Mageia forum

I tried Mageia 9 alpha1. Live Plasma.
In virtualbox, though I did not find anything very interesting, it works normally.
On my hardware, it could not boot into desktop environment.
EFI boot and legacy boot have the same situation.
I selected the default first boot option. It displayed three line and reboot immediately without any green “OK”.
I selected the second option “use non-free NVIDIA drivers”. It displayed a lot of “OK” and then:

[FAILED] Failed to start drakx-installer-xsetup.service.

Then stuck here.
Weird, I do not have NVIDIA graphic cards, I have only AMD FM2+ APU.

thanks for trying a Mga9-alpha1.
these are the sort of findings that are important as QA cannot have every combination of hardware to test the ISOs on.

do you already have a Mageia system on this machine?

if so please post output of

Code: Select all: inxi -Fxxx

benmc wrote:if so please post output of
Code: Select all
inxi -Fxxx

Oh...my...god. Generally speaking I am a weird person trying to hide myself. Though, due to my love to Mageia, I would like to tell my hardware information.

Code: Select all: System: Host: localhost Kernel: 5.10.16-desktop-1.mga8 x86_64 bits: 64 compiler: gcc v: 10.2.1 Desktop: KDE Plasma 5.20.4 tk: Qt 5.15.2 wm: kwin_x11 dm: SDDM Distro: Mageia 8 mga8 Machine: Type: Desktop Mobo: MSI model: A88XM-E45 (MS-7721) v: 5.0 serial: N/A UEFI: American Megatrends v: 25.6 date: 12/15/2014 CPU: Info: Quad Core model: AMD Eng Sample ZD352058A4451_39/35/16_9904_760 bits: 64 type: MCP arch: Piledriver rev: 1 L1 cache: 192 KiB L2 cache: 2 MiB flags: avx lm nx pae sse sse2 sse3 sse4_1 sse4_2 sse4a ssse3 svm bogomips: 27947 Speed: 1397 MHz min/max: 1400/3500 MHz boost: enabled volts: 1.4 V ext-clock: 100 MHz Core speeds (MHz): 1: 1397 2: 1398 3: 1396 4: 1396 Graphics: Device-1: AMD Trinity [Radeon HD 7560D] vendor: Micro-Star MSI driver: radeon v: kernel bus ID: 00:01.0 chip ID: 1002:9904 Display: x11 server: Mageia X.org 1.20.10 compositor: kwin_x11 driver: ati,radeon unloaded: fbdev,modesetting,vesa resolution: 640x480 s-dpi: 96 OpenGL: renderer: AMD ARUBA (DRM 2.50.0 / 5.10.16-desktop-1.mga8 LLVM 11.0.1) v: 4.3 Mesa 20.3.4 compat-v: 3.1 direct render: Yes Audio: Device-1: AMD Trinity HDMI Audio vendor: Micro-Star MSI driver: snd_hda_intel v: kernel bus ID: 00:01.1 chip ID: 1002:9902 Device-2: AMD FCH Azalia vendor: Micro-Star MSI driver: snd_hda_intel v: kernel bus ID: 00:14.2 chip ID: 1022:780d Sound Server: ALSA v: k5.10.16-desktop-1.mga8 Network: Device-1: Realtek RTL8111/8168/8411 PCI Express Gigabit Ethernet vendor: Micro-Star MSI driver: r8169 v: kernel port: e000 bus ID: 01:00.0 chip ID: 10ec:8168 IF: enp1s0 state: up speed: 100 Mbps duplex: full mac: 44:8a:5b:81:c2:9b Drives: Local Storage: total: 30.02 GiB used: 7.68 GiB (25.6%) ID-1: /dev/sda type: USB model: AI Mass size: 30.02 GiB serial: Storage Partition: ID-1: / size: 2.78 GiB used: 203 MiB (7.1%) fs: overlay source: ERR-102 Swap: Alert: No Swap data was found. Sensors: System Temperatures: cpu: 12.8 C mobo: N/A Fan Speeds (RPM): N/A Info: Processes: 240 Uptime: 6m wakeups: 1 Memory: 5.56 GiB used: 2.19 GiB (39.3%) Init: systemd v: 246 runlevel: 5 target: graphical.target Compilers: gcc: N/A Packages: rpm: 3073 Shell: Bash (su) v: 5.1.4 running in: konsole inxi: 3.2.01

Yes an AMD Sample CPU. I have been using it for a long term. But I also tried an old and stable model A4-6300 and got the same problem.
In a Live system. And I changed the resolution, the graphic driver in Mageia 8 has no problem.

thanks for the info.

you verified the download?
if you burned a USB with *isodumper* (Mageia USB tool) it will have checked the burn integrity.
have you checked the burn integrity?

you may need to edit the grub2 boot menu by highlighting an entry and clicking *e*.
then remove *quiet* & *splash* from the boot line.
you may need to add *nomodeset* as well.
*ctrl+x* will then start the live system bootup.
let us know where it stops

it may pay also to create a bug report for this issue.

If this is your first time, consult https://wiki.mageia.org/en/How_to_report_a_bug_properly

benmc wrote:it may pay also to create a bug report for this issue.

https://bugs.mageia.org/show_bug.cgi?id=31130

in your bug report you advise you did manage to get to the desktop by removing *splash* & *quiet*. correct?

when you next boot the Live, can you run

Code: Select all: drakx11

when at the desktop.
see if your graphics card is listed. if so, select it and then attempt an install.
if not, definitely note that on the bug report. sometimes drivers do get dropped.

the installer should keep your graphics card info for the installation, so there *should* be no issue for the reboot.

reply here or the bug report.

thanks for your endeavours.

benmc wrote:in your bug report you advise you did manage to get to the desktop by removing *splash* & *quiet*. correct?

Errr... In fact:
Default first boot option: Failed.
Removing “splash” & “quiet” only: Failed.
Adding “nomodeset” only: Succeeded.
Removing “splash” & “quiet” and adding “nomodeset”: Succeeded.

benmc wrote:when you next boot the Live, can you run
Code: Select all
drakx11
when at the desktop.
see if your graphics card is listed. if so, select it and then attempt an install.

I run “drakx11” and it showed a list of graphic cards. And my seletion “Radeon HD 6400 and later (radeon)” is highlighted already. I selected it and tested, got error. Tried installing. Booting installed system got same problem. Default boot option rebooted quickly. Adding “nomodeset” got

Code: Select all: [FAILED] Failed to start drakx-installer-xsetup.service.

again.

What is next? Should I do something more to help the problem?

One thing to try is choosing the card and then do not do the test. Don't know if that still applies but that used to work in some cases.

Yes, the test may leave behind an ~/.Xauthority filed owned by root, which makes starting the Xserver fail for a regular user. So don't use the test, mostly that will not even work as the card is already in use by another driver.

Germ wrote:One thing to try is choosing the card and then do not do the test. Don't know if that still applies but that used to work in some cases.

doktor5000 wrote:Yes, the test may leave behind an ~/.Xauthority filed owned by root, which makes starting the Xserver fail for a regular user. So don't use the test, mostly that will not even work as the card is already in use by another driver.

Failed.

gaojicainiao wrote:I tried Mageia 9 alpha1. Live Plasma.
In virtualbox, though I did not find anything very interesting, it works normally.
On my hardware, it could not boot into desktop environment.
EFI boot and legacy boot have the same situation.
I selected the default first boot option. It displayed three line and reboot immediately without any green “OK”.
I selected the second option “use non-free NVIDIA drivers”. It displayed a lot of “OK” and then:
[FAILED] Failed to start drakx-installer-xsetup.service.

Then stuck here.
Weird, I do not have NVIDIA graphic cards, I have only AMD FM2+ APU.

Reporting Mageia 9 beta 1 testing result. Same hardware, same problem as the previous above.

Please report it in the bug.
Thanks.

The bug thread discontinuted, I guessed I should not report again because it may be bothersome, also nobody tells me to do some new operations to test.
RC1 not usable too. At that time, I thought I should wait for the stable version to be released. But finally Mageia 9 does not boot into desktop normally on my hardwares.
So it is time to say “Goodbye”. But the more important words are “Thank you”. Mageia helped me a lot for years.
If you are interested in it, I still would like to test as you say, before I learn another distribution and get start using.

FWIW, the stable version was released this weekend. Although there won't be much differences to RC1.

Also what do you expect, if nobody is able to reproduce your bug? And you already have a workaround which allows it to boot normally - why is it in particular that you want to say "goodbye" ?

doktor5000 wrote:And you already have a workaround which allows it to boot normally - why is it in particular that you want to say "goodbye" ?

Oh. Sorry. I should have said that this workaround is not normal, the graphic driver is not used, so the resolution is not correct, also the GPU suffers, running in a low efficiency.

Trying to sum up what we know so far...
On a default install, or Live, this system need "nomodeset" added to kernel boot command line.
It then can boot to desktop, but the problem is low resolution and performance (already mentioned in first post in bug)

I think problem is best handled in the bug, so i think you should update it that the problem remain with RC1 (or final, if you can try that), and summarise similar to what I just did, if correct.
I guess the driver experts can give some hint.
If not solveable on mga9 maybe use elder kernel (I have a system that needed that...) and/or driver from mga8.

I have been working on a wiki page about selecting driver.
https://wiki.mageia.org/en/Setup_the_graphical_server

I found a scary thing!
Several other new version distributions are not able to boot into desktop environment on my hardwares too!
It looks like it is a kernel bug, it is not Mageia, it is Linux abandoning me!

It is often so that hardware manufacturers stop updating their drivers, and distributions like Mageia do not want to ship drivers with non fixed security issues.
It may also be that drivers should need to be updated to fit newer kernel.
Strangely, manufacturers do not care much about reputation for reliability, cost of ownership, environment, hassle...?
That is why i suggest to try downgrading.
Example: The machine i write on now runs Mageia 9 since a couple months, but suspend-resume do not work with kernel > 6.1x. It runs nicely with last kernel from Mageia 8 backport.

morgano wrote:Trying to sum up what we know so far...
On a default install, or Live, this system need "nomodeset" added to kernel boot command line.
~
If not solveable on mga9 maybe use elder kernel (I have a system that needed that...) and/or driver from mga8.

either the Live or your installed system, read as required.

you can install the Mga8 kernel (while in the Live or installed system) by having a read here => viewtopic.php?f=36&t=4231
add on-line repos in the Live (or installed) menu, then manually remove all but the (2) Live system ones (or just) edit and core-release and nonfree-release . Manually edit them from *9* to *8*, then save.

or you can use CLI method for the installed system.

eg:

Code: Select all: # urpmi.removemedia -a # urpmi.addmedia --distrib 'ftp://ftp.nluug.nl/pub/os/Linux/distr/mageia/distrib/8/$ARCH'

then manually choose 2 kernels, a very old one, say 5.10 series, and a 5.15 series.

When you install the Live system, both these older kernels will also be installed, as will the 6.4 series one used by Mga9 Live.
Unlike some other Installers, the Mageia installer copies the Live system as is, including any additions / removals, configurations, documents...etc you have made.

Upon reboot, choose one or the other older kernel (in the boot menu recovery option).
Let us know if there is any improvement.

Not the newest M/b, nor the eldest:

UEFI: American Megatrends v: 25.6 date: 12/15/2014

a bit messy explanation

, will try to do better.

Then re-add the Mga9 repos.
TIA

I decide to use a new grahpic card.

I got Mageia 9 installed, feeling excellent.
But my affection with Linux dropped a lot because of this kernel bug.

According to kernel developers, is this bug hopefully to be fixed soon? If not, then I fall back to Microsoft Windows to avoid the next graphic card not usable in the future.

Well, did you report it upstream? And in any case you'd have to ask the kernel developers, how should we answer that question ?

Mageia forum

[Solved]Kernel bug on graphic driver.

[Solved]Kernel bug on graphic driver.

Re: Mageia 9 alpha1 unusable on AMD FM2+ APU.

Re: Mageia 9 alpha1 unusable on AMD FM2+ APU.

Re: Mageia 9 alpha1 unusable on AMD FM2+ APU.

Re: Mageia 9 alpha1 unusable on AMD FM2+ APU.

Re: Mageia 9 alpha1 unusable on AMD FM2+ APU.

Re: Mageia 9 alpha1 unusable on AMD FM2+ APU.

Re: Mageia 9 alpha1 unusable on AMD FM2+ APU.

Re: Mageia 9 alpha1 unusable on AMD FM2+ APU.

Re: Mageia 9 alpha1 unusable on AMD FM2+ APU.

Re: Mageia 9 alpha1 unusable on AMD FM2+ APU.

Re: Mageia 9 alpha1 unusable on AMD FM2+ APU.

Mageia 9 beta1 unusable on AMD FM2+ APU.

Re: Mageia 9 alpha1 unusable on AMD FM2+ APU.

Re: Mageia 9 alpha1 unusable on AMD FM2+ APU.

Re: Mageia 9 alpha1 unusable on AMD FM2+ APU.

Re: Mageia 9 alpha1 unusable on AMD FM2+ APU.

Re: Mageia 9 alpha1 unusable on AMD FM2+ APU.

Re: Mageia 9 alpha1 unusable on AMD FM2+ APU.

Re: Mageia 9 alpha1 unusable on AMD FM2+ APU.

Re: Mageia 9 alpha1 unusable on AMD FM2+ APU.

Re: Mageia 9 alpha1 unusable on AMD FM2+ APU.

Re: Mageia 9 alpha1 unusable on AMD FM2+ APU.

Re: [Solved]Kernel bug on graphic driver.

Re: [Solved]Kernel bug on graphic driver.