Quote from: woodl1 on September 06, 2017, 10:47:21 AM

Quote from: woodl1 on September 06, 2017, 06:21:54 AM

Quote from: woodl1 on September 05, 2017, 07:49:59 PM

Quote from: octomob on September 04, 2017, 12:05:24 AM

Anone managed to start NVOC version 19 with built-in VGA? I heared this version finally have a fix, for using on rigs built on GP106-100 GPUs

As I can see on the first page, it is still advised to start without monitor attached if you have only p106 in your rig.

I have tried to start full headless, with pure p106 mode enabled, but all I get are errors in logs, mining process doesn't start. Does anyone know why this could happen?

(I have asus z270-p with 8 p106-100, screen was attached only to modify bios settings as described for z270-a, usb stick was never attached with monitor at the same time)

Code:

Sep 5 15:13:51 m1-desktop systemd[1]: Started Detect the available GPUs and deal with any system changes.
Sep 5 15:13:51 m1-desktop systemd[1]: Starting Light Display Manager...
Sep 5 15:13:51 m1-desktop systemd[1]: Started Light Display Manager.
Sep 5 15:13:51 m1-desktop lightdm[5561]: /etc/modprobe.d is not a file
Sep 5 15:13:51 m1-desktop lightdm[5561]: message repeated 4 times: [ /etc/modprobe.d is not a file]
Sep 5 15:13:51 m1-desktop lightdm[5561]: update-alternatives: error: no alternatives for x86_64-linux-gnu_gfxcore_conf
Sep 5 15:13:52 m1-desktop lightdm[5561]: /etc/modprobe.d is not a file
Sep 5 15:13:52 m1-desktop lightdm[5561]: message repeated 4 times: [ /etc/modprobe.d is not a file]
Sep 5 15:13:52 m1-desktop lightdm[5561]: update-alternatives: error: no alternatives for x86_64-linux-gnu_gfxcore_conf
Sep 5 15:13:52 m1-desktop systemd[1]: lightdm.service: Main process exited, code=exited, status=1/FAILURE
Sep 5 15:13:52 m1-desktop systemd[1]: lightdm.service: Unit entered failed state.
Sep 5 15:13:52 m1-desktop systemd[1]: lightdm.service: Failed with result 'exit-code'.
Sep 5 15:13:52 m1-desktop systemd[1]: lightdm.service: Service hold-off time over, scheduling restart.
Sep 5 15:13:52 m1-desktop systemd[1]: Stopped Light Display Manager.
Sep 5 15:13:52 m1-desktop systemd[1]: Starting Detect the available GPUs and deal with any system changes...
Sep 5 15:13:54 m1-desktop gpu-manager[5613]: /etc/modprobe.d is not a file
Sep 5 15:13:54 m1-desktop gpu-manager[5613]: message repeated 4 times: [ /etc/modprobe.d is not a file]
Sep 5 15:13:54 m1-desktop gpu-manager[5613]: update-alternatives: error: no alternatives for x86_64-linux-gnu_gfxcore_conf
Sep 5 15:13:54 m1-desktop systemd[1]: Started Detect the available GPUs and deal with any system changes.
Sep 5 15:13:54 m1-desktop systemd[1]: Starting Light Display Manager...
Sep 5 15:13:54 m1-desktop systemd[1]: Started Light Display Manager.
Sep 5 15:13:54 m1-desktop lightdm[5640]: /etc/modprobe.d is not a file
Sep 5 15:13:54 m1-desktop lightdm[5640]: message repeated 4 times: [ /etc/modprobe.d is not a file]
Sep 5 15:13:54 m1-desktop lightdm[5640]: update-alternatives: error: no alternatives for x86_64-linux-gnu_gfxcore_conf
Sep 5 15:13:54 m1-desktop lightdm[5640]: /etc/modprobe.d is not a file
Sep 5 15:13:54 m1-desktop lightdm[5640]: message repeated 4 times: [ /etc/modprobe.d is not a file]
Sep 5 15:13:54 m1-desktop lightdm[5640]: update-alternatives: error: no alternatives for x86_64-linux-gnu_gfxcore_conf
Sep 5 15:13:54 m1-desktop systemd[1]: lightdm.service: Main process exited, code=exited, status=1/FAILURE
Sep 5 15:13:54 m1-desktop systemd[1]: lightdm.service: Unit entered failed state.
Sep 5 15:13:54 m1-desktop systemd[1]: lightdm.service: Failed with result 'exit-code'.
Sep 5 15:13:55 m1-desktop systemd[1]: lightdm.service: Service hold-off time over, scheduling restart.
Sep 5 15:13:55 m1-desktop systemd[1]: Stopped Light Display Manager.
Sep 5 15:13:55 m1-desktop systemd[1]: Starting Detect the available GPUs and deal with any system changes...
Sep 5 15:13:56 m1-desktop gpu-manager[5692]: /etc/modprobe.d is not a file
Sep 5 15:13:56 m1-desktop gpu-manager[5692]: message repeated 4 times: [ /etc/modprobe.d is not a file]
Sep 5 15:13:56 m1-desktop gpu-manager[5692]: update-alternatives: error: no alternatives for x86_64-linux-gnu_gfxcore_conf
Sep 5 15:13:56 m1-desktop systemd[1]: Started Detect the available GPUs and deal with any system changes.
Sep 5 15:13:56 m1-desktop systemd[1]: Starting Light Display Manager...
Sep 5 15:13:57 m1-desktop systemd[1]: Started Light Display Manager.

i also have following errors in syslog, in clean nvos019 install. I suppose it's kinda tails from previous usage of "SanDisk_Cruzer_Blade", because I've never used such device.

Code:

Sep 6 02:13:06 m1-desktop systemd[1]: dev-disk-by\x2did-usb\x2dSanDisk_Cruzer_Blade_4C530001260812105231\x2d0:0\x2dpart1.device: Job dev-disk-by\x2did-usb\x2dSanDisk_Cruzer_Blade_4C530001260812105231\x2d0:0\x2dpart1.device/start timed out.
Sep 6 02:13:06 m1-desktop systemd[1]: Timed out waiting for device dev-disk-by\x2did-usb\x2dSanDisk_Cruzer_Blade_4C530001260812105231\x2d0:0\x2dpart1.device.
Sep 6 02:13:06 m1-desktop systemd[1]: Dependency failed for /mnt/usb-SanDisk_Cruzer_Blade_4C530001260812105231-0:0-part1.
Sep 6 02:13:06 m1-desktop systemd[1]: mnt-usb\x2dSanDisk_Cruzer_Blade_4C530001260812105231\x2d0:0\x2dpart1.mount: Job mnt-usb\x2dSanDisk_Cruzer_Blade_4C530001260812105231\x2d0:0\x2dpart1.mount/start failed with result 'dependency'.
Sep 6 02:13:06 m1-desktop systemd[1]: dev-disk-by\x2did-usb\x2dSanDisk_Cruzer_Blade_4C530001260812105231\x2d0:0\x2dpart1.device: Job dev-disk-by\x2did-usb\x2dSanDisk_Cruzer_Blade_4C530001260812105231\x2d0:0\x2dpart1.device/start failed with result 'timeout'.
Sep 6 02:13:06 m1-desktop systemd[1]: dev-disk-by\x2duuid-55184403759586FB.device: Job dev-disk-by\x2duuid-55184403759586FB.device/start timed out.
Sep 6 02:13:06 m1-desktop systemd[1]: Timed out waiting for device dev-disk-by\x2duuid-55184403759586FB.device.
Sep 6 02:13:06 m1-desktop systemd[1]: Dependency failed for /mnt/55184403759586FB.
Sep 6 02:13:06 m1-desktop systemd[1]: mnt-55184403759586FB.mount: Job mnt-55184403759586FB.mount/start failed with result 'dependency'.
Sep 6 02:13:06 m1-desktop systemd[1]: Startup finished in 7.010s (kernel) + 1min 30.221s (userspace) = 1min 37.231s.
Sep 6 02:13:06 m1-desktop systemd[1]: dev-disk-by\x2duuid-55184403759586FB.device: Job dev-disk-by\x2duuid-55184403759586FB.device/start failed with result 'timeout'.

Can anybody guess what to do to make this thing work?

thanks in advance!

ok i'm talking with myself but this becomes useful to solve the problem

After a little research i've finally launched mining full headless rig of asus z270-p with 7 gp106-100 onboard. Problem was in xorg configuration, nvidia utility helped to solve it:

Code:

sudo nvidia-xconfig -a --cool-bits=28 --allow-empty-initial-configuration
sudo reboot

Now all works fine including overclocking and powerlimits, just one trouble left - if I connect 8-th GPU, nvOS hangs on boot Sad

I included an unverified logic which conditionally reconfigures the xorg.conf in v0019 when:

Code:

P106_100_FULL_HEADLESS_MODE="YES"

perhaps I should have included the forced reconfiguration everytime instead. you can change this portion of 3main to change this manually:

find this area in 3main by ctrl + f with gedit or using the vi cmds to search for: P106_100_FULL_HEADLESS_MODE

then change this code block:

Code:

UPDATE="YES"

if [ $P106_100_FULL_HEADLESS_MODE == "YES" ]
then
if grep -q "XORG_UPDATED" /home/m1/xorg_flag;
then
sleep 2
UPDATE="NO"
fi
if [ $UPDATE == "YES" ]
then
sudo nvidia-xconfig --enable-all-gpus --cool-bits=12
cd /home/m1
echo XORG_UPDATED > '/home/m1/xorg_flag'
sleep 4
echo "XORG UPDATED"
echo ''
echo "Rebooting in 5"
echo ''
echo "disconnect monitor if connected"
sleep 5
sudo reboot
fi
sleep 4

to:

Code:

UPDATE="YES"

if [ $UPDATE == "YES" ]
then
sudo nvidia-xconfig --enable-all-gpus --cool-bits=12
sleep 4
echo "XORG UPDATED"
fi
sleep 4