KVM sometimes does not wake screen (Debian)

Level1 Store KVM Helpdesk
1

KVM Model:

HDMI 2.1 KVM w/EDID & Serial Control - Single Monitor - Four Computer

USB Interface (5-gigabit or 10-gigbit):

10-gigabit

OS

PikaOS (Debian-based)

CPU

AMD Ryzen 5 7600 6-Core Processor

GPU

NVIDIA GeForce RTX 4070 SUPER

Motherboard

MSI MPG B650I EDGE WIFI

Problem:

Similarly to this issue in the FAQs, my Debian-based PC sometimes will not wake when connected via the KVM. All the USB inputs are consistently detected, but the display won’t turn on. When it doesn’t, I simply keep unplugging and plugging the HDMI until it eventually does work.

This occurs whether I plug it into my mobo HDMI or my graphics card HDMI.

Other Notes:

As suggested in the above post, I SSHd into my machine to see the output of journalctl and dmesg after plugging in my PC to the KVM (see link for content):

journalctl --follow

dmesg -w

When the screen does wake, I see the following output in my journalctl

gnome-shell[3747] Created gbm renderer for '/dev/dri/card1'
gnome-shell[3747] Failed to create colord device for 'xandr-Samsung ...': already exists
2

What monitor?

3

Samsung LS27R75-HTOM900006

4

Hm, try seeing if the eco plus or eco timer is on in the monitor settings? Those should be off

5

Hi Amber, it looks like they are both off.

6

Hmmm, weird. I’m sure EDID emulation would probably fix the issue. EDID emulation usually makes it so the monitor doesn’t disappear/re-handshake the connection. i.e. the disconnection and sleep states don’t happen. I’m not sure exactly how to do this on Linux… I have seen Wendell recommend xrandr

https://manpages.debian.org/unstable/x11-xserver-utils/xrandr.1.en.html

We also have a hardware feeder, but software should work as well.

7

Will continue to debug and update you if I find a solution here! So far it doesn’t wake more often than it does, so I’ll be on top of this :slight_smile:

8

Hi @Level1_Amber, I’m not having any luck in my research on how this works. Any tips on how to proceed? It seems very random when the KVM does/does not work for my Linux setup.

If you have other advice on how to proceed with debugging, that would be much appreciated.

9

@wendell

10

did you have any luck setting up xrandr? It could be a cabling issue as when the monitor comes back it takes a few handshakes and will give up. Thats why unplugging and plugging fixes it – it does more handshakes (forces more handshakes).

xrandr solves it by just telling the gpu to always assume the monitor is there so it sort of skips the handshake process.

sometimes theres a clue about that in dmesg but I didnt see it here

11

Can I ask what you mean by “setting up xrandr”? The command works for me, here’s what I’m doing ssh’d into my linux machine (which is currently connected to the KVM, display not awake):

> export DISPLAY=:0 XAUTHORITY=[...]

> xrandr
Screen 0: minimum 16 x 16, current 2560 x 1440, maximum 32767 x 32767
HDMI-1 connected primary 2560x1440+0+0 (normal left inverted right x axis y axis) 600mm x 340mm
   2560x1440    143.91*+
   1920x1440    143.90  
   1600x1200    143.89  
   1440x1080    143.80  
   1400x1050    143.89  
   1280x1024    143.79  
   1280x960     143.86  
   1152x864     143.92  
   1024x768     143.87  
   800x600      143.83  
   640x480      143.85  
   320x240      142.05  
   1920x1200    143.89  
   1680x1050    143.88  
   1440x900     143.86  
   1280x800     143.84  
   1152x720     143.77  
   960x600      143.72  
   928x580      143.50  
   800x500      143.68  
   768x480      143.69  
   720x480      143.85  
   640x400      143.37  
   320x200      141.40  
   2048x1152    143.88  
   1920x1080    143.88  
   1600x900     143.93  
   1368x768     143.77  
   1280x720     143.67  
   1024x576     143.91  
   864x486      143.63  
   720x400      143.88  
   640x350      143.57

and btw, here’s the dgrep output (sorry i think the copy-paste link expired):

> sudo dmesg -w



[ 1356.615705] usb 1-1.2: new high-speed USB device number 21 using xhci_hcd
[ 1356.815752] usb 1-1.2: new high-speed USB device number 22 using xhci_hcd
[ 1356.922165] usb 1-1-port2: attempt power cycle
[ 1357.622854] usb 1-1.2: new full-speed USB device number 23 using xhci_hcd
[ 1357.743366] usb 1-1.2: New USB device found, idVendor=0495, idProduct=8095, bcdDevice= 0.01
[ 1357.743371] usb 1-1.2: New USB device strings: Mfr=1, Product=2, SerialNumber=4
[ 1357.743373] usb 1-1.2: Product: M15i Mobile Headphone Amplifier
[ 1357.743374] usb 1-1.2: Manufacturer: Questyle
[ 1357.743375] usb 1-1.2: SerialNumber: 2024503357
[ 1358.961465] usbhid 1-1.2:1.2: couldn't find an input interrupt endpoint
[ 1358.963483] hid-generic 0003:0495:8095.0016: hiddev98,hidraw2: USB HID v1.00 Device [Questyle M15i Mobile Headphone Amplifier] on usb-0000:0e:00.0-1.2/input4
[ 1359.165620] usb 1-1.4: new full-speed USB device number 24 using xhci_hcd
[ 1359.425774] usb 1-1.4: New USB device found, idVendor=fffe, idProduct=004b, bcdDevice= 1.06
[ 1359.425778] usb 1-1.4: New USB device strings: Mfr=1, Product=2, SerialNumber=0
[ 1359.425780] usb 1-1.4: Product: Neo Ergo
[ 1359.425781] usb 1-1.4: Manufacturer:  
[ 1359.563293] input:   Neo Ergo as /devices/pci0000:00/0000:00:02.1/0000:03:00.0/0000:04:0c.0/0000:0e:00.0/usb1/1-1/1-1.4/1-1.4:1.0/0003:FFFE:004B.0017/input/input41
[ 1359.637781] hid-generic 0003:FFFE:004B.0017: input,hidraw3: USB HID v1.11 Keyboard [  Neo Ergo] on usb-0000:0e:00.0-1.4/input0
[ 1359.638164] hid-generic 0003:FFFE:004B.0018: hiddev99,hidraw4: USB HID v1.11 Device [  Neo Ergo] on usb-0000:0e:00.0-1.4/input1
[ 1359.638632] input:   Neo Ergo Mouse as /devices/pci0000:00/0000:00:02.1/0000:03:00.0/0000:04:0c.0/0000:0e:00.0/usb1/1-1/1-1.4/1-1.4:1.2/0003:FFFE:004B.0019/input/input42
[ 1359.638751] input:   Neo Ergo System Control as /devices/pci0000:00/0000:00:02.1/0000:03:00.0/0000:04:0c.0/0000:0e:00.0/usb1/1-1/1-1.4/1-1.4:1.2/0003:FFFE:004B.0019/input/input43
[ 1359.690001] input:   Neo Ergo Consumer Control as /devices/pci0000:00/0000:00:02.1/0000:03:00.0/0000:04:0c.0/0000:0e:00.0/usb1/1-1/1-1.4/1-1.4:1.2/0003:FFFE:004B.0019/input/input44
[ 1359.690099] input:   Neo Ergo Keyboard as /devices/pci0000:00/0000:00:02.1/0000:03:00.0/0000:04:0c.0/0000:0e:00.0/usb1/1-1/1-1.4/1-1.4:1.2/0003:FFFE:004B.0019/input/input45
[ 1359.725932] hid-generic 0003:FFFE:004B.0019: input,hidraw6: USB HID v1.11 Mouse [  Neo Ergo] on usb-0000:0e:00.0-1.4/input2
12

Perhaps just disable screen sleeping… if the problem is really the lack of being able to wake.

When I see “single monitor four computer”… Does it definitely have its own power source? Does it share ground with the screen you are trying to power? And finally, does having more monitors plugged in solve the issue? (if its the last one just get dummy HDMI)…

Something just smells like hardware issue to me, doesn’t mean it is one though. Maybe you got a dud, maybe this really is software.

13

To clarify, by “wake”, I mean whenever I turn my PC on from suspend/off, the screen does not turn on. What you suggested @GridAndGhost was Amber’s first suggestion, no luck there - but thanks for the suggestion anyway.

Whenever it randomly starts working, I can switch between my PC and Mac (the other connected PC) with no issues. However, since time of original post, my display has only woken once.

14

you will need to read a little more on what xrandr does for you. basically you can set a modeline and your monitor will always be in that mode instead of being probed by your gpu

it probably still is the handshake/cabling issue but that does away with the handshake that fails intermittently

15

I’m going to go with your screen not liking the swapping or some part of your hardware not liking it… It could literally be your keyboard/mouse/audio if they’re not evenly distributed and asking for too much power/providing too much data. Soooo many things with KVMs lol, I’ve had mice not want to switch over and other mice switch over easily. Had to do with whether or not my kvm was powered, and manufacturer of the mouse. Silly stuff like that.