On Sat, Jan 18, 2020 at 08:24:55PM +0100, Cliff McDiarmid via blfs-support
wrote:
>
>
> >at a guess there is something wrong with that directory
> >(non-existent, or wrong owner or wrong perms). Possibly
> >>http://www.mgreene.org/?p=302[http://www.mgreene.org/?p=302]
>
> Thanks Ken. That's what I thought originally. Followed those suggestions as
> part of the 'solving' but it made no difference.
>
> Reinsalled GDM. No difference.
>
> >Pasting from that, since you cannot run X (note that his log was in
> >.local/share not .local)
> >Simple solution, chown -R gdm.gdm /var/lib/gdm
>
> This time did: 'chgrp gdm /var/lib/gdm/.local/share' and 'chown gdm
> /var/lib/gdm/.local/share'
>
> drwxrwx--T 6 gdm gdm 4096 Jan 17 22:50 .
> drwxr-xr-x 31 root root 4096 Jul 10 2019 ..
> drwx------ 6 gdm gdm 4096 Jan 18 15:34 .cache
> drwx------ 7 gdm gdm 4096 Apr 1 2019 .config
> -rw------- 1 root root 16 Apr 1 2019 .esd_auth
> -rw------- 1 root root 34826 Jan 17 00:24 .ICEauthority
> drwxr-xr-x 3 root root 4096 Mar 17 2019 .local
> drwx------ 3 gdm gdm 4096 Apr 1 2019 .nv
>
I notice that two of those directories are from January and March
2017. That suggests that the base LFS system is 8.2 or before ?
I don't use systemd, let alone gdm, so I can't comment on any known
issues along the way - but I start to wonder if versions have got
out of kilter. I have no idea what mix of packages / versions you
are running, perhaps summarising (gcc, binutils, glibc, kernel
headers, meson, systemd, kernel, glib, gtk3, gnome) might help if
someone who uses gdm reads this.
> I don't get the access errors as before but now get:
>
> (EE) NVIDIA: Failed to initialize the NVIDIA kernel module. Please see the
> [ 2.778] (EE) NVIDIA: system's kernel log for additional error
> messages and
> [ 2.778] (EE) NVIDIA: consult the NVIDIA README for details.
> No apparent reason for this. As mentioned have tried recovering the LFS from
> backup but the errors return.
>
> Cliff
Not sure if I've parsed that last sentenc, so let me try what I
think you are saying:
1. On the broken system, you changed ownership of
/var/lib/gdm/.local/share to gdm:gdm (but not the .local directory
itself - I have no idea what else, if anything that contains).
2. This possibly made progress, i.e. X tried to load the nvidia
module but failed. I regard that as 'possible' progress because I
suspect that happens after opening the log.
3. You restored from backup. I think you are saying that this used
to work for a while, even without changing perms, but now you are
back to the failure to open the log ? ~Or are you back to the
nvidia error ?
For the nvidia kernel module - had you updated the kernel since
your last usable backup ('usable' in the sense of "I've restored
this once and gdm worked") ?
Sorry, this is all outside my areas of adequacy :)
ĸen
--
The politics of wizardry were either very simple, and resolved by
someone ceasing to breathe, or as complex as one ball of yarn in a
room with three bright-eyed little kittens. - Unseen Academicals
--
http://lists.linuxfromscratch.org/listinfo/blfs-support
FAQ: http://www.linuxfromscratch.org/blfs/faq.html
Unsubscribe: See the above information page
