You are not logged in.
I am using Linux-zen. So recently a while back i installed NVIDIA beta dkms and NVIDIA beta utils from AUR replacing the stable packages from official repos and everything seemed fine. I got an update from AUR and installed it but i after restarting, it seems to get stuck up on tlp system startup/shutdown. luckily i had a timeshift snap just before the upgrade and was able to revert back but rn i have stalled upgrading nvidia cause whenever i upgrade it, i am not able to boot up. I tried reinstalling the stable package but the same thing. What to do?
I am upgrading nvidia from 440.59-1 to 440.64-1. Please lemme know any other details if i need to post, Im a newbie. Also another thing interestingly is that i have set up howdy and when ever i update the nvidia drivers, it doesnt work showing some error "Error while calling cudaGetDevice(&the_device_id)" dont know if its relevant or not.
Last edited by jessepinkman2802 (2020-03-10 16:34:40)
Offline
Post the journal from a failing boot in [ code ] [ /code] tags, e.g.
sudo journalctl -b-1where the -1 refers to a the boot prior to this one. Maybe also post the output during the update.
That userspace utilities run into problems after the update is normal, you need to reboot so that kernel driver/userspace libraries for the nvidia modules are back in sync.
Mod note: Not an installation issue, moving to NC
Offline
Post the journal from a failing boot in [ code ] [ /code] tags, e.g.
sudo journalctl -b-1where the -1 refers to a the boot prior to this one. Maybe also post the output during the update.
That userspace utilities run into problems after the update is normal, you need to reboot so that kernel driver/userspace libraries for the nvidia modules are back in sync.
Mod note: Not an installation issue, moving to NC
this is the error is i am getting in howdy
Exception ignored in thread started by: <function init_detector at 0x7fd3d14021f0>
Traceback (most recent call last):
File "/lib/security/howdy/compare.py", line 43, in init_detector
face_encoder = dlib.face_recognition_model_v1(PATH + "/dlib-data/dlib_face_recognition_resnet_model_v1.dat")
RuntimeError: Error while calling cudaGetDevice(&the_device_id) in file /home/red/.cache/yay/python-dlib-cuda/src/dlib-19.19/dlib/cuda/gpu_data.cpp:201. code: 804, reason: forward compatibility was attempted on non supported HW
^CTraceback (most recent call last):
File "/lib/security/howdy/compare.py", line 154, in <module>
lock.acquire()I just updated nvidia and am pretty sure i wont be able to boot up, so do you want me to type that code in the terminal from boot screen?
Last edited by jessepinkman2802 (2020-03-10 14:25:57)
Offline
You will regardless of what happens with the nvidia module able to boot to a terminal with
systemd.unit=multi-user.targeton the kernel parameters and then upload the output of that command https://wiki.archlinux.org/index.php/Li … in_clients
Offline
You will regardless of what happens with the nvidia module able to boot to a terminal with
systemd.unit=multi-user.targeton the kernel parameters and then upload the output of that command https://wiki.archlinux.org/index.php/Li … in_clients
I actually fixed it, I ran
mkinitcpio -p linuxand rebooted. It worked. It showed some kernel version mismatch or something when I ran the journal command.
Offline
Typically you forgot to mount the boot parrtition when this happens.
Please always remember to mark resolved threads by editing your initial posts subject - so others will know that there's no task left, but maybe a solution to find.
Thanks.
Offline