Re: [PLUG] Slackware login pain
On 1/8/19 8:22 PM, Ben Koenig wrote: I'm trying to figure out what's broken here and I just can't see it. The subject of this thread was something about problems logging in. Turned out to be an issue with having the wrong nvidia driver installed. I should have started a new thread. I'll do that now. -- Regards, Dick Steffens ___ PLUG mailing list PLUG@pdxlinux.org http://lists.pdxlinux.org/mailman/listinfo/plug
Re: [PLUG] Slackware login pain
I'm trying to figure out what's broken here and I just can't see it. The subject of this thread was something about problems logging in. Turned out to be an issue with having the wrong nvidia driver installed. And now you say that X launches the nvidia-settings program runs as expected. Looks to me like everything is working the way it's supposed On Tue, Jan 8, 2019 at 9:54 AM Dick Steffens wrote: > On 1/7/19 3:37 PM, Ben Koenig wrote: > > Wait a second why is nvidia-settings trying to create an xorg.conf > > file? I thought the program was nvidia-xconfig .. > > It occurred to me that if I have a correct nVidia setup on my Ubuntu > machine, which has a very similar monitor setup, I could compare the > xorg.config files with each other and maybe edit the one on the > Slackware machine to match as appropriate. With the exception of the > nVidia model, the two files are identical. On the Ubuntu machine I have > a GeForce 210. > > I hadn't heard of nvidia-xconfig, so I ran it. > > root@ENU-2:/etc/X11# nvidia-xconfig > > Using X configuration file: "/etc/X11/xorg.conf". > Backed up file '/etc/X11/xorg.conf' as > '/etc/X11/xorg.conf.nvidia-xconfig-original' > Backed up file '/etc/X11/xorg.conf' as '/etc/X11/xorg.conf.backup' > New X configuration file written to '/etc/X11/xorg.conf' > > root@ENU-2:/etc/X11# > > When I log out I'm back at run level 3. I run startx again and it comes > up with mirrored screens. I run nvidia-settings from the menu and I can > uncover the smaller screen, put it to the left, set it as the primary > display, click apply, and I get what I want. The smaller monitor on the > left with the menu bar at the top and the launcher bar at the bottom, > and the larger monitor to the right with nothing but wallpaper. I can't > save the configuration from this invocation, so I quit. open a terminal, > log in as root, and it still shows the settings I set up as rsteff. I > can save to xorg.conf. I log out, log back in, and the setting are back > to mirrored, with the menu and launch bars back on the right hand monitor. > > I thought xorg.conf is where those settings were being stored. Is there > somewhere else I need to be looking? > > -- > Regards, > > Dick Steffens > > ___ > PLUG mailing list > PLUG@pdxlinux.org > http://lists.pdxlinux.org/mailman/listinfo/plug > ___ PLUG mailing list PLUG@pdxlinux.org http://lists.pdxlinux.org/mailman/listinfo/plug
Re: [PLUG] Slackware login pain
On 1/7/19 3:37 PM, Ben Koenig wrote: Wait a second why is nvidia-settings trying to create an xorg.conf file? I thought the program was nvidia-xconfig .. It occurred to me that if I have a correct nVidia setup on my Ubuntu machine, which has a very similar monitor setup, I could compare the xorg.config files with each other and maybe edit the one on the Slackware machine to match as appropriate. With the exception of the nVidia model, the two files are identical. On the Ubuntu machine I have a GeForce 210. I hadn't heard of nvidia-xconfig, so I ran it. root@ENU-2:/etc/X11# nvidia-xconfig Using X configuration file: "/etc/X11/xorg.conf". Backed up file '/etc/X11/xorg.conf' as '/etc/X11/xorg.conf.nvidia-xconfig-original' Backed up file '/etc/X11/xorg.conf' as '/etc/X11/xorg.conf.backup' New X configuration file written to '/etc/X11/xorg.conf' root@ENU-2:/etc/X11# When I log out I'm back at run level 3. I run startx again and it comes up with mirrored screens. I run nvidia-settings from the menu and I can uncover the smaller screen, put it to the left, set it as the primary display, click apply, and I get what I want. The smaller monitor on the left with the menu bar at the top and the launcher bar at the bottom, and the larger monitor to the right with nothing but wallpaper. I can't save the configuration from this invocation, so I quit. open a terminal, log in as root, and it still shows the settings I set up as rsteff. I can save to xorg.conf. I log out, log back in, and the setting are back to mirrored, with the menu and launch bars back on the right hand monitor. I thought xorg.conf is where those settings were being stored. Is there somewhere else I need to be looking? -- Regards, Dick Steffens ___ PLUG mailing list PLUG@pdxlinux.org http://lists.pdxlinux.org/mailman/listinfo/plug
Re: [PLUG] Slackware login pain
On 1/7/19 3:37 PM, Ben Koenig wrote: Wait a second why is nvidia-settings trying to create an xorg.conf file? I thought the program was nvidia-xconfig .. Are you running nvidia-settings from a graphical terminal? With X running? Yes. It's available from the Applications > Settings menu as NVIDIA X Server Settings One of the options is "Save to X configuration file" It's the approach I always used in Ubuntu. Another thing. I now have a mirrored situation whereas before, the single desktop was spread across both monitors. -- Regards, Dick Steffens ___ PLUG mailing list PLUG@pdxlinux.org http://lists.pdxlinux.org/mailman/listinfo/plug
Re: [PLUG] Slackware login pain
Wait a second why is nvidia-settings trying to create an xorg.conf file? I thought the program was nvidia-xconfig .. Are you running nvidia-settings from a graphical terminal? With X running? On Mon, Jan 7, 2019 at 3:15 PM Dick Steffens wrote: > On 1/7/19 2:59 PM, Ben Koenig wrote: > > the blacklist file prevents Nouveau from loading, but if you need to > double > > check that it actually worked you use 'lsmod' > > I pipe the output to grep to shorten the list: > > > > $ lsmod |grep nouveau > > root@ENU-2:~# lsmod | grep nouveau > root@ENU-2:~# > > > > This will output a single line if it finds the nouveau kernel module, or > do > > nothing if nouveau is not loaded. Do the same for the nvidia driver: > > $ lsmod |grep nvidia > > root@ENU-2:~# lsmod | grep nvidia > nvidia_drm 36083 2 > nvidia_modeset 1100086 3 nvidia_drm > nvidia 14351241 84 nvidia_modeset > ipmi_msghandler37262 1 nvidia > drm_kms_helper119698 2 i915,nvidia_drm > drm 323926 6 i915,drm_kms_helper,nvidia_drm > i2c_core 48117 7 > drm,i915,i2c_i801,i2c_dev,drm_kms_helper,i2c_algo_bit,nvidia > root@ENU-2:~# > > > If both drivers are loaded at the same time, you will have problems > > If neither driver is loaded, you will have 2D performance only. > > If nouveau is loaded, but NOT nvidia then nvidia-settings will crash and > > burn. > > > > > > Find which of the 2 modules you have currently loaded in your kernel with > > lsmod, > > Then check the nvidia packages you have installed: > > $ ls /var/log/packages/ |grep nvidia > > root@ENU-2:~# ls /var/log/packages | grep nvidia > nvidia-legacy390-driver-390.87-x86_64-1_SBo > nvidia-legacy390-kernel-390.87_4.4.157-x86_64-1_SBo > root@ENU-2:~# > > > None of the above suggestions are fixes, just triage. What is broken > > depends entirely on the output of lsmod vs the nvidia programs you have > > installed. > > nvidia-settings runs. > > /etc/X11/xorg.conf 73 lines, starting with: > > # nvidia-settings: X configuration file generated by nvidia-settings > # nvidia-settings: version 390.87 (root@ENU-2) Sun Jan 6 15:29:25 PST > 2019 > > It also has a set of lines: > > Section "Device" > Identifier "Device0" > Driver "nvidia" > VendorName "NVIDIA Corporation" > BoardName "GeForce GT 610" > EndSection > > (When I saved that configuration there were two screens. In this file I > only see screen 0. Should I try running nvidia-setting again as root, > which I did the first time) and see if the changes stick? > > -- > Regards, > > Dick Steffens > > ___ > PLUG mailing list > PLUG@pdxlinux.org > http://lists.pdxlinux.org/mailman/listinfo/plug > ___ PLUG mailing list PLUG@pdxlinux.org http://lists.pdxlinux.org/mailman/listinfo/plug
Re: [PLUG] Slackware login pain
On 1/7/19 2:59 PM, Ben Koenig wrote: the blacklist file prevents Nouveau from loading, but if you need to double check that it actually worked you use 'lsmod' I pipe the output to grep to shorten the list: $ lsmod |grep nouveau root@ENU-2:~# lsmod | grep nouveau root@ENU-2:~# This will output a single line if it finds the nouveau kernel module, or do nothing if nouveau is not loaded. Do the same for the nvidia driver: $ lsmod |grep nvidia root@ENU-2:~# lsmod | grep nvidia nvidia_drm 36083 2 nvidia_modeset 1100086 3 nvidia_drm nvidia 14351241 84 nvidia_modeset ipmi_msghandler 37262 1 nvidia drm_kms_helper 119698 2 i915,nvidia_drm drm 323926 6 i915,drm_kms_helper,nvidia_drm i2c_core 48117 7 drm,i915,i2c_i801,i2c_dev,drm_kms_helper,i2c_algo_bit,nvidia root@ENU-2:~# If both drivers are loaded at the same time, you will have problems If neither driver is loaded, you will have 2D performance only. If nouveau is loaded, but NOT nvidia then nvidia-settings will crash and burn. Find which of the 2 modules you have currently loaded in your kernel with lsmod, Then check the nvidia packages you have installed: $ ls /var/log/packages/ |grep nvidia root@ENU-2:~# ls /var/log/packages | grep nvidia nvidia-legacy390-driver-390.87-x86_64-1_SBo nvidia-legacy390-kernel-390.87_4.4.157-x86_64-1_SBo root@ENU-2:~# None of the above suggestions are fixes, just triage. What is broken depends entirely on the output of lsmod vs the nvidia programs you have installed. nvidia-settings runs. /etc/X11/xorg.conf 73 lines, starting with: # nvidia-settings: X configuration file generated by nvidia-settings # nvidia-settings: version 390.87 (root@ENU-2) Sun Jan 6 15:29:25 PST 2019 It also has a set of lines: Section "Device" Identifier "Device0" Driver "nvidia" VendorName "NVIDIA Corporation" BoardName "GeForce GT 610" EndSection (When I saved that configuration there were two screens. In this file I only see screen 0. Should I try running nvidia-setting again as root, which I did the first time) and see if the changes stick? -- Regards, Dick Steffens ___ PLUG mailing list PLUG@pdxlinux.org http://lists.pdxlinux.org/mailman/listinfo/plug
Re: [PLUG] Slackware login pain
the blacklist file prevents Nouveau from loading, but if you need to double check that it actually worked you use 'lsmod' I pipe the output to grep to shorten the list: $ lsmod |grep nouveau This will output a single line if it finds the nouveau kernel module, or do nothing if nouveau is not loaded. Do the same for the nvidia driver: $ lsmod |grep nvidia If both drivers are loaded at the same time, you will have problems If neither driver is loaded, you will have 2D performance only. If nouveau is loaded, but NOT nvidia then nvidia-settings will crash and burn. Find which of the 2 modules you have currently loaded in your kernel with lsmod, Then check the nvidia packages you have installed: $ ls /var/log/packages/ |grep nvidia None of the above suggestions are fixes, just triage. What is broken depends entirely on the output of lsmod vs the nvidia programs you have installed. On Mon, Jan 7, 2019 at 10:35 AM Dick Steffens wrote: > On 1/5/19 8:30 PM, King Beowulf wrote: > > > <...> (stuff I did yesterday) > > > now comment out the nouveau blacklist in > > /etc/modprobe.d/BLACKLIST-nouveau.conf: > > > > # Do not load the kernel nouveau dri module, since it interferes with > both > > # the nv and binary nvidia drivers. > > > > #blacklist nouveau > > > > (once you re-install nvidia, you will need to blacklist the nouveau > > module again) > > > > reboot. Once the console comes back up, log in as regular user and > > > > startx > > > > note any errors on screen > > Nothing I noticed. > > > and in /var/log/Xorg.0.log > > [ 153.273] (EE) Failed to load module "fbdev" (module does not exist, 0) > > I see from > > https://www.linuxquestions.org/questions/slackware-14/what-happened-to-xf86-video-fbdev-917338/ > that it's a don't care. > > > If the DE starts up, do the mouse, keyboard work work? Any startup > errors? > > Mouse and keyboard work. > > nvidia-settings also works, and I was able to change screens so that the > smaller monitor is the primary display and is on the left, and the > larger monitor is on the right, and the cursor goes from the left > monitor to the right monitor in the expected way. > > > depending on your use case, the nouveau driver may be plenty good > > enough. If not, reinstall the nvidia-390 drivers from Slackbuilds.org if > > nouveau works. > > > > I recommend sticking with runlevel 3 until everything is configured to > > your liking. X is a weird beast. best not to jab it with a stick. > > It looks like it's working, so I'll go back to run level 4 and make sure. > > Working mostly, but the /etc/X11/xorg.conf I saved from nvidia-settings > didn't hold. It looks like it's not using the nvidia driver. When > attempting to run nvidia-settings I get the error message: > > ERROR: The control display is undefined; please run `nvidia-settings > --help` for usage information. > > I note that you said above that I'd have to blacklist nouveau again. I > have the file > > /etc/modprobe.d/BLACKLIST-nouveau.conf > > What else do I need to do to blacklist nouveau again? > > -- > Regards, > > Dick Steffens > > > ___ > PLUG mailing list > PLUG@pdxlinux.org > http://lists.pdxlinux.org/mailman/listinfo/plug > ___ PLUG mailing list PLUG@pdxlinux.org http://lists.pdxlinux.org/mailman/listinfo/plug
Re: [PLUG] Slackware login pain
On 1/5/19 8:30 PM, King Beowulf wrote: <...> (stuff I did yesterday) now comment out the nouveau blacklist in /etc/modprobe.d/BLACKLIST-nouveau.conf: # Do not load the kernel nouveau dri module, since it interferes with both # the nv and binary nvidia drivers. #blacklist nouveau (once you re-install nvidia, you will need to blacklist the nouveau module again) reboot. Once the console comes back up, log in as regular user and startx note any errors on screen Nothing I noticed. and in /var/log/Xorg.0.log [ 153.273] (EE) Failed to load module "fbdev" (module does not exist, 0) I see from https://www.linuxquestions.org/questions/slackware-14/what-happened-to-xf86-video-fbdev-917338/ that it's a don't care. If the DE starts up, do the mouse, keyboard work work? Any startup errors? Mouse and keyboard work. nvidia-settings also works, and I was able to change screens so that the smaller monitor is the primary display and is on the left, and the larger monitor is on the right, and the cursor goes from the left monitor to the right monitor in the expected way. depending on your use case, the nouveau driver may be plenty good enough. If not, reinstall the nvidia-390 drivers from Slackbuilds.org if nouveau works. I recommend sticking with runlevel 3 until everything is configured to your liking. X is a weird beast. best not to jab it with a stick. It looks like it's working, so I'll go back to run level 4 and make sure. Working mostly, but the /etc/X11/xorg.conf I saved from nvidia-settings didn't hold. It looks like it's not using the nvidia driver. When attempting to run nvidia-settings I get the error message: ERROR: The control display is undefined; please run `nvidia-settings --help` for usage information. I note that you said above that I'd have to blacklist nouveau again. I have the file /etc/modprobe.d/BLACKLIST-nouveau.conf What else do I need to do to blacklist nouveau again? -- Regards, Dick Steffens ___ PLUG mailing list PLUG@pdxlinux.org http://lists.pdxlinux.org/mailman/listinfo/plug
Re: [PLUG] Slackware login pain
On 1/6/19 12:51 AM, Ben Koenig wrote: No, that looks fine. Don't delete them. Most of those files should have been replaced when you reinstalled the packages as suggested by Ed. All you need to do is reboot so that nouveau loads properly and everything should be set. Reboot, see what it gives you. Ideally you will have /etc/inittab set to runlevel 3 and it will give a non-graphical login. Take it slow, slackware is going to launch everything one layer at a time so sit back and enjoy the show. Yes. ENU-2 came up in run level 3, and startx worked. I created ~/.ssh/config on both machines, and put in ForwardX11Trusted yes ssh -x from ENU-1 to ENU-2 (Ubuntu to Slackware) works, and I can run Thunar while sitting at ENU-1. Thanks to all for the support. -- Regards, Dick Steffens ___ PLUG mailing list PLUG@pdxlinux.org http://lists.pdxlinux.org/mailman/listinfo/plug
Re: [PLUG] Slackware login pain
On 1/5/19 9:46 PM, Daniel Johnson wrote: On Slackware does run level 3 start sshd? Just making sure so he doesn't kill a secondary login. Also are you sure none of the libGL and glx stuff is part of mesa or X? I'm used to package managers doing the right thing, and if I have to touch a system .so that usually means a third party installer was behaving badly. Just after sending my reply to Ed last night we had a power failure, so any secondary logins are killed. As you've seen in my response to Ben, I do have that level of things working. Thanks for your thoughts. -- Regards, Dick Steffens ___ PLUG mailing list PLUG@pdxlinux.org http://lists.pdxlinux.org/mailman/listinfo/plug
Re: [PLUG] Slackware login pain
No, that looks fine. Don't delete them. Most of those files should have been replaced when you reinstalled the packages as suggested by Ed. All you need to do is reboot so that nouveau loads properly and everything should be set. Reboot, see what it gives you. Ideally you will have /etc/inittab set to runlevel 3 and it will give a non-graphical login. Take it slow, slackware is going to launch everything one layer at a time so sit back and enjoy the show. On Sat, Jan 5, 2019 at 10:16 PM Dick Steffens wrote: > On 1/5/19 8:30 PM, King Beowulf wrote: > > Hello Dick, > > The other replies have some good ideas, but lets go back to basics since > > you can ssh in. I do not think there is a need to reinstall from > > scratch just yet. > > > > When you leave the login manager, X tries to load all the other stuff > > and can't, so exits and the login manager reloads. Thus . If you select > > the option to instead exit, a list gives you the option to exit to the > > console (no X) instead of rebooting/shut down, as though you had set > > inittab to runlevel 3 (console) and not runlevel 4 (GUI) > > > > When X is running, you can access virtual terminals via: > > > > ctrl-alt-F1 log messages for running X > > ctrl-alt-F[2-6] text console login > > ctrl-alt-F7 return to X screen > > > > Since you had too new of a nvidia driver, lets return to stock and ty > again. > > > > Remove Nvidia and check operation > > = > > 1. log in via ssh. edit /etc/inittab to remove GUI login: > > > > # Default runlevel. (Do not set to 0 or 6) > > id:3:initdefault: > > > > 2. reboot. After, you should have a login prompt. Does your keyboard > > work? Can you log in? If No, post back any errors. > > Yes. I'm logged in as rsteff. > > > > *IF YES* > > > > log in as root. > > I switched to root with su- > > > remove nvidia drivers and reinstall xserver and mesa: > > > > nvidia-switch --remove > > nvidia-switch command not found > > > slackpkg remove nvidia-driver > > The file(s) nvidia-driver can't be removed -- package not installed. > > > slackpkg remove nvidia-kernel > > The file(s) nvidia-kernel can't be removed -- package not installed. > > I expected those responses since I removed both of those packages on > Friday. > > > slackpkg reinstall mesa > > Successful. > > > slackpkg reinstall xorg-server > > > > (make sure you pick the newest versions of both) > > ncurses window has four entries, all stared [*] > > [*] xorg-server-1.18.3-x86_64-5_slack14.2.txz > [*] xorg-server-xephyr-1.18.3-x86_64-5_slack14.2.txz > [*] xorg-server-xnest-1.18.3-x86_64-5_slack14.2.txz > [*] xorg-server-xvfb-1.18.3-x86_64-5_slack14.2.txz > > I'm going out on a limb and assuming I need all of those. > > Successful. > > > now, check for busted nvidia symlinks and delete them > > > > ls -l /usr/lib64/libGL* > root@ENU-2:~# ls -l /usr/lib64/libGL* > -rwxr-xr-x 1 root root 1462 May 9 2016 /usr/lib64/libGL.la* > lrwxrwxrwx 1 root root 14 Jan 5 22:03 /usr/lib64/libGL.so -> > libGL.so.1.2.0* > lrwxrwxrwx 1 root root 14 Jan 5 22:03 /usr/lib64/libGL.so.1 -> > libGL.so.1.2.0* > -rwxr-xr-x 1 root root 422392 May 9 2016 /usr/lib64/libGL.so.1.2.0* > -rwxr-xr-x 1 root root 1005 May 9 2016 /usr/lib64/libGLESv1_CM.la* > lrwxrwxrwx 1 root root 21 Jan 5 22:03 /usr/lib64/libGLESv1_CM.so -> > libGLESv1_CM.so.1.1.0* > lrwxrwxrwx 1 root root 21 Jan 5 22:03 /usr/lib64/libGLESv1_CM.so.1 > -> libGLESv1_CM.so.1.1.0* > -rwxr-xr-x 1 root root 18072 May 9 2016 > /usr/lib64/libGLESv1_CM.so.1.1.0* > -rwxr-xr-x 1 root root987 May 9 2016 /usr/lib64/libGLESv2.la* > lrwxrwxrwx 1 root root 18 Jan 5 22:03 /usr/lib64/libGLESv2.so -> > libGLESv2.so.2.0.0* > lrwxrwxrwx 1 root root 18 Jan 5 22:03 /usr/lib64/libGLESv2.so.2 -> > libGLESv2.so.2.0.0* > -rwxr-xr-x 1 root root 30360 May 9 2016 /usr/lib64/libGLESv2.so.2.0.0* > lrwxrwxrwx 1 root root 17 Dec 23 13:29 /usr/lib64/libGLEW.so -> > libGLEW.so.1.13.0* > lrwxrwxrwx 1 root root 17 Dec 23 13:29 /usr/lib64/libGLEW.so.1.13 -> > libGLEW.so.1.13.0* > -rwxr-xr-x 1 root root 509520 Nov 23 2015 /usr/lib64/libGLEW.so.1.13.0* > lrwxrwxrwx 1 root root 19 Dec 23 13:29 /usr/lib64/libGLEWmx.so -> > libGLEWmx.so.1.13.0* > lrwxrwxrwx 1 root root 19 Dec 23 13:29 /usr/lib64/libGLEWmx.so.1.13 > -> libGLEWmx.so.1.13.0* > -rwxr-xr-x 1 root root 448488 Nov 23 2015 /usr/lib64/libGLEWmx.so.1.13.0* > -rwxr-xr-x 1 root root 1254 Feb 20 2013 /usr/lib64/libGLU.la* > lrwxrwxrwx 1 root root 15 Dec 23 13:29 /usr/lib64/libGLU.so -> > libGLU.so.1.3.1* > lrwxrwxrwx 1 root root 15 Dec 23 13:29 /usr/lib64/libGLU.so.1 -> > libGLU.so.1.3.1* > -rwxr-xr-x 1 root root 516416 Feb 20 2013 /usr/lib64/libGLU.so.1.3.1* > root@ENU-2:~# > > > ls -l /usr/lib64/libGLES* > root@ENU-2:~# ls -l /usr/lib64/xorg/modules/extensions/libglx* > -rwxr-xr-x 1 root root 1503 Oct 18 2017 > /usr/lib64/xorg/modules/extensions/libglx.la* > -rwxr-xr-x 1 root root 285992 Oct 18 2017 > /
Re: [PLUG] Slackware login pain
On 1/5/19 8:30 PM, King Beowulf wrote: Hello Dick, The other replies have some good ideas, but lets go back to basics since you can ssh in. I do not think there is a need to reinstall from scratch just yet. When you leave the login manager, X tries to load all the other stuff and can't, so exits and the login manager reloads. Thus . If you select the option to instead exit, a list gives you the option to exit to the console (no X) instead of rebooting/shut down, as though you had set inittab to runlevel 3 (console) and not runlevel 4 (GUI) When X is running, you can access virtual terminals via: ctrl-alt-F1 log messages for running X ctrl-alt-F[2-6] text console login ctrl-alt-F7 return to X screen Since you had too new of a nvidia driver, lets return to stock and ty again. Remove Nvidia and check operation = 1. log in via ssh. edit /etc/inittab to remove GUI login: # Default runlevel. (Do not set to 0 or 6) id:3:initdefault: 2. reboot. After, you should have a login prompt. Does your keyboard work? Can you log in? If No, post back any errors. Yes. I'm logged in as rsteff. *IF YES* log in as root. I switched to root with su- remove nvidia drivers and reinstall xserver and mesa: nvidia-switch --remove nvidia-switch command not found slackpkg remove nvidia-driver The file(s) nvidia-driver can't be removed -- package not installed. slackpkg remove nvidia-kernel The file(s) nvidia-kernel can't be removed -- package not installed. I expected those responses since I removed both of those packages on Friday. slackpkg reinstall mesa Successful. slackpkg reinstall xorg-server (make sure you pick the newest versions of both) ncurses window has four entries, all stared [*] [*] xorg-server-1.18.3-x86_64-5_slack14.2.txz [*] xorg-server-xephyr-1.18.3-x86_64-5_slack14.2.txz [*] xorg-server-xnest-1.18.3-x86_64-5_slack14.2.txz [*] xorg-server-xvfb-1.18.3-x86_64-5_slack14.2.txz I'm going out on a limb and assuming I need all of those. Successful. now, check for busted nvidia symlinks and delete them ls -l /usr/lib64/libGL* root@ENU-2:~# ls -l /usr/lib64/libGL* -rwxr-xr-x 1 root root 1462 May 9 2016 /usr/lib64/libGL.la* lrwxrwxrwx 1 root root 14 Jan 5 22:03 /usr/lib64/libGL.so -> libGL.so.1.2.0* lrwxrwxrwx 1 root root 14 Jan 5 22:03 /usr/lib64/libGL.so.1 -> libGL.so.1.2.0* -rwxr-xr-x 1 root root 422392 May 9 2016 /usr/lib64/libGL.so.1.2.0* -rwxr-xr-x 1 root root 1005 May 9 2016 /usr/lib64/libGLESv1_CM.la* lrwxrwxrwx 1 root root 21 Jan 5 22:03 /usr/lib64/libGLESv1_CM.so -> libGLESv1_CM.so.1.1.0* lrwxrwxrwx 1 root root 21 Jan 5 22:03 /usr/lib64/libGLESv1_CM.so.1 -> libGLESv1_CM.so.1.1.0* -rwxr-xr-x 1 root root 18072 May 9 2016 /usr/lib64/libGLESv1_CM.so.1.1.0* -rwxr-xr-x 1 root root 987 May 9 2016 /usr/lib64/libGLESv2.la* lrwxrwxrwx 1 root root 18 Jan 5 22:03 /usr/lib64/libGLESv2.so -> libGLESv2.so.2.0.0* lrwxrwxrwx 1 root root 18 Jan 5 22:03 /usr/lib64/libGLESv2.so.2 -> libGLESv2.so.2.0.0* -rwxr-xr-x 1 root root 30360 May 9 2016 /usr/lib64/libGLESv2.so.2.0.0* lrwxrwxrwx 1 root root 17 Dec 23 13:29 /usr/lib64/libGLEW.so -> libGLEW.so.1.13.0* lrwxrwxrwx 1 root root 17 Dec 23 13:29 /usr/lib64/libGLEW.so.1.13 -> libGLEW.so.1.13.0* -rwxr-xr-x 1 root root 509520 Nov 23 2015 /usr/lib64/libGLEW.so.1.13.0* lrwxrwxrwx 1 root root 19 Dec 23 13:29 /usr/lib64/libGLEWmx.so -> libGLEWmx.so.1.13.0* lrwxrwxrwx 1 root root 19 Dec 23 13:29 /usr/lib64/libGLEWmx.so.1.13 -> libGLEWmx.so.1.13.0* -rwxr-xr-x 1 root root 448488 Nov 23 2015 /usr/lib64/libGLEWmx.so.1.13.0* -rwxr-xr-x 1 root root 1254 Feb 20 2013 /usr/lib64/libGLU.la* lrwxrwxrwx 1 root root 15 Dec 23 13:29 /usr/lib64/libGLU.so -> libGLU.so.1.3.1* lrwxrwxrwx 1 root root 15 Dec 23 13:29 /usr/lib64/libGLU.so.1 -> libGLU.so.1.3.1* -rwxr-xr-x 1 root root 516416 Feb 20 2013 /usr/lib64/libGLU.so.1.3.1* root@ENU-2:~# ls -l /usr/lib64/libGLES* root@ENU-2:~# ls -l /usr/lib64/xorg/modules/extensions/libglx* -rwxr-xr-x 1 root root 1503 Oct 18 2017 /usr/lib64/xorg/modules/extensions/libglx.la* -rwxr-xr-x 1 root root 285992 Oct 18 2017 /usr/lib64/xorg/modules/extensions/libglx.so* root@ENU-2:~# ls -l /usr/lib64/xorg/modules/extensions/libglx* root@ENU-2:~# ls -l /usr/lib64/xorg/modules/extensions/libglx* -rwxr-xr-x 1 root root 1503 Oct 18 2017 /usr/lib64/xorg/modules/extensions/libglx.la* -rwxr-xr-x 1 root root 285992 Oct 18 2017 /usr/lib64/xorg/modules/extensions/libglx.so* root@ENU-2:~# Are all of those related to nVidia, and should I delete them all? If yes, is this the right command for each group: rm /usr/lib64/libGL* rm /usr/lib64/libGLES* rm /usr/lib64/xorg/modules/extensions/libglx* I'll pick up the rest of this tomorrow after I hear from you about my delete questions. Thanks! -- Regards, Dick Steffens ___ PLUG mailing list PLUG@pdxl
Re: [PLUG] Slackware login pain
On Slackware does run level 3 start sshd? Just making sure so he doesn't kill a secondary login. Also are you sure none of the libGL and glx stuff is part of mesa or X? I'm used to package managers doing the right thing, and if I have to touch a system .so that usually means a third party installer was behaving badly. On Sat, Jan 5, 2019, 8:30 PM King Beowulf On 1/5/19 2:34 PM, Dick Steffens wrote: > > Sometime yesterday (see thread Can't kill process) I lost the ability to > > log in to my Slackware machine, ENU-2, from its keyboard. I can log in > > via ssh from another machine. I have successfully mounted a USB stick > > and copied all the directories in my Downloads directory and copied them > > to this machine (the one I'm typing on). > > > > Just a few minutes ago I thought I'd try logging in from ENU-2's > > keyboard. I had a look a the options and saw one of them is called > > Failsafe. So, I gave it a try. While it works, it provides only a small > > terminal window, which works for awhile, but then stops responding to > > anything other than d. I do that enough times and it logs me out. > > I can log back in to Failsafe, but not KDE, KDE (Failsafe), or XFCE. > > When I enter my password, the screen briefly shows the beginnings of the > > initial phases of logging in, but then goes blank, and returns to the > > login screen. > > > > At this point I'm seriously considering starting over. I have all the > > stuff I downloaded for various add on programs, so reinstalling them > > will take a little less time than before. > > > > I there any good reason I shouldn't go this route? > > > > Hello Dick, > > The other replies have some good ideas, but lets go back to basics since > you can ssh in. I do not think there is a need to reinstall from > scratch just yet. > > When you leave the login manager, X tries to load all the other stuff > and can't, so exits and the login manager reloads. Thus . If you select > the option to instead exit, a list gives you the option to exit to the > console (no X) instead of rebooting/shut down, as though you had set > inittab to runlevel 3 (console) and not runlevel 4 (GUI) > > When X is running, you can access virtual terminals via: > > ctrl-alt-F1 log messages for running X > ctrl-alt-F[2-6] text console login > ctrl-alt-F7 return to X screen > > Since you had too new of a nvidia driver, lets return to stock and ty > again. > > Remove Nvidia and check operation > = > 1. log in via ssh. edit /etc/inittab to remove GUI login: > > # Default runlevel. (Do not set to 0 or 6) > id:3:initdefault: > > 2. reboot. After, you should have a login prompt. Does your keyboard > work? Can you log in? If No, post back any errors. > > *IF YES* > > log in as root. remove nvidia drivers and reinstall xserver and mesa: > > nvidia-switch --remove > slackpkg remove nvidia-driver > slackpkg remove nvidia-kernel > > slackpkg reinstall mesa > slackpkg reinstall xorg-server > > (make sure you pick the newest versions of both) > > now, check for busted nvidia symlinks and delete them > > ls -l /usr/lib64/libGL* > ls -l /usr/lib64/libGLES* > ls -l /usr/lib64/xorg/modules/extensions/libglx* > > now comment out the nouveau blacklist in > /etc/modprobe.d/BLACKLIST-nouveau.conf: > > # Do not load the kernel nouveau dri module, since it interferes with both > # the nv and binary nvidia drivers. > > #blacklist nouveau > > (once you re-install nvidia, you will need to blacklist the nouveau > module again) > > reboot. Once the console comes back up, log in as regular user and > > startx > > note any errors on screen and in /var/log/Xorg.0.log > > If the DE starts up, do the mouse, keyboard work work? Any startup errors? > > depending on your use case, the nouveau driver may be plenty good > enough. If not, reinstall the nvidia-390 drivers from Slackbuilds.org if > nouveau works. > > I recommend sticking with runlevel 3 until everything is configured to > your liking. X is a weird beast. best not to jab it with a stick. > > -Ed > > ___ > PLUG mailing list > PLUG@pdxlinux.org > http://lists.pdxlinux.org/mailman/listinfo/plug > ___ PLUG mailing list PLUG@pdxlinux.org http://lists.pdxlinux.org/mailman/listinfo/plug
Re: [PLUG] Slackware login pain
On 1/5/19 7:52 PM, wes wrote: > On Sat, Jan 5, 2019 at 6:18 PM Dick Steffens wrote: > >> Would the >> solution there of deleting all .serverauthXXX and .Xauthority files as >> root be something I should do here? >> > > Yes, you should try this. Also look for a .ICEauthority file to delete or > rename. > > -wes No. Not at this time. Dick's issue is a borked nvidia install. Let's not get go fishing for red herrings. -Ed signature.asc Description: OpenPGP digital signature ___ PLUG mailing list PLUG@pdxlinux.org http://lists.pdxlinux.org/mailman/listinfo/plug
Re: [PLUG] Slackware login pain
On 1/5/19 2:34 PM, Dick Steffens wrote: > Sometime yesterday (see thread Can't kill process) I lost the ability to > log in to my Slackware machine, ENU-2, from its keyboard. I can log in > via ssh from another machine. I have successfully mounted a USB stick > and copied all the directories in my Downloads directory and copied them > to this machine (the one I'm typing on). > > Just a few minutes ago I thought I'd try logging in from ENU-2's > keyboard. I had a look a the options and saw one of them is called > Failsafe. So, I gave it a try. While it works, it provides only a small > terminal window, which works for awhile, but then stops responding to > anything other than d. I do that enough times and it logs me out. > I can log back in to Failsafe, but not KDE, KDE (Failsafe), or XFCE. > When I enter my password, the screen briefly shows the beginnings of the > initial phases of logging in, but then goes blank, and returns to the > login screen. > > At this point I'm seriously considering starting over. I have all the > stuff I downloaded for various add on programs, so reinstalling them > will take a little less time than before. > > I there any good reason I shouldn't go this route? > Hello Dick, The other replies have some good ideas, but lets go back to basics since you can ssh in. I do not think there is a need to reinstall from scratch just yet. When you leave the login manager, X tries to load all the other stuff and can't, so exits and the login manager reloads. Thus . If you select the option to instead exit, a list gives you the option to exit to the console (no X) instead of rebooting/shut down, as though you had set inittab to runlevel 3 (console) and not runlevel 4 (GUI) When X is running, you can access virtual terminals via: ctrl-alt-F1 log messages for running X ctrl-alt-F[2-6] text console login ctrl-alt-F7 return to X screen Since you had too new of a nvidia driver, lets return to stock and ty again. Remove Nvidia and check operation = 1. log in via ssh. edit /etc/inittab to remove GUI login: # Default runlevel. (Do not set to 0 or 6) id:3:initdefault: 2. reboot. After, you should have a login prompt. Does your keyboard work? Can you log in? If No, post back any errors. *IF YES* log in as root. remove nvidia drivers and reinstall xserver and mesa: nvidia-switch --remove slackpkg remove nvidia-driver slackpkg remove nvidia-kernel slackpkg reinstall mesa slackpkg reinstall xorg-server (make sure you pick the newest versions of both) now, check for busted nvidia symlinks and delete them ls -l /usr/lib64/libGL* ls -l /usr/lib64/libGLES* ls -l /usr/lib64/xorg/modules/extensions/libglx* now comment out the nouveau blacklist in /etc/modprobe.d/BLACKLIST-nouveau.conf: # Do not load the kernel nouveau dri module, since it interferes with both # the nv and binary nvidia drivers. #blacklist nouveau (once you re-install nvidia, you will need to blacklist the nouveau module again) reboot. Once the console comes back up, log in as regular user and startx note any errors on screen and in /var/log/Xorg.0.log If the DE starts up, do the mouse, keyboard work work? Any startup errors? depending on your use case, the nouveau driver may be plenty good enough. If not, reinstall the nvidia-390 drivers from Slackbuilds.org if nouveau works. I recommend sticking with runlevel 3 until everything is configured to your liking. X is a weird beast. best not to jab it with a stick. -Ed signature.asc Description: OpenPGP digital signature ___ PLUG mailing list PLUG@pdxlinux.org http://lists.pdxlinux.org/mailman/listinfo/plug
Re: [PLUG] Slackware login pain
On Sat, Jan 5, 2019 at 6:18 PM Dick Steffens wrote: > Would the > solution there of deleting all .serverauthXXX and .Xauthority files as > root be something I should do here? > Yes, you should try this. Also look for a .ICEauthority file to delete or rename. -wes ___ PLUG mailing list PLUG@pdxlinux.org http://lists.pdxlinux.org/mailman/listinfo/plug
Re: [PLUG] Slackware login pain
I need an email app for my phone that doesn't force top posting. Gah! Anyways. Ok you can login to VT2. Startx doesn't work. Could be the display manager is locking a resource you need. startxfce4: X server already running on display :1 Is a clue. Could maybe only need to tell it to use display :2 They start at 0 so I'm curious what else might be running. Try shutting off that service and see if you can manually start x then. Could be a problem with X such as a driver. But it looks like xfce4 is running in X already. The EGl error might be a clue. That is hardware graphics acceleration. Maybe the login screen doesn't need acceleration, but the full desktop does. Which graphics driver are you using? Do you have options like open source vs nvidia official, or perhaps booting into an old kernel? On Sat, Jan 5, 2019, 6:15 PM Dick Steffens On 1/5/19 5:12 PM, Daniel Johnson wrote: > > If your goal is learning then it's worth it. If your goal was just having > > it work I'm not sure you would have picked Slackware as your distro so > I'm > > going with learning. > > A little of both. I've been using the easy to install and run stuff for > something over 10 years, but lately there have been unresolved issues, > mostly minor annoyances. So, hearing that Slackware still used the old > school approach instead of Systemd, I thought I'd try learning more. It > always made sense to me that everything should be in a file somewhere, > and those should be editable and fixable. > > > Being able to login via ssh means the password isn't corrupted. Being > able > > to type a username probably means the keyboard input is fine. What's > left? > > > > PAM (Pluggable Authentication Modules), or the display manager are the > only > > ones that come to mind. > > > > Can you switch VT (virtual terminal) and login from the console? > > Yes. > > > Something > > like Ctrl-Alt-F2 ... If you can that indicates another part that isn't > > broken. If you get that far can you startX? > No. > > rsteff@ENU-2:~$startx > xauth: /home/rsteff/.serverauth.1274 does not exist > X.Org X Server 1.18.3 > Release Date: 2016-04-04 > X Protocol Version 11, Revision 0 > Build Operating System: Slackware 14.2 Slackware Linux Project > Current Operating System: Linux ENU-2 4.4.157 #2 SMP Fri Sep 21 0:36:59 > CDT 2018 x86_64 > Kernel command line: BOOT_IMAGE=dev000:\EFI\SLACKWARE\vmlinuz > root=/dev/sda2 vga=normal ro ro > Build Date: 18 October 2017 12:21:16PM > > Current version of pixman: 0.34.0 > Before reporting problems, check htp://wiki.x.org > to make sure that you have the latest version. > Markers: (--) probed, (**) from config file, (==) default setting, > (++) from command line, (!!) notice, informational, > (WW) warning, (EE) error, (NI) not implemented, (??) Unknown. > (==) Log file: "/var/log/Xorg.1.log", Time: Sat Jan 5 17:41:53 2019 > (==) Using system config directory "/usr/share/X11/xorg.conf.d" > resize called 3200 1000 > /usr/bin/startxfce4: X server already running on display :1 > xfce4-session: error while loading shared libraries: libEGL.so.1 cannot > open shared object file: No such file or directory > xinit: connection to X serve lost > > waiting for X server to shut down (II) Server terminated successfully > (0). Closing log file > > rsteff:@ENU-2:~$ > > > > Could be a display manager problem. KDE likes to install KDM(KDE Display > > Manager) GNOME has GDM. There are a few others xfce has one it usually > > comes with. They can usually start the other desktops though. Try > replacing > > whatever display manager you have setup with a different one and see if > > that works. > > I have tried KDE Plasma Workspace, KDE Plasma Workspace (failsafe > session), XFCE Session, and Failsafe. The only one of those that worked > at all was Failsafe, and that for only a short while. > > From the > xauth: /home/rsteff/.serverauth.1274 does not exist > line it looks like I'm missing an important file. Is there a way to > properly create that file from the terminal, either the one I got with > Ctrl-Alt-F2 or from an ssh login? > > I just searched on "Slackware .serverauth.1274 does not exist" and found: > > > https://www.linuxquestions.org/questions/slackware-14/xauth-file-home-user-serverauth-1436-does-not-exist-4175576123/ > > It talks about trouble with an nVidia driver. As I think about it, > that's about the time I started having this problem. I tried to use a > newer driver when I should have been using an older one. Would the > solution there of deleting all .serverauthXXX and .Xauthority files as > root be something I should do here? > > Thanks for showing me another way to look for the problem. > > -- > Regards, > > Dick Steffens > > > ___ > PLUG mailing list > PLUG@pdxlinux.org > http://lists.pdxlinux.org/mailman/listinfo/plug > ___ PLUG mailing list PLUG@pdxlinux.org http://lists.pdxlinux.org/mailman/
Re: [PLUG] Slackware login pain
On 1/5/19 5:12 PM, Daniel Johnson wrote: If your goal is learning then it's worth it. If your goal was just having it work I'm not sure you would have picked Slackware as your distro so I'm going with learning. A little of both. I've been using the easy to install and run stuff for something over 10 years, but lately there have been unresolved issues, mostly minor annoyances. So, hearing that Slackware still used the old school approach instead of Systemd, I thought I'd try learning more. It always made sense to me that everything should be in a file somewhere, and those should be editable and fixable. Being able to login via ssh means the password isn't corrupted. Being able to type a username probably means the keyboard input is fine. What's left? PAM (Pluggable Authentication Modules), or the display manager are the only ones that come to mind. Can you switch VT (virtual terminal) and login from the console? Yes. Something like Ctrl-Alt-F2 ... If you can that indicates another part that isn't broken. If you get that far can you startX? No. rsteff@ENU-2:~$startx xauth: /home/rsteff/.serverauth.1274 does not exist X.Org X Server 1.18.3 Release Date: 2016-04-04 X Protocol Version 11, Revision 0 Build Operating System: Slackware 14.2 Slackware Linux Project Current Operating System: Linux ENU-2 4.4.157 #2 SMP Fri Sep 21 0:36:59 CDT 2018 x86_64 Kernel command line: BOOT_IMAGE=dev000:\EFI\SLACKWARE\vmlinuz root=/dev/sda2 vga=normal ro ro Build Date: 18 October 2017 12:21:16PM Current version of pixman: 0.34.0 Before reporting problems, check htp://wiki.x.org to make sure that you have the latest version. Markers: (--) probed, (**) from config file, (==) default setting, (++) from command line, (!!) notice, informational, (WW) warning, (EE) error, (NI) not implemented, (??) Unknown. (==) Log file: "/var/log/Xorg.1.log", Time: Sat Jan 5 17:41:53 2019 (==) Using system config directory "/usr/share/X11/xorg.conf.d" resize called 3200 1000 /usr/bin/startxfce4: X server already running on display :1 xfce4-session: error while loading shared libraries: libEGL.so.1 cannot open shared object file: No such file or directory xinit: connection to X serve lost waiting for X server to shut down (II) Server terminated successfully (0). Closing log file rsteff:@ENU-2:~$ Could be a display manager problem. KDE likes to install KDM(KDE Display Manager) GNOME has GDM. There are a few others xfce has one it usually comes with. They can usually start the other desktops though. Try replacing whatever display manager you have setup with a different one and see if that works. I have tried KDE Plasma Workspace, KDE Plasma Workspace (failsafe session), XFCE Session, and Failsafe. The only one of those that worked at all was Failsafe, and that for only a short while. From the xauth: /home/rsteff/.serverauth.1274 does not exist line it looks like I'm missing an important file. Is there a way to properly create that file from the terminal, either the one I got with Ctrl-Alt-F2 or from an ssh login? I just searched on "Slackware .serverauth.1274 does not exist" and found: https://www.linuxquestions.org/questions/slackware-14/xauth-file-home-user-serverauth-1436-does-not-exist-4175576123/ It talks about trouble with an nVidia driver. As I think about it, that's about the time I started having this problem. I tried to use a newer driver when I should have been using an older one. Would the solution there of deleting all .serverauthXXX and .Xauthority files as root be something I should do here? Thanks for showing me another way to look for the problem. -- Regards, Dick Steffens ___ PLUG mailing list PLUG@pdxlinux.org http://lists.pdxlinux.org/mailman/listinfo/plug
Re: [PLUG] Slackware login pain
If your goal is learning then it's worth it. If your goal was just having it work I'm not sure you would have picked Slackware as your distro so I'm going with learning. Being able to login via ssh means the password isn't corrupted. Being able to type a username probably means the keyboard input is fine. What's left? PAM (Pluggable Authentication Modules), or the display manager are the only ones that come to mind. Can you switch VT (virtual terminal) and login from the console? Something like Ctrl-Alt-F2 ... If you can that indicates another part that isn't broken. If you get that far can you startX? Could be a display manager problem. KDE likes to install KDM(KDE Display Manager) GNOME has GDM. There are a few others xfce has one it usually comes with. They can usually start the other desktops though. Try replacing whatever display manager you have setup with a different one and see if that works. On Sat, Jan 5, 2019, 2:34 PM Dick Steffens Sometime yesterday (see thread Can't kill process) I lost the ability to > log in to my Slackware machine, ENU-2, from its keyboard. I can log in > via ssh from another machine. I have successfully mounted a USB stick > and copied all the directories in my Downloads directory and copied them > to this machine (the one I'm typing on). > > Just a few minutes ago I thought I'd try logging in from ENU-2's > keyboard. I had a look a the options and saw one of them is called > Failsafe. So, I gave it a try. While it works, it provides only a small > terminal window, which works for awhile, but then stops responding to > anything other than d. I do that enough times and it logs me out. > I can log back in to Failsafe, but not KDE, KDE (Failsafe), or XFCE. > When I enter my password, the screen briefly shows the beginnings of the > initial phases of logging in, but then goes blank, and returns to the > login screen. > > At this point I'm seriously considering starting over. I have all the > stuff I downloaded for various add on programs, so reinstalling them > will take a little less time than before. > > I there any good reason I shouldn't go this route? > > -- > Regards, > > Dick Steffens > > ___ > PLUG mailing list > PLUG@pdxlinux.org > http://lists.pdxlinux.org/mailman/listinfo/plug > ___ PLUG mailing list PLUG@pdxlinux.org http://lists.pdxlinux.org/mailman/listinfo/plug