System crash when chaging network configuration

Hi,

I have a Thinkpad T470s, and I recently upgraded to FC40 and since I have a pretty annoying problem with my wired network adapter.

If I change the configuration of my NIC or even unplug/plug the wire, the system freezes and I have to reset it.

I did several updates and until now I am having the same problem.

[root@machine ~] # uname -a
Linux 6.8.7-300.fc40.x86_64 #1 SMP PREEMPT_DYNAMIC Wed Apr 17 19:21:08 UTC 2024 x86_64 GNU/Linux

[root@machine ~] # lspci -s 00:1f.6 -vvv
00:1f.6 Ethernet controller: Intel Corporation Ethernet Connection I219-LM (rev 21)
Subsystem: Lenovo Device 224b
Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx+
Status: Cap+ 66MHz- UDF- FastB2B- ParErr- DEVSEL=fast >TAbort- <TAbort- SERR- <PERR- INTx-
Latency: 0
Interrupt: pin A routed to IRQ 130
Region 0: Memory at dc100000 (32-bit, non-prefetchable) [size=128K]
Capabilities: [c8] Power Management version 3
Flags: PMEClk- DSI+ D1- D2- AuxCurrent=0mA PME(D0+,D1-,D2-,D3hot+,D3cold+)
Status: D0 NoSoftRst+ PME-Enable- DSel=0 DScale=1 PME-
Capabilities: [d0] MSI: Enable+ Count=1/1 Maskable- 64bit+
Address: 00000000fee00298 Data: 0000
Capabilities: [e0] PCI Advanced Features
AFCap: TP+ FLR+
AFCtrl: FLR-
AFStatus: TP-
Kernel driver in use: e1000e
Kernel modules: e1000e

This is the last entries on my journald system before the last crash

May 05 09:40:01 machine.domain.tld kernel: e1000e 0000:00:1f.6 enp0s31f6: NIC Link is Up 1000 Mbps Half Duplex, Flow Control: Rx/Tx
May 05 09:40:01 machine.domain.tld kernel: BUG: scheduling while atomic: kworker/0:3/214/0x00000002
May 05 09:40:01 machine.domain.tld kernel: Modules linked in: uinput snd_seq_dummy snd_hrtimer bridge stp llc nf_nat_tftp nf_conntrack_tftp nf_conn>
May 05 09:40:01 machine.domain.tld kernel: snd_pcm_dmaengine kvm_intel intel_pmc_bxt snd_hda_codec_realtek iTCO_vendor_support mei_wdt snd_hda_cod>
May 05 09:40:01 machine.domain.tld kernel: CPU: 0 PID: 214 Comm: kworker/0:3 Not tainted 6.8.7-300.fc40.x86_64 #1
May 05 09:40:01 machine.domain.tld kernel: Hardware name: LENOVO 20JTS0NJ00/20JTS0NJ00, BIOS N1WET74W (1.53 ) 02/26/2024
May 05 09:40:01 machine.domain.tld kernel: Workqueue: events e1000_watchdog_task [e1000e]
May 05 09:40:01 machine.domain.tld kernel: Call Trace:
May 05 09:40:01 machine.domain.tld kernel:
May 05 09:40:01 machine.domain.tld kernel: dump_stack_lvl+0x6a/0x90
May 05 09:40:01 machine.domain.tld kernel: __schedule_bug+0x56/0x70
May 05 09:40:01 machine.domain.tld kernel: __schedule+0x10a7/0x1530
May 05 09:40:01 machine.domain.tld kernel: ? ktime_get+0x3c/0xa0
May 05 09:40:01 machine.domain.tld kernel: ? lapic_next_deadline+0x2c/0x40
May 05 09:40:01 machine.domain.tld kernel: ? hrtimer_start_range_ns+0x267/0x350
May 05 09:40:01 machine.domain.tld kernel: schedule+0x32/0xd0
May 05 09:40:01 machine.domain.tld kernel: schedule_hrtimeout_range_clock+0xe7/0x1e0
May 05 09:40:01 machine.domain.tld kernel: ? __pfx_hrtimer_wakeup+0x10/0x10
May 05 09:40:01 machine.domain.tld kernel: usleep_range_state+0x67/0x90
May 05 09:40:01 machine.domain.tld kernel: e1000e_read_phy_reg_mdic+0x87/0x280 [e1000e]
May 05 09:40:01 machine.domain.tld kernel: e1000e_update_stats+0x51a/0x720 [e1000e]
May 05 09:40:01 machine.domain.tld kernel: e1000_watchdog_task+0xe5/0xaa0 [e1000e]
May 05 09:40:01 machine.domain.tld kernel: process_one_work+0x16d/0x330
May 05 09:40:01 machine.domain.tld kernel: worker_thread+0x273/0x3c0
May 05 09:40:01 machine.domain.tld kernel: ? __pfx_worker_thread+0x10/0x10
May 05 09:40:01 machine.domain.tld kernel: kthread+0xe5/0x120
May 05 09:40:01 machine.domain.tld kernel: ? __pfx_kthread+0x10/0x10
May 05 09:40:01 machine.domain.tld kernel: ret_from_fork+0x31/0x50
May 05 09:40:01 machine.domain.tld kernel: ? __pfx_kthread+0x10/0x10

Regards.

Hi everyone,

I’m encountering an issue on my Fedora system, and I’m looking for some help to resolve it. Below is the detailed error message from the logs:

mai 14 16:50:39 fedora kernel: watchdog: BUG: soft lockup - CPU#7 stuck for 26s! [kworker/7:0:8963]
mai 14 16:50:39 fedora kernel: Modules linked in: ...
mai 14 16:50:39 fedora kernel: CPU: 7 PID: 8963 Comm: kworker/7:0 Tainted: G        W          6.8.9-300.fc40.x86_64 #1
mai 14 16:50:39 fedora kernel: Hardware name: LENOVO 20S1SH7100/20S1SH7100, BIOS N2XET40W (1.30 ) 10/31/2023
mai 14 16:50:39 fedora kernel: Workqueue: events linkwatch_event
mai 14 16:50:39 fedora kernel: RIP: 0010:native_queued_spin_lock_slowpath+0x72/0x2d0
mai 14 16:50:39 fedora kernel: Code: ...
mai 14 16:50:39 fedora kernel: RSP: 0018:ffffaab09118bba0 EFLAGS: 00000202
mai 14 16:50:39 fedora kernel: RAX: 0000000000000001 RBX: ffff93464198b428 RCX: 0000000000000000
mai 14 16:50:39 fedora kernel: RDX: 0000000000000000 RSI: 0000000000000001 RDI: ffff93464198b428
mai 14 16:50:39 fedora kernel: RBP: ffff93464b093138 R08: 0000000000000000 R09: 0000000000000000
mai 14 16:50:39 fedora kernel: R10: ffff93464b093200 R11: 0000000000000010 R12: ffff93464198b428
mai 14 16:50:39 fedora kernel: R13: ffff93464b093000 R14: ffffaab09118bc97 R15: ffffaab09118bcb8
mai 14 16:50:39 fedora kernel: FS:  0000000000000000(0000) GS:ffff934785780000(0000) knlGS:0000000000000000
mai 14 16:50:39 fedora kernel: CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
mai 14 16:50:39 fedora kernel: CR2: 00007f8453643000 CR3: 0000000239428004 CR4: 00000000003706f0
mai 14 16:50:39 fedora kernel: DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
mai 14 16:50:39 fedora kernel: DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
mai 14 16:50:39 fedora kernel: Call Trace:
mai 14 16:50:39 fedora kernel:  <IRQ>
mai 14 16:50:39 fedora kernel:  ? watchdog_timer_fn+0x1ea/0x270
mai 14 16:50:39 fedora kernel:  ? __pfx_watchdog_timer_fn+0x10/0x10
mai 14 16:50:39 fedora kernel:  ? __hrtimer_run_queues+0x113/0x280
mai 14 16:50:39 fedora kernel:  ? ktime_get_update_offsets_now+0x49/0x110
mai 14 16:50:39 fedora kernel:  ? hrtimer_interrupt+0xf8/0x230
mai 14 16:50:39 fedora kernel:  ? __sysvec_apic_timer_interrupt+0x4a/0x140
mai 14 16:50:39 fedora kernel:  ? sysvec_apic_timer_interrupt+0x6d/0x90
mai 14 16:50:39 fedora kernel:  </IRQ>
mai 14 16:50:39 fedora kernel:  <TASK>
mai 14 16:50:39 fedora kernel:  ? asm_sysvec_apic_timer_interrupt+0x1a/0x20
mai 14 16:50:39 fedora kernel:  ? native_queued_spin_lock_slowpath+0x72/0x2d0
mai 14 16:50:39 fedora kernel:  _raw_spin_lock+0x29/0x30
mai 14 16:50:39 fedora kernel:  e1000e_get_stats64+0x22/0x120 [e1000e]
mai 14 16:50:39 fedora kernel:  dev_get_stats+0x65/0x150
mai 14 16:50:39 fedora kernel:  ? __nla_reserve+0x3c/0x50
mai 14 16:50:39 fedora kernel:  rtnl_fill_stats+0x3b/0x130
mai 14 16:50:39 fedora kernel:  rtnl_fill_ifinfo+0x83d/0x1550
mai 14 16:50:39 fedora kernel:  ? __kmalloc_node_track_caller+0x33a/0x4d0
mai 14 16:50:39 fedora kernel:  ? __alloc_skb+0x8a/0x1a0
mai 14 16:50:39 fedora kernel:  rtmsg_ifinfo_build_skb+0xae/0x110
mai 14 16:50:39 fedora kernel:  rtmsg_ifinfo+0x3c/0x90
mai 14 16:50:39 fedora kernel:  netdev_state_change+0x89/0x90
mai 14 16:50:39 fedora kernel:  linkwatch_do_dev+0x4f/0x60
mai 14 16:50:39 fedora kernel:  __linkwatch_run_queue+0xdf/0x220
mai 14 16:50:39 fedora kernel:  linkwatch_event+0x31/0x40
mai 14 16:50:39 fedora kernel:  process_one_work+0x16f/0x330
mai 14 16:50:39 fedora kernel:  worker_thread+0x273/0x3c0
mai 14 16:50:39 fedora kernel:  ? __pfx_worker_thread+0x10/0x10
mai 14 16:50:39 fedora kernel:  kthread+0xe5/0x120
mai 14 16:50:39 fedora kernel:  ? __pfx_kthread+0x10/0x10
mai 14 16:50:39 fedora kernel:  ret_from_fork+0x31/0x50
mai 14 16:50:39 fedora kernel:  ? __pfx_kthread+0x10/0x10
mai 14 16:50:39 fedora kernel:  ret_from_fork_asm+0x1b/0x30
mai 14 16:50:39 fedora audit[1000]: AVC avc:  denied  { read } for  pid=1000 comm=72733A6D61696E20513A526567 name="sessions" dev="tmpfs" ino=90 scontext=system_u:system_r:syslogd_t:s0 tcontext=system_u:object_r:systemd_logind_sessions_t:s0 tclass

Every time I switch from Ethernet to Wi-Fi, the system completely freezes. This issue started occurring after updating to kernel version 6.8.xxx.

I’d check under BIOS to see if there’s a setting to handle wifi and Ethernet activations. On my Dell 5591, there’s a setting to automatically disable wifi if Ethernet is plugged in, and I’m just assuming that wouldn’t work anywhere near ideally from within operating systems and never tried it :stuck_out_tongue:

I’d also make sure your BIOS is up-to-date. I’ve also heard some Intel Ethernet adapters around the I219 model (maybe I211?) had firmware updates available in other communities that fixed issues.


Looking at that at little closer, it sounds like native_queued_spin_lock_slowpath means that communications with the CPU wasn’t fast enough (deadline scheduling?) One idea could be to make sure the drive is using another scheduler or none (not sure how to do it offhand but NVMe defaults to none)


I don’t know what my revision is, but I have that exact Ethernet (I219-LM) on my laptop. I can try messing with it later to see if I experience anything.

I don’t imagine it’s something specific to Fedora, but on openSUSE TW with kernel 6.8.9-1, hotplugging my Ethernet works fine without any errors in dmesg. I have Intel Corporation Ethernet Connection (7) I219-LM (rev 10)


I see on a Windows driver thread that there’s apparently different drivers for I219 based on the revision. I have rev 10 and apparently use a shared driver between I217/I218/I219 models. Interestingly I see a lot of rev 10 stuff in my lspci and wonder if it’s a chipset thing (I have Coffee Lake).

rev 18+ has a different driver, and yours being 21 might imply some notable difference.

Laptop vendors often try to be “helpful” with switching between wired and WiFi network, particularly with the current emphasis on minimizing power consumption and prolonging runtime on battery. It is not unusual to have issues with some use cases – maybe they only test disabling wired connections when undocking. Sometimes there are workarounds (e.g., manually switch to wifi before disconnecting ethernet cable).

Do you boot more than one OS? Many systems enable some basic networking at boot to support PXE (Preboot eXecution Environment). There can be conflicts due to differences in firmware support between PXE, Wiindows, and Linux. Intel usually fixes those with firmware updates (and corresponding linux module changes), but vendors often add their own tweaks to firmware so you need to get updates from Lenovo as well as Fedora. When reporting issues it is best to fully update Fedora packages and also vendor firmware (e.g., with fwupdtool) so you aren’t wasting time on solved issues.

1 Like

The problem is indeed related to the Ethernet network card e1000 (00:1f.6 Ethernet controller: Intel Corporation Ethernet Connection (10) I219-LM). The system freezes only when I disconnect and reconnect the UTP cable.

I disabled Wi-Fi through the graphical interface and left only Ethernet enabled.

I also disabled the simultaneous operation of Wi-Fi and Ethernet in the BIOS. Now, when I connect to Wi-Fi, it is automatically turned off.

I hope that a new kernel or firmware update will include a fix for this issue.

According to J Forbes, it should be addressed in 6.8.10
https://bugzilla.redhat.com/show_bug.cgi?id=2276325#c6

1 Like

Thank you, @filippo2 Waiting eagerly. To avoid worsening or creating a problem, I will only use Wi-Fi and not install or change anything until the new kernel is available.

1 Like

Once again, “Solved” isn’t appropriate until the fix arrives, but it could help others with the same issue to have a “waiting for upstream patch” status.

1 Like

Hello everyone,

The latest update I received fixed the problem. In this update, the Linux kernel 6.8.10-300.fc40.x86_64 #1 SMP PREEMPT_DYNAMIC Fri May 17 21:20:54 UTC 2024 was installed.

Now everything is working fine, even with both interfaces connected at the same time. I have disconnected and reconnected the UTP cable without any freezes so far.

Thank you to everyone who helped me.

1 Like