kallweit; +Cc: Realtek linux nic maintainers, Francois Romieu, netdev
Hello,
Thank you for the quick reply.
Heiner Kallweit, 2026-02-02 18:25+0100:
>Thanks for the report. Interesting issue .. The network driver (r8169) has no link to the input system.
>Therefore it's likely your network manager which is causing this behavior.
I doubt it, the network is simply managed by Debian's ifupdown. Nothing
fancy like NetworkManager for instance. By the way, I tried after
switching to systemd-networkd, and the issue remains.
(I also tried after enabling ifplugd, which deconfigures the interface
when a carrier loss is detected, and that does avoid the problem, just
like when I do the same manually: shutdown only happens when I re-plug
the cable into an interface that is still configured.)
I would more likely suspect something weird in the motherboard itself.
But this is indeed very hard to tell. I reported the issue to the
manufacturer as well (Trigkey), but I am not sure they will reply. I do
not see how the r8169 driver could be responsible for that issue, I
think it merely creates conditions in which some hardware bug appears.
But since r8168 does not trigger that issue, I felt I should report it
just in case it could be useful. Do not waste too much time on this. :-)
Regards,
--
Tanguy Ortolo
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2026-02-03 10:01 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-02-02 17:00 Weird issue with r8169: shutdown on carrier comeback Tanguy Ortolo
2026-02-02 17:25 ` Heiner Kallweit
2026-02-03 10:01 ` Tanguy Ortolo
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox