# Unable to see support for OpenGL 1.5 and above on linux

**URL:** <https://forums.developer.nvidia.com/t/unable-to-see-support-for-opengl-1-5-and-above-on-linux/61517>\
**Category:** Linux\
**Created:** [May 21, 2018, 11:31pm UTC](https://forums.developer.nvidia.com/t/unable-to-see-support-for-opengl-1-5-and-above-on-linux/61517 "2018-05-21T23:31:37Z")\
**Posts on this page:** 11\
**Page:** 1

<div class="post-metadata">

**Author:** ![hemantk65](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@hemantk65](https://forums.developer.nvidia.com/u/hemantk65)\
**Post date:** [May 21, 2018, 11:31pm UTC](https://forums.developer.nvidia.com/t/unable-to-see-support-for-opengl-1-5-and-above-on-linux/61517/1 "2018-05-21T23:31:37Z")

</div>

I am running a on an RHEL 7.4 machine with a Quadro K2200 and 390.48 drivers.  
This machine is unable to support anything above OpenGL 1.4

```auto
[h.challa@cbatch91806 ~]$ glxinfo | grep OpenGL
OpenGL vendor string: VMware, Inc.
OpenGL renderer string: Gallium 0.4 on llvmpipe (LLVM 3.8, 256 bits)
OpenGL version string: 1.4 (3.0 Mesa 17.0.6)
OpenGL extensions:

```

Steps used for installing drivers -

1. Create a New file: /etc/modprobe.d/nvidia-installer-disable-nouveau.conf with contents, as below:

```auto
blacklist nouveau
options nouveau modeset=0

```

2. Get NVIDIA drivers from:

```auto
http://www.nvidia.com/download/driverResults.aspx/134262/en-us

```

3. Install by running:

```auto
# sh NVIDIA-Linux-x86_64-390.48.run

```

Checking that the driver was installed correctly -

```auto
# lshw -C display | egrep -i 'product| driver' | grep -A1 K2200
       product: GM107GL [Quadro K2200]
       configuration: driver=nvidia latency=0
       product: GM107GL [Quadro K2200]
       configuration: driver=nvidia latency=0

```

I will attach the nvidia-bug-report log shortly.  
[glxinfo.log](https://forums.developer.nvidia.com/uploads/short-url/u6s4UTZ41ZT73NLeMI9shQO3XVN.log) (24.9 KB)  
[nvidia-installer.log](https://forums.developer.nvidia.com/uploads/short-url/cO8hAYND7wcAgsPPdOmyPfKuKK.log) (27.1 KB)  
[nvidia-bug-report.log.gz](https://forums.developer.nvidia.com/uploads/short-url/oM1VNQz8vyDCEU8AuQB2x7MY20f.gz) (115 KB)

---

<div class="post-metadata">

**Author:** ![generix](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/generix/32/16024_2.png) [@generix](https://forums.developer.nvidia.com/u/generix)\
**Post date:** [May 22, 2018, 9:19am UTC](https://forums.developer.nvidia.com/t/unable-to-see-support-for-opengl-1-5-and-above-on-linux/61517/2 "2018-05-22T09:19:34Z")

</div>

This looks like you’re trying to do remote GL, are you using ssh?  
The driver is installed fine, there’s just no monitor connected to the Quadro, so the Xserver is stopping. So either connect a monitor to it or add  
Option “AllowEmptyInitialConfiguration” “true”  
to the device section of your xorg.conf. Maybe also add the BusID of the Quadro you want to use.  
If you want a remote display, use x11vnc to connect to the running Xserver.

---

<div class="post-metadata">

**Author:** ![hemantk65](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@hemantk65](https://forums.developer.nvidia.com/u/hemantk65)\
**Post date:** [May 22, 2018, 9:43pm UTC](https://forums.developer.nvidia.com/t/unable-to-see-support-for-opengl-1-5-and-above-on-linux/61517/3 "2018-05-22T21:43:49Z")

</div>

Yes, I think I forgot to mention that. This is a headless server and I am trying to get remote GL working through ssh.

I generated an xorg.conf file using the following command, rebooted and tested again with both the default display and ‘DISPLAY=:0’ but I still see the same issue.

```auto
#nvidia-xconfig -a --allow-empty-initial-configuration

```

[xorg.conf.txt](https://forums.developer.nvidia.com/uploads/short-url/o1xtVOz3U08BkCz485UpCVK33HX.txt) (2.18 KB)  
[Xorg.0.log](https://forums.developer.nvidia.com/uploads/short-url/xyiBNonKjbPJo5KRs8loVzasMpb.log) (4.53 KB)

---

<div class="post-metadata">

**Author:** ![generix](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/generix/32/16024_2.png) [@generix](https://forums.developer.nvidia.com/u/generix)\
**Post date:** [May 22, 2018, 9:53pm UTC](https://forums.developer.nvidia.com/t/unable-to-see-support-for-opengl-1-5-and-above-on-linux/61517/4 "2018-05-22T21:53:45Z")

</div>

IndirectGL means that rendering takes place on your client machine. So both machines will have to have a nvidia card and driver installed.

---

<div class="post-metadata">

**Author:** ![hemantk65](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@hemantk65](https://forums.developer.nvidia.com/u/hemantk65)\
**Post date:** [May 22, 2018, 9:58pm UTC](https://forums.developer.nvidia.com/t/unable-to-see-support-for-opengl-1-5-and-above-on-linux/61517/5 "2018-05-22T21:58:44Z")

</div>

Is there a way to only have the remote machine with an nvidia card and the client machine is just used to ssh and see the display?

---

<div class="post-metadata">

**Author:** ![generix](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/generix/32/16024_2.png) [@generix](https://forums.developer.nvidia.com/u/generix)\
**Post date:** [May 23, 2018, 8:01am UTC](https://forums.developer.nvidia.com/t/unable-to-see-support-for-opengl-1-5-and-above-on-linux/61517/6 "2018-05-23T08:01:54Z")

</div>

Like said, start an xserver on the remote machine, you might have to fake a monitor using  
Option “ConnectedMonitor” “DFP-0”  
if AllowEmptyInitialConfiguration isn’t sufficient and run x11vnc to share that screen. Then simply use an vnc client to connect.  
Otherwise, you would have to use something like VirtualGL.

---

<div class="post-metadata">

**Author:** ![hemantk65](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@hemantk65](https://forums.developer.nvidia.com/u/hemantk65)\
**Post date:** [June 6, 2018, 6:33pm UTC](https://forums.developer.nvidia.com/t/unable-to-see-support-for-opengl-1-5-and-above-on-linux/61517/7 "2018-06-06T18:33:29Z")

</div>

We did add a monitor to the machine but it still claims that no display devices have been connected. The machine is at a remote location and we are checking if it is definitely plugged into the GPU and not the motherboard I/O.

I also wanted to check if the xorg.conf needs to be updated once the monitor has been connected?

```auto
# nvidia-xconfig --query-gpu-info
Number of GPUs: 2

GPU #0:
  Name : Quadro K2200
  UUID : GPU-55661a88-cffc-eb7f-38be-6a81dc700dcc
  PCI BusID : PCI:59:0:0

  Number of Display Devices: 0

GPU #1:
  Name : Quadro K2200
  UUID : GPU-9fb03c8b-6bc5-8630-1ee4-dc4b5b73756a
  PCI BusID : PCI:175:0:0

  Number of Display Devices: 0

```

[xorg.conf.log](https://forums.developer.nvidia.com/uploads/short-url/8ksLfKdrDrcm5uwoTdFwdCpV0pH.log) (2.36 KB)

---

<div class="post-metadata">

**Author:** ![generix](https://sea2.discourse-cdn.com/nvidia/user_avatar/forums.developer.nvidia.com/generix/32/16024_2.png) [@generix](https://forums.developer.nvidia.com/u/generix)\
**Post date:** [June 6, 2018, 7:01pm UTC](https://forums.developer.nvidia.com/t/unable-to-see-support-for-opengl-1-5-and-above-on-linux/61517/8 "2018-06-06T19:01:14Z")

</div>

This is rather odd. It seems like your xorg.conf isn’t even used.  
Please use this minimal /etc/X11/xorg.conf

```auto
Section "Device"
    Identifier "Device0"
    Driver "nvidia"
    VendorName "NVIDIA Corporation"
    BoardName "Quadro K2200"
    BusID "PCI:59:0:0"
    Option "AllowEmptyInitialConfiguration" "True"
    Option "ConnectedMonitor" "DFP-0"
EndSection

```

And attach the resulting Xorg.0.log

---

<div class="post-metadata">

**Author:** ![hemantk65](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@hemantk65](https://forums.developer.nvidia.com/u/hemantk65)\
**Post date:** [June 6, 2018, 10:32pm UTC](https://forums.developer.nvidia.com/t/unable-to-see-support-for-opengl-1-5-and-above-on-linux/61517/9 "2018-06-06T22:32:26Z")

</div>

The monitor was in the wrong port, so we moved it to the GPU. Swapped out the config file and restarted the system.

```auto
[h.challa@cbatch91806 ~]$ nvidia-xconfig --query-gpu-info
Number of GPUs: 2

GPU #0:
  Name : Quadro K2200
  UUID : GPU-55661a88-cffc-eb7f-38be-6a81dc700dcc
  PCI BusID : PCI:59:0:0

  Number of Display Devices: 1

  Display Device 0 (CRT-0):
      EDID Name : Samsung S24C650
      Minimum HorizSync : 30.000 kHz
      Maximum HorizSync : 81.000 kHz
      Minimum VertRefresh : 56 Hz
      Maximum VertRefresh : 75 Hz
      Maximum PixelClock : 170.000 MHz
      Maximum Width : 1920 pixels
      Maximum Height : 1080 pixels
      Preferred Width : 1920 pixels
      Preferred Height : 1080 pixels
      Preferred VertRefresh : 60 Hz
      Physical Width : 520 mm
      Physical Height : 290 mm

GPU #1:
  Name : Quadro K2200
  UUID : GPU-9fb03c8b-6bc5-8630-1ee4-dc4b5b73756a
  PCI BusID : PCI:175:0:0

  Number of Display Devices: 0

```

---

<div class="post-metadata">

**Author:** ![hemantk65](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@hemantk65](https://forums.developer.nvidia.com/u/hemantk65)\
**Post date:** [June 6, 2018, 10:34pm UTC](https://forums.developer.nvidia.com/t/unable-to-see-support-for-opengl-1-5-and-above-on-linux/61517/10 "2018-06-06T22:34:07Z")

</div>

Will update another Xorg.0.log in a few minutes. Looks like the config wasn’t swapped out correctly.

---

<div class="post-metadata">

**Author:** ![hemantk65](https://developer.download.nvidia.com/images/forums/profile-default-devtalk-84.png) [@hemantk65](https://forums.developer.nvidia.com/u/hemantk65)\
**Post date:** [June 6, 2018, 11:10pm UTC](https://forums.developer.nvidia.com/t/unable-to-see-support-for-opengl-1-5-and-above-on-linux/61517/11 "2018-06-06T23:10:50Z")

</div>

Here is the updated Xorg.0.log  
[Xorg.0.log](https://forums.developer.nvidia.com/uploads/short-url/pY71rtATqY25sSnp45uaLIsLa68.log) (25.6 KB)