Running tigervncserver@:1.service causes Xorg.0.log to fill up

[Richard Mixon]

I'm running both TigerVNC  1.12 (tigervnc-standalone-server on Ubuntu 22.04) and noVNC 1.40. 

The Xorg.0.log file fills up with the following repeating sequence:

[   904.333] (II) modeset(0): EDID vendor "DVA", prod id 60001
[   904.333] (II) modeset(0): DDCModeFromDetailedTiming: 1280x800 Warning: We only handle separate sync.
[   904.333] (II) modeset(0): Using hsync ranges from config file
[   904.333] (II) modeset(0): Using vrefresh ranges from config file
[   904.333] (II) modeset(0): Printing DDC gathered Modelines:
[   904.333] (II) modeset(0): Modeline "1024x768"x0.0   65.00  1024 1048 1184 1344  768 771 777 806 -hsync -vsync (48.4 kHz eP)
[   904.333] (II) modeset(0): Modeline "1280x768"x0.0   79.50  1280 1344 1472 1664  768 771 778 798 -hsync +vsync (47.8 kHz e)
[   904.333] (II) modeset(0): Modeline "1360x768"x0.0   85.50  1360 1424 1536 1792  768 771 777 795 +hsync +vsync (47.7 kHz e)
[   904.333] (II) modeset(0): Modeline "1366x768"x0.0   85.50  1366 1436 1579 1792  768 771 774 798 +hsync +vsync (47.7 kHz e)
[   904.333] (II) modeset(0): Modeline "1280x800"x0.0   83.50  1280 1352 1480 1680  800 803 809 831 -hsync +vsync (49.7 kHz e)
[   904.333] (II) modeset(0): Modeline "1440x900"x0.0  106.50  1440 1520 1672 1904  900 903 909 934 -hsync +vsync (55.9 kHz e)
[   904.333] (II) modeset(0): Modeline "1680x1050"x0.0  146.25  1680 1784 1960 2240  1050 1053 1059 1089 -hsync +vsync (65.3 kHz e)
[   904.333] (II) modeset(0): Modeline "800x600"x0.0   40.00  800 840 968 1056  600 601 605 628 +hsync +vsync (37.9 kHz e)
[   904.333] (II) modeset(0): Modeline "640x480"x0.0   25.18  640 656 752 800  480 490 492 525 -hsync -vsync (31.5 kHz e)
[   904.333] (II) modeset(0): Modeline "720x400"x0.0   28.32  720 738 846 900  400 412 414 449 -hsync +vsync (31.5 kHz e)
[   904.334] (II) modeset(0): Modeline "1280x960"x0.0  108.00  1280 1376 1488 1800  960 961 964 1000 +hsync +vsync (60.0 kHz e)
[   904.334] (II) modeset(0): Modeline "1280x1024"x0.0  108.00  1280 1328 1440 1688  1024 1025 1028 1066 +hsync +vsync (64.0 kHz e)
[   904.334] (II) modeset(0): Modeline "1280x720"x60.0   74.48  1280 1336 1472 1664  720 721 724 746 -hsync +vsync (44.8 kHz e)
[   904.334] (II) modeset(0): Modeline "1400x1050"x0.0  101.00  1400 1448 1480 1560  1050 1053 1057 1080 +hsync -vsync (64.7 kHz e)
[   904.334] (II) modeset(0): Modeline "1600x1200"x0.0  162.00  1600 1664 1856 2160  1200 1201 1204 1250 +hsync +vsync (75.0 kHz e)
[   904.334] (II) modeset(0): Modeline "1920x1200"x0.0  154.00  1920 1968 2000 2080  1200 1203 1209 1235 +hsync -vsync (74.0 kHz e)

The sequence repeats faster than every second. If I stop the tigervncserver@:1.service service the messages stop.

I've tried changing the resolution in the ~?.vnc/'config file but that does not seem to help.

Below is the info on my hardware setup (it's an embedded box with Atom CPU)

Thank you much ... in advance - Richard

n2kowner@RichZ3:~$ sudo lshw -short
H/W path           Device     Class          Description
========================================================
                              system         VT_mIPC_BT35M X64 (Type1 - SKU0)
/0                            bus            Type2 - Board Product Name
/0/0                          memory         64KiB BIOS
/0/4                          processor      Intel(R) Atom(TM) CPU  E3845  @ 1.91GHz
/0/4/8                        memory         32KiB L1 cache
/0/4/9                        memory         1MiB L2 cache
/0/7                          memory         24KiB L1 cache
/0/25                         memory         2GiB System Memory
/0/25/0                       memory         2GiB SODIMM DDR3 Synchronous 1333 MHz (0.8 ns)
/0/25/1                       memory         SODIMM DDR3 Synchronous 1333 MHz (0.8 ns) [empty]
/0/100                        bridge         Atom Processor Z36xxx/Z37xxx Series SoC Transaction Register
/0/100/2           /dev/fb0   display        Atom Processor Z36xxx/Z37xxx Series Graphics & Display
/0/100/13          scsi1      storage        Atom Processor E3800 Series SATA AHCI Controller
/0/100/13/0.0.0    /dev/sda   disk           8012MB mSATA mini 3SE4
/0/100/13/0.0.0/1  /dev/sda1  volume         127MiB Windows FAT volume
/0/100/13/0.0.0/2  /dev/sda2  volume         7512MiB EXT4 volume
/0/100/14                     bus            Atom Processor Z36xxx/Z37xxx, Celeron N2000 Series USB xHCI
/0/100/14/0        usb1       bus            xHCI Host Controller
/0/100/14/0/2                 bus            USB 2.0 Hub
/0/100/14/0/2/2               communication  USBCAN (CAN-USB Gateway)
/0/100/14/0/2/4               bus            USB hub
/0/100/14/0/2/4/1             communication  VCOM Port
/0/100/14/0/2/4/2  input21    input          eGalax Inc. eGalaxTouch EXC7200-747Cv1.002           Stylus
/0/100/14/0/3                 input          USB Receiver
/0/100/14/0/3/0    input17    input          Logitech M510
/0/100/14/0/4      input16    input          Dell Dell USB Keyboard
/0/100/14/1        usb2       bus            xHCI Host Controller
/0/100/18                     generic        Atom Processor Z36xxx/Z37xxx Series LPIO2 DMA Controller
/0/100/18.1                   bus            Atom Processor Z36xxx/Z37xxx Series LPIO2 I2C Controller #1
/0/100/1a                     generic        Atom Processor Z36xxx/Z37xxx Series Trusted Execution Engine
/0/100/1b          card0      multimedia     Atom Processor Z36xxx/Z37xxx Series High Definition Audio Controller
/0/100/1b/0        input25    input          HDA Intel PCH HDMI/DP,pcm=3
/0/100/1b/1        input26    input          HDA Intel PCH HDMI/DP,pcm=7
/0/100/1c                     bridge         Atom Processor E3800 Series PCI Express Root Port 1
/0/100/1c/0        eth0       network        82574L Gigabit Network Connection
/0/100/1c.1                   bridge         Atom Processor E3800 Series PCI Express Root Port 2
/0/100/1c.2                   bridge         Atom Processor E3800 Series PCI Express Root Port 3
/0/100/1e                     generic        Atom Processor Z36xxx/Z37xxx Series LPIO1 DMA Controller
/0/100/1e.5                   bus            Atom Processor Z36xxx/Z37xxx Series LPIO1 SPI Controller
/0/100/1f                     bridge         Atom Processor Z36xxx/Z37xxx Series Power Control Unit
/0/100/1f/0                   system         PnP device PNP0b00
/0/100/1f/1                   system         PnP device PNP0c02
/0/100/1f/2                   input          PnP device PNP0303
/0/100/1f/3                   input          PnP device PNP0f03
/0/100/1f/4                   communication  PnP device PNP0501
/0/100/1f/5                   communication  PnP device PNP0501
/0/100/1f/6                   system         PnP device PNP0c02
/0/100/1f/7                   system         PnP device PNP0c31
/0/100/1f.3                   bus            Atom Processor E3800/CE2700 Series SMBus Controller
/1                            power          SR Real Battery
/2                            power          OEM_Define5
/3                 input0     input          Power Button
/4                 input1     input          Sleep Button
/5                 input2     input          Power Button
/6                 input6     input          Video Bus
n2kowner@RichZ3:~$
H/W path           Device     Class          Description
========================================================
                              system         VT_mIPC_BT35M X64 (Type1 - SKU0)
/0                            bus            Type2 - Board Product Name
/0/0                          memory         64KiB BIOS
/0/4                          processor      Intel(R) Atom(TM) CPU  E3845  @ 1.91GHz
/0/4/8                        memory         32KiB L1 cache
/0/4/9                        memory         1MiB L2 cache
/0/7                          memory         24KiB L1 cache
/0/25                         memory         2GiB System Memory
/0/25/0                       memory         2GiB SODIMM DDR3 Synchronous 1333 MHz (0.8 ns)
/0/25/1                       memory         SODIMM DDR3 Synchronous 1333 MHz (0.8 ns) [empty]
/0/100                        bridge         Atom Processor Z36xxx/Z37xxx Series SoC Transaction Register
/0/100/2           /dev/fb0   display        Atom Processor Z36xxx/Z37xxx Series Graphics & Display
/0/100/13          scsi1      storage        Atom Processor E3800 Series SATA AHCI Controller
/0/100/13/0.0.0    /dev/sda   disk           8012MB mSATA mini 3SE4
/0/100/13/0.0.0/1  /dev/sda1  volume         127MiB Windows FAT volume
/0/100/13/0.0.0/2  /dev/sda2  volume         7512MiB EXT4 volume
/0/100/14                     bus            Atom Processor Z36xxx/Z37xxx, Celeron N2000 Series USB xHCI
/0/100/14/0        usb1       bus            xHCI Host Controller
/0/100/14/0/2                 bus            USB 2.0 Hub
/0/100/14/0/2/2               communication  USBCAN (CAN-USB Gateway)
/0/100/14/0/2/4               bus            USB hub
/0/100/14/0/2/4/1             communication  VCOM Port
/0/100/14/0/2/4/2  input21    input          eGalax Inc. eGalaxTouch EXC7200-747Cv1.002           Stylus
/0/100/14/0/3                 input          USB Receiver
/0/100/14/0/3/0    input17    input          Logitech M510
/0/100/14/0/4      input16    input          Dell Dell USB Keyboard
/0/100/14/1        usb2       bus            xHCI Host Controller
/0/100/18                     generic        Atom Processor Z36xxx/Z37xxx Series LPIO2 DMA Controller
/0/100/18.1                   bus            Atom Processor Z36xxx/Z37xxx Series LPIO2 I2C Controller #1
/0/100/1a                     generic        Atom Processor Z36xxx/Z37xxx Series Trusted Execution Engine
/0/100/1b          card0      multimedia     Atom Processor Z36xxx/Z37xxx Series High Definition Audio Controller
/0/100/1b/0        input25    input          HDA Intel PCH HDMI/DP,pcm=3
/0/100/1b/1        input26    input          HDA Intel PCH HDMI/DP,pcm=7
/0/100/1c                     bridge         Atom Processor E3800 Series PCI Express Root Port 1
/0/100/1c/0        eth0       network        82574L Gigabit Network Connection
/0/100/1c.1                   bridge         Atom Processor E3800 Series PCI Express Root Port 2
/0/100/1c.2                   bridge         Atom Processor E3800 Series PCI Express Root Port 3
/0/100/1e                     generic        Atom Processor Z36xxx/Z37xxx Series LPIO1 DMA Controller
/0/100/1e.5                   bus            Atom Processor Z36xxx/Z37xxx Series LPIO1 SPI Controller
/0/100/1f                     bridge         Atom Processor Z36xxx/Z37xxx Series Power Control Unit
/0/100/1f/0                   system         PnP device PNP0b00
/0/100/1f/1                   system         PnP device PNP0c02
/0/100/1f/2                   input          PnP device PNP0303
/0/100/1f/3                   input          PnP device PNP0f03
/0/100/1f/4                   communication  PnP device PNP0501
/0/100/1f/5                   communication  PnP device PNP0501
/0/100/1f/6                   system         PnP device PNP0c02
/0/100/1f/7                   system         PnP device PNP0c31
/0/100/1f.3                   bus            Atom Processor E3800/CE2700 Series SMBus Controller
/1                            power          SR Real Battery
/2                            power          OEM_Define5
/3                 input0     input          Power Button
/4                 input1     input          Sleep Button
/5                 input2     input          Power Button
/6                 input6     input          Video Bus
n2kowner@RichZ3:~$

[Pierre Ossman]

On 08/02/2024 19:55, Richard Mixon wrote:
> I'm running both TigerVNC 1.12 (tigervnc-standalone-server on Ubuntu
> 22.04) and noVNC 1.40.
> The Xorg.0.log file fills up with the following repeating sequence:
>
> [ 904.333] (II) modeset(0): EDID vendor "DVA", prod id 60001

That TigerVNC server is, as the name suggests, standalone and does not
interact with the normal Xorg server. As such, I don't see that it could
be TigerVNC that is causing this issue.

I would guess that there is some service that affects both the local and
VNC session that gets confused and starts doing something silly.

Are you using the same user in VNC as well as the local session? That is
known to cause weird effects.

Regards
--
Pierre Ossman Software Development
Cendio AB https://cendio.com
Teknikringen 8 https://twitter.com/ThinLinc
583 30 Linköping https://facebook.com/ThinLinc
Phone: +46-13-214600

A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?

[Richard Mixon]

>On 08/02/2024 19:55, Richard Mixon wrote:
>> I'm running both TigerVNC 1.12 (tigervnc-standalone-server on Ubuntu
>> 22.04) and noVNC 1.40.
>> The Xorg.0.log file fills up with the following repeating sequence:
>>
>> [ 904.333] (II) modeset(0): EDID vendor "DVA", prod id 60001

>That TigerVNC server is, as the name suggests, standalone and does not
> interact with the normal Xorg server. As such, I don't see that it could
>be TigerVNC that is causing this issue.

>I would guess that there is some service that affects both the local and
>VNC session that gets confused and starts doing something silly.

>Are you using the same user in VNC as well as the local session? That is
>known to cause weird effects.

The tigervncserver@:1.service run as the session user, in this case n2kowner (the user whose session we wish to connect to).

The noVNC.service run as root - but again, the problem occurs even if the noVNC.service is stopped.

Thank you much - Richard

[Richard Mixon]

I reinstalled on a fresh OS install, but this time did not install package "tigervnc-xorg-extension", just installed via:

• sudo apt install tigervnc-standalone-server igervnc-common dbus-x11
  

With the noVNC.service stopped, I am still able to access via vncviewer over SSH tunnel. But the 'modeset(0): EDID vendor "DVA",' sequence is still filling up Xorg.0.log.

Configuration file /etc/tigervnc/vncserver-config-defaults and vncserver-config-mandatory have not been modified.

File /etc/tigervnc/vncserver.users has a single (uncommented) line: :1=n2kowner

Any ideas at all?

Thank you much - Richard

[Pierre Ossman]

On 09/02/2024 16:53, Richard Mixon wrote:
>
>> Are you using the same user in VNC as well as the local session? That is
>> known to cause weird effects.
>
> The tigervncserver@:1.service run as the session user, in this case
> n2kowner (the user whose session we wish to connect to).
> The noVNC.service run as root - but again, the problem occurs even if the
> noVNC.service is stopped.
>

I'm afraid vncserver is not for accessing an existing session. You
should look into the Xorg extension (libvnc.so) or x0vncserver for that
use case.

[Richard Mixon]

On Mon, Feb 12, 2024 at 8:27 AM Pierre Ossman <oss...@cendio.se> wrote:
On 09/02/2024 16:53, Richard Mixon wrote:
>
>> Are you using the same user in VNC as well as the local session? That is
>> known to cause weird effects.
>
> The tigervncserver@:1.service run as the session user, in this case
> n2kowner (the user whose session we wish to connect to).
> The noVNC.service run as root - but again, the problem occurs even if the
> noVNC.service is stopped.
>
I'm afraid vncserver is not for accessing an existing session. You
should look into the Xorg extension (libvnc.so) or x0vncserver for that
use case.

OK, should I x0vncserver even though the only issue we have is:

Flooding of the Xorg.0.log file - the VNC function works fine otherwise

??

Previously TigerVNC 1.7 running as an init.d service under older Ubuntu (12.04) worked fine, Our project/task was to update to a more recent version of Ubuntu (22.04) and it seemed to make sense to update TigerVNC and noVNC (different product I realize) also.

Our use case runs on an isolated IP network (marine industrial location):

1. Our black box running Ubuntu 22.04 starts up and auto logs in as user A and runs an application on X display "1:".
2. On same box TigerVNC server runs as user A via service tigervncserver@:1.service
3. On same box noVNC runs runs as user root via service noVNC.service, autoconnects to VNC user A on display "1:"
4. A smart display (contains embedded HTML browser) connects  to noVNC to display our application.

SmartDisplay-->noVNC-->TigerVNC-->application

Any advice is appreciated

Thank you much - Richard

--

Richard Mixon

2209 E. Constance Way

Phoenix, AZ 85042

mobile: 480-577-6834

[Pierre Ossman]

On 12/02/2024 18:56, Richard Mixon wrote:
> On Mon, Feb 12, 2024 at 8:27 AM Pierre Ossman <oss...@cendio.se> wrote:
>> I'm afraid vncserver is not for accessing an existing session. You
>> should look into the Xorg extension (libvnc.so) or x0vncserver for that
>> use case.
>
>
> OK, should I x0vncserver even though the only issue we have is:
>
> Flooding of the Xorg.0.log file - the VNC function works fine otherwise
> ??
>

I'm afraid the setup you describe is simply not possible. vncserver
cannot be running on the same display as the local session. It would
simply refuse to start in that case. So there is some crucial detail
missing here.

>
> 1. Our black box running Ubuntu 22.04 starts up and auto logs in as

> user A and runs an application on X display "1:".

> 2. On same box TigerVNC server runs as user A via service tigervncserver@
> :1.service

Why is the system configured to both autologin, and to start vncserver?
What happens if you remove the local login?

[Richard Mixon]

> Flooding of the Xorg.0.log file - the VNC function works fine otherwise
> ??
>

I'm afraid the setup you describe is simply not possible. vncserver
cannot be running on the same display as the local session. It would
simply refuse to start in that case. So there is some crucial detail
missing here.

>
>     1. Our black box running Ubuntu 22.04 starts up and auto logs in as
>     user A and runs an application on X display "1:".
>     2. On same box TigerVNC server runs as user A via service tigervncserver@
>     :1.service

Why is the system configured to both autologin, and to start vncserver?
What happens if you remove the local login?

I see the confusion/what's missing. The autologin runs a startup script that does the following:

1. Sets "export DISPLAY=:1"
2. Runs a subscript that launches a management dialog on DISPLAY ":0" using this command:
• DISPLAY=:0 yad  --on-top --center --button=gtk-ok:0 --image=gtk-dialog-info --text="${msg}" --title="${title}"
  
3. Starts these services:
• sudo systemctl start tigervncserver@:1.service
  
• sudo systemctl start noVNC.service
  
4. Starts the main application (on DISPLAY ":1" of course)
5. At this point I can see the application - either
• Using an SSH tunnel and vncviewer on a different Linux computer; or
• Using an HTML browser from another computer pointing at the address noVNC advertises.

Thank you much - Richard

[Richard Mixon]

One other observation. File ~/.vnc/config has the following contents:

• #geometry=2000x1536
  geometry=1400x1050
  

When I changed it to this, 

• #geometry=2000x1536
  #geometry=1400x1050
  geometry=1920x1200
  

the volume of logging to Xorg.0.log reduced somewhat - but it still fills up within a day.

[Pierre Ossman]

On 14/02/2024 18:52, Richard Mixon wrote:
>
> I see the confusion/what's missing. The autologin runs a startup script
> that does the following:
>

Ah, that explains things.

But then we're back to the problem that running multiple sessions as the
same user often has weird effects. In this case, the main application is
also partially in one session, partially in another.

I would suggest testing the following:

1. Starting the vncserver as a different user than the local one

2. Starting the main application from inside the VNC session, rather
than from the local session

Hopefully, one or both of these will stop confusing whatever component
is fiddling with the display settings.

[Richard Mixon]

OK, things are working fine now - once I realized that when tigervncserver@:1.service starts it creates a complete new desktop session in addition to the original desktop session. You as much as said this,  I just did not understand at the time.

And the issue with the Xorg.0.log filling up was a application script loop that was starting the app over and over.

Thank you much. Now if I can figure out how to make noVNC into a proper systemd service unit I will be all set. But I'll try and submit that question from the noVNC mailing list once I"m approved.

Thank you much - Richard