Skip to main content
Topic: [SOLVED] X will not start, dbus error (Read 1541 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

[SOLVED] X will not start, dbus error

X will not start.
I don't know if it's novideo or what. I'm on Artix S6. All I can think of that I did that may have caused this was a system update, followed by installing mullvad-vpn-bin and mullvad-vpn-s6, then a reboot leaving me where I am now.
I removed nvidia-dkms as a test then rebooted, but no success, x still won't start and gives the same error.
The error is "failed to connect to socket /run/dbus/system_bus_socket: No such file or directory"
Any help would be greatly appreciated.

Re: X will not start, dbus error

Reply #1
Do you have dbus and dbus-s6 installed?

Re: X will not start, dbus error

Reply #2
It looks like you do not have the dbus service running.

Re: X will not start, dbus error

Reply #3
Do you have dbus and dbus-s6 installed?
They were already installed, yeah.
It looks like you do not have the dbus service running.
Hmm ok. I've never had to mess with dbus before. Not sure if it's useful, but I tried the command dbus-launch. Appears to start a dbus server. X wouldn't start still. Am I missing something?

Re: X will not start, dbus error

Reply #4
Hmm ok. I've never had to mess with dbus before. Not sure if it's useful, but I tried the command dbus-launch. Appears to start a dbus server. X wouldn't start still. Am I missing something?
I'm just guessing here, but probably dbus and elogind are not in your default bundle. You can check that with sudo s6-rc-db contents default. The way we handled managing bundles changed earlier this year to match upstream. The support for the old way was dropped a while ago, so if you did not migrate to the new way, it's possible that bundle contents were dropped on an upgrade. Check out the wiki for details. You can use the s6-rc command to start/stop services as usual of course.

 

Re: X will not start, dbus error

Reply #5
I'm just guessing here, but probably dbus and elogind are not in your default bundle. You can check that with sudo s6-rc-db contents default. The way we handled managing bundles changed earlier this year to match upstream. The support for the old way was dropped a while ago, so if you did not migrate to the new way, it's possible that bundle contents were dropped on an upgrade. Check out the wiki for details. You can use the s6-rc command to start/stop services as usual of course.
So, I put dbus and elogind in my default bundles, and now they appear to automatically start running as they would before. I no longer get the error message in op.
But X still will not start. Appears to be an nvidia problem.  ::) Must be from my initial uninstall and reinstall of nvidia-dkms.
Error in log appears to be nouveau and nv modules not loading.

I've tried reinstalling all nvidia packages including nvidia-utils-s6. I've checked that nouveau is blacklisted in the confs (maybe I missed some?).
I don't know.



Re: X will not start, dbus error

Reply #8
This is a weird one. I'm not really sure nouveau failing to load is a fatal or error or not. It looks like it continues on just find after that but xorg decides to shut itself down for no particular reason.

Re: X will not start, dbus error

Reply #9
This is a weird one. I'm not really sure nouveau failing to load is a fatal or error or not. It looks like it continues on just find after that but xorg decides to shut itself down for no particular reason.
I'm no expert on any of this, but that's what it seemed like to me at first too.


Re: X will not start, dbus error

Reply #11
Probably you haven't recompiled nvidia-dkms, so the modules aren't loaded. Test with:
Code: [Select]
lsmod | grep nvidia[code]
[/code]

If there is no output, see: https://wiki.archlinux.org/title/Dynamic_Kernel_Module_Support
I ran
Code: [Select]
lsmod | grep nvidia
and got this output:
Code: [Select]
nvidia uvm     2609152 0
nvidia drm     73728 0
nvidia modeset    1163264 1 nvidia drm
nvidia    39153664 2 nvidia_uvm, nvidia_nodeset

I ran "sudo dkms autoinstall" and got
Code: [Select]
Error! Could not locate dkms.conf file.
File: /var/lib/dkms/exfat-5.8/1/source/dkms.conf

"dkms status" gives this
Code: [Select]
exfat/206.297a573: added
exfat-5.8/1: added
nvidia 510.68.02, 5.17.4-artix1-1, x86_64: installed
vmware-workstation/16.2.1_18811642, 5.17.4-artix1-1, x86_64: installed
I assume exfat 206.xx is the newer version and the older 5.8 version folder is conflicting? Should I remove the older folder if so?

I'd troubleshoot further, but I have to get to bed.

Re: X will not start, dbus error

Reply #12
From the X log, it seems you have two video cards, or at least Intel video kernel module in addition to nvidia. Can you give the output of
Code: [Select]
inxi -F
as root? What are your system specifications in relation to video hardware?

P.S. You should blacklist nouveau if you are using the proprietary driver.

Re: X will not start, dbus error

Reply #13
From the X log, it seems you have two video cards, or at least Intel video kernel module in addition to nvidia. Can you give the output of
Code: [Select]
inxi -F
as root? What are your system specifications in relation to video hardware?

P.S. You should blacklist nouveau if you are using the proprietary driver.
I am on a hybrid laptop with an integrated Intel GPU and dedicated NVIDIA GPU
Here's the output of "inxi -F" (I cannot find a good enough picture to text service to convert all of it to text)

Nouveau has been blacklisted since I started using Artix. I checked all the relevant ".conf" files I know, and that still seems to be the case.

Re: X will not start, dbus error

Reply #14
By the way, here is what the Xorg.0.log looks like with the "nvidia-dkms" package removed:

https://bin.snopyta.org/?17c1d7091363ee62#G3FWmPY84A167NvUHtxpa56zerMoFxnXpzWRWGWqcPMY

It really seems like xorg shuts down for no reason... hopefully I can figure this out and remember to do more backups lol.
As always, I appreciate the help. I don't feel like I'd get very far on most other forums.