Computer Auto Reboots After Changing Monitor

Hello. Sorry if this happens to be the wrong venue for this, but I’m not super savvy with this stuff and I’m at my wits end with it regardless. Haha. So. Thank you and sorry in advance.

I’m running Fedora 43 on an older desktop with an nvidia 1080ti, old intel i7, and an ASROCK Extreme motherboard. I’ve tried to ensure all my “wake on” settings were disabled on the mother board, as well.

I just changed my monitor out. My old one had a DVI cable. The new one has a cord that looks kind of like an HDMI cable but has one corner cut out of it.

Upon swapping monitors, every time I hit the “power off”, it immediately reboots.

The monitor is made by AOC. I have no devices connected to it (just power cable and the output to the desktop). It is connected directly into the nvidia 1080ti.

As a workaround, you can, from a terminal, type

shutdown now

The issue you have sounds like it has nothing to do with the monitor, but rather for some reason the function of your power button has changed from shutdown to reboot.

You can look up how to change the function of your power button.

Hello and thanks for chiming in.

Upon typing the command, the computer powers down and still immediately starts itself back up.

I’m not pressing any buttons, etc. I’m open to consider it’s “something else,” but I’m not sure what that something else is. The monitor was just the only thing I’ve changed.

I’m not sure if nvidia needs something adjusted since changing monitors (and cable), etc.

But yes. No change.

In that case, it could be some peripheral.

I have had an AOC monitor and it all worked fine.

Can you try a HDMI cable?

cat /proc/acpi/wakeup

I ran that command earlier and I disabled everything that was enabled. There was one that when I disabled it it turned off my keyboard and mouse. That one I left enabled. (I couldn’t use my keyboard or mouse when I turned it off. I had to force power down on that one).

Regarding the HDMI cable, believe it or not, I’m not sure I even have one at the moment. I moved somewhat recently and I don’t even own a TV.

This sounds like it may be a DP (display port) to HDMI cable – especially if it has HDMI on one end and the other connector (DP ?) on the other end. If it has the same connector on both ends then it may be a DP cable.

HDMI cables are inexpensive so the suggestion to try a direct cable with HDMI on both ends sounds like a good idea.

I’m def considering this if it ends up being the last resort. Where I’m at, I need to rely on public transit and I live in a quite inconvenient place, so I’ll bite that bullet if I need to ultimately.

I’m trying to troubleshoot it with ChatGPT (I don’t really expect it to “solve” it but I’m hoping it can at least guide me. I’m not very strong with diagnosing/troubleshooting).

mrdaisybates@hattivatti-tietokone:~$ echo $XDG_SESSION_TYPE
last -x | head -20
journalctl -b -1 -p 3
journalctl -b -1 | grep -Ei 'nvidia|nvrm|drm|edid|flip|timeout|xid|suspend|resume'
rpm -qa | grep -E 'akmod-nvidia|kmod-nvidia|xorg-x11-drv-nvidia-power'
systemctl status nvidia-suspend.service nvidia-resume.service nvidia-hibernate.service
wayland
mrdaisy* tty2         tty2             Wed Mar 18 16:55   still logged in
mrdaisy* seat0        login screen     Wed Mar 18 16:55   still logged in
reboot   system boot  6.19.8-200.fc43* Wed Mar 18 16:55   still running
shutdown system down  6.19.8-200.fc43* Wed Mar 18 16:55 - 16:55  (00:00)
mrdaisy* tty2         tty2             Wed Mar 18 16:49 - 16:55  (00:05)
mrdaisy* seat0        login screen     Wed Mar 18 16:49 - down   (00:05)
reboot   system boot  6.19.8-200.fc43* Wed Mar 18 16:49 - 16:55  (00:06)
shutdown system down  6.19.8-200.fc43* Wed Mar 18 16:48 - 16:49  (00:00)
reboot   system boot  6.19.8-200.fc43* Wed Mar 18 16:47 - 16:48  (00:00)
shutdown system down  6.19.8-200.fc43* Wed Mar 18 16:47 - 16:47  (00:00)
reboot   system boot  6.19.8-200.fc43* Wed Mar 18 16:47 - 16:47  (00:00)
shutdown system down  6.19.8-200.fc43* Wed Mar 18 15:44 - 16:47  (01:02)
mrdaisy* tty2         tty2             Wed Mar 18 15:40 - down   (00:03)
mrdaisy* seat0        login screen     Wed Mar 18 15:40 - down   (00:03)
reboot   system boot  6.19.8-200.fc43* Wed Mar 18 15:40 - 15:44  (00:04)
shutdown system down  6.19.8-200.fc43* Wed Mar 18 15:39 - 15:40  (00:00)
mrdaisy* tty2         tty2             Wed Mar 18 15:33 - 15:39  (00:06)
mrdaisy* seat0        login screen     Wed Mar 18 15:33 - down   (00:06)
reboot   system boot  6.19.8-200.fc43* Wed Mar 18 15:33 - 15:39  (00:06)
shutdown system down  6.19.8-200.fc43* Wed Mar 18 15:32 - 15:33  (00:00)
Mar 18 16:49:05 hattivatti-tietokone kernel: scsi 10:0:0:1: Wrong diagnostic page; asked for 1 got 8
Mar 18 16:49:05 hattivatti-tietokone kernel: scsi 10:0:0:1: Failed to get diagnostic page 0x1
Mar 18 16:49:05 hattivatti-tietokone kernel: scsi 10:0:0:1: Failed to bind enclosure -19
Mar 18 16:49:07 hattivatti-tietokone kernel: 
Mar 18 16:49:16 hattivatti-tietokone firewalld[1002]: ERROR: NAME_CONFLICT: new_policy_object(): 'docker>
Mar 18 16:49:22 hattivatti-tietokone gdm-password][2798]: gkr-pam: unable to locate daemon control file
Mar 18 16:55:05 hattivatti-tietokone systemd-coredump[7650]: Failed to send coredump datagram: Connectio>
Mar 18 16:55:05 hattivatti-tietokone polkitd[866]: Error converting subject to JS object: Process 3409 t>
Mar 18 16:55:05 hattivatti-tietokone dbus-broker-launch[837]: Activation request for 'org.freedesktop.Av>
Mar 18 16:55:06 hattivatti-tietokone dbus-broker-launch[837]: Activation request for 'org.freedesktop.Co>
Mar 18 16:55:06 hattivatti-tietokone dbus-broker-launch[837]: Activation request for 'org.freedesktop.nm>
Mar 18 16:55:06 hattivatti-tietokone dbus-broker-launch[837]: Activation request for 'org.freedesktop.nm>
Mar 18 16:55:06 hattivatti-tietokone dbus-broker-launch[837]: Activation request for 'org.freedesktop.nm>
lines 1-13/13 (END)

Try disconnecting your monitor and shutting down either or both previous ways.

Hello, MatH. I did indeed try that and it does seem to be forcibly restarting still. I’m curious if something in the configuration of the computer itself changed when I made this change.

I’ve tried a second DP port on the GPU and that has not made any difference. I’ve also tried unplugging the cable as I’m powering it down.

ChatGPT seems to believe something in the OS is maybe calling it to wakeup or something. Most of my terminal outputs have been too long to paste. I might see if I can get it to produce a summary of what it believes the issue is or isn’t and share it here since it’s over my head and maybe someone else can make more sense of it.

System:

  • Fedora 43
  • NVIDIA GTX 1080 Ti
  • Proprietary NVIDIA driver via akmods
  • GNOME/Wayland
  • Monitor changed from DVI-era setup to an AOC monitor using DisplayPort

Symptoms:

  • The machine appears to power off/reboot on its own after the monitor change.
  • This is not just a black screen: last -x shows repeated real boot cycles.

What seems likely:

  1. Some software/user-space path is initiating a normal shutdown/poweroff.
    Why:
    • Logs show an orderly shutdown, not a crash/panic.
    • systemd-logind says: “The system will power off now!”
    • Filesystems unmount cleanly and systemd-shutdown runs normally.

What seems less likely:
2. A GPU crash / kernel panic / hard lockup.
Why:

  • The shutdown path looks graceful, not like a hard crash.
  • No obvious panic/oops/Xid-style smoking gun was found in the logs collected so far.
  1. A bad DisplayPort cable or a single bad GPU DP port.
    Why:

    • Tested multiple DP cables.
    • Tested multiple DP ports on the 1080 Ti.
    • Same behavior.
  2. Monitor USB / monitor hub / KVM weirdness.
    Why:

    • No USB devices are connected through the monitor.
  3. Front-panel power/reset switch as the main cause.
    Why:

    • We tested ignoring logind power/reboot key handling and behavior did not change.
    • That makes a simple hardware power-button path less convincing.

Still possible:
6. A Fedora/GNOME/Wayland/NVIDIA power-management or session-management issue.
Why:

  • The problem started right after the monitor swap.
  • But the current logs point more toward software-triggered shutdown than display failure.

Important log details:

  • echo $XDG_SESSION_TYPE = wayland
  • systemd-logind: “The system will power off now!”
  • Shutdown sequence is clean/orderly.
  • NVIDIA version in use: 580.126.18
  • nvidia-powerd reports unsupported system for this GPU, but that did not look like the direct cause.

Main question:
What could be causing Fedora 43 / GNOME / systemd-logind / systemd to initiate a normal poweroff on its own in this setup, especially after a monitor/display change?

This is an activation request… now is it relevant and what made dbus request it…?

Do you have ethernet and ‘wake on LAN’ active?

Also, please state if a post is an LLM so we can discount its hallucinations. (I dont want to come across as rude, :slight_smile: )


And def not. Human as can be. Just progressing slowly and it’s getting late in my timezone, so I’m just trying to use the options I have.

As far as I know, nothing should be able to wake it. I don’t have anything at the bios level anyways. Not as far as I can see/know. Firmware is up to date, as well.

I don’t have an ethernet or LAN cable connected, either. Just wifi.

1 Like

I’ve made a little progress!

ChatGPT suggested I attempt changing my “Deep Sleep” on my AsRock Mobo to “S5”. When that was enabled, it seems to stay fully powered off as far as I can tell.

It still does not tell me what the cause was, so it’s basically a bandaide. (And no, I didn’t mess with any bios settings prior to the monitor fiasco. I mean, I have in the past messed with them, but not recently).

ChatGPT felt I should say this:

The system appears to be shutting down normally, but with Deep Sleep disabled the motherboard was still allowing a wake/power-on event from S5; enabling Deep Sleep for S5 stopped the automatic power-back-on, so this looks like a firmware/standby-power wake issue rather than a Fedora/NVIDIA/display crash.

I’m not going to “close” this or mark it solved because I’m not sure that it classifies as such (I still don’t know what the cause was, even if I’ve found a workaround).

I’m mainly leaving this here in case someone who actually “gets” all of this can find some value in it, has some follow up questions (that don’t require a lot of technical skill to give), and if there actually is potentially a bug or adjustment that needs to be made somewhere on the Fedora end (regardless of what ChatGPT said. I’d prefer some expert opinion on that).

Anyways. Thanks again everyone. If it remains in active and closes itself, it was a pleasure. Otherwise I’ll try and pay attention if anyone asks for further troubleshooting.

It does classify as a solution. The bios setting seems to be the issue and is not fedora related at all. That means it was hardware/bios related and has nothing to do with the OS.

How so, @computersavvy ?

In the decade I’ve had the machine, I’ve never needed to toggle that deep sleep setting from anything other than disabled and it’s always powered down fine. And while toggling that setting is blocking the bad behavior, it doesn’t mean it was the cause of the bad behavior. Again. I disabled the setting probably 7 years ago or so and it’s never “needed” to be on since.

Furthermore, I’ve lost my ability to use any kind of wake up functionality now because this setting appears as if it would be blocking it. For the record, I don’t currently need it, but my point is, I’ve reduced the functionality of my computer in a way that I’ve never needed to do before just so a symptom would not appear. If I disable the setting, it will probably go right back to what it was doing before. If it were a medical condition, we’d call it “managed” not “cured.”

I’m not super tech savvy here, and I know that it’s possibly a hardware issue…it’s possible it could still have something to do with the OS, too. The cause isn’t definitively known, even if certain things are or are not more likely.

So, from that stance, yes, if the possibility exists that Fedora has some issues, it’d be nice if they were known. I like Fedora. I’d like to keep using Fedora. And I’d like people using Fedora to have the best possible user experience, and if something on my machine can assist with that and it’s not too hard to dig out, I’d be happy to provide that.

1 Like

Actually my statement above was incomplete.
The bios settings are somewhat relevant to newer kernels so it is possible that the issue cropped up with recent kernel & driver changes.

It may be worth a bug report to see if it can be reproduced and if there might be a fix.

1 Like

And I don’t even know what all goes into this with Fedora. The forums here are as deep as I’ve gone with reporting anything with them. I just always assumed members from the team watched the forums and created their own issues and such based on what they observed here. I didn’t see it listed in the help area, but now I see the bugzilla thing. I’m not even sure what I’d say there other than just kind of linking them to this post.

Since my last post, I have discovered that changing that setting has caused it to reset the LEDs on my keyboard, mouse, etc upon every boot cycle, and that is def something I’d prefer to not be the case if at all possible, so I’d def love if there were an option to not do that.

You’re suggesting there is a potential kernal issue?

Fedora seems to update quite regularly, and I’m regular about updating (I check every time I sit down and get up). With the updates being small and frequent, it’s entirely possible I may have updated something at a similar time to changing the monitor (as in just before), but in that case, I would have likely expected the computer to fail to shutdown just before I changed the monitor. My memory is fuzzy because it’s been a couple days, but I don’t believe I updated right as I was changing my monitor. I can’t say 100% that I didn’t, but it’s the kind of detail that would have likely stood out in my mind as I was writing up this post to begin with.

Try to boot into another OS, but in my practice most mysterious problems with startup and reboot occurred when my power supply was slowly dying not able to give what is needed to new upgrades.

Linking them to this post is a great start, they will ask you questions if needed. They do not watch here, only in bugzilla. But they will look at this thread if you link them to it.