I’ve been upgrading to LEAP 15.3 from an earlier version of openSUSE. I successfully installed Leap 15.2 and then attempted to upgrade to LEAP 15.3 using
I had a glitch in the middle of it when it told me it failed to remove the Linux Kernel but then moved on. When I rebooted the machine it failed to load Plasma. I got the message “Welcome to openSUSE LEAP 15.3” and offered me a command line login. Before the Welcome message it displayed “Failed to start Load Kernel Modules”.
Information about your graphic hardware, /var/log/Xorg.log (although depending on display manager it may be somewhere else) and output of “journalctl -b” immediately after boot would certainly help.
Is there a rescue or failsafe option available at the Grub menu? If yes, what happens if you use it? If yes, what does rpm -qa | grep rnel report if you are able to login?
Is there a prior kernel option available at the Grub menu? If yes, what happens if you use it, and what does rpm -qa | grep rnel report if you are able to login?
I booted up in recovery mode and tried **rpm -qa | grep rnel. **There are 13 entries displayed, but I’m not sure what I should be looking for. Should I expect more than 13? Fewer?
My nVidia card is a GeForce GTX 980. Interestingly, it tells me that my kernel module is nouveau.
I’me afraid that, in the 16 or 17 years I’ve been using Linux I’ve never, as far as I recall, connected to the internet from a command line. How is this done?
When you are asked to provide command output you are really expected to provide command output, not your description of it. Others may see something that neither you nor I recognize.
You apparently have another system where you can access this site. So you could login over network and simply copy-paste output. In the worst case you can copy output to USB stick.
Interestingly, it tells me that my kernel module is nouveau.
Why is it interesting? This is default driver for nVidia cards. What did you expect?
I’me afraid that, in the 16 or 17 years I’ve been using Linux I’ve never, as far as I recall, connected to the internet from a command line. How is this done?
Are you using wicked or NetworkManager? Do you have wired or wireless connection?
I’m using my iPad to enter stuff in this forum. Anything I enter here I literally have to read from my PC then type up. There’s no way for me to copy and paste between the devices. That’s why the idea of connecting to the internet view commands would be useful. The Xorg.log output says my PC has Samba running, so that’s maybe a start.
I’m using wicked over a wired connection on my PC. And at least the correct driver is loaded. I thought maybe one explanation was that the driver had been lost in the upgrade. It’s reassuring that it hasn’t.
Then you should already be connected. What “ping 8.8.8.8” and “ping google.com” say? Otherwise you need to explain how you established connection from GUI.
There’s an app called Termius. You can use it to ssh from your ipad into your computer if it’s in the same LAN.
I have nvidia cards in all my boxes, too. Actually, landing in CLI after upgrade was quite “daily” business for me until it became officially recommended to keep third party repos active when upgrading which I did with the nvidia repo.
The “failed to delete …” seems to be quite common. I had it and others, too. Don’t bother. Since you got the “welcome to Leap 15.3” I’d think you should have got a working system and kernel. “failed to load kernel modules” may be related to any hardware driver but might as well be the nvidia driver. So, you may try what I usually did, start YaST in terminal. Login as root as and enter “yast”. Add / activate the nvidia repo. Go to software management and (force) install the nvidia driver. I not sure which one is suitable for your card. I have G04 here and your card is listed as supported. (You may have had G03 before which isn’t supported by nvidia anymore.) If you enter G04 into the search field you should get 5 results of which you can choose 4 when activating “nvidia-computeG04”. In case of conflicts it may be nouveau - you can delete such conflicting packages.
For me, that has always done the trick. But wait, after all that typing:
**Did you do the “zypper up” after the upgrade? **It is cruicial to get the new repos active and I think I had a similar issue. Just make sure the nvidia repo is activated.
You can upgrade your system by using zypper dup. When using this command, take the following into account:
If you are using zypper dup and there is not enough disk space available, the upgrade fails. In this case, use --pkg-cache-dir *DIR*
to set an alternative package cache directory. - All packages with unresolved dependencies will be removed. Packages installed from disabled repositories are kept as long as their dependencies are satisfied. Packages that are no longer available in the repositories are considered orphaned. Such packages are uninstalled if their dependencies cannot be satisfied. If they can be satisfied, such packages stay installed.
zypper dup ensures that all installed packages come from one of the available repositories. It does not consider the version or architecture, but prevents changing the vendor of the installed packages by default, using the --no-allow-vendor-change option. If you have third-party repositories enabled, some repositories may break during the upgrade. In this case, use --allow-vendor-change instead.
The app is actually called Termius. You should find it in the appstore via the search function.