Created attachment 867823 [details]
Journal sequence of inoperable start run

Just I found a difference in the Journal,
Please filter "wlan"

The description is too ambiguous and it's difficult to judge what is wrong.
How often is "sometimes"?  Can you reproduce it reliably?
And, is it a regression from the earlier releases?

(In reply to Peter Thoms from comment #2)
> Just I found a difference in the Journal,
> Please filter "wlan"

What does it mean exactly?

Hello, Thank you very much for working on the bug!

Yes, it is an regression to 15.4
Estimated about 5 start up and one fail.

The symptom is still not clear.  It happens occasionally after a few reboots, even if you don't touch systems (no update, etc)?

For verifying whether it's a kernel problem, you can try to install the old Leap 15.4 kernel on the Leap 15.5 system.  Grab kernel-default.rpm, kernel-default-extra.rpm and kernel-default-optional.rpm from OBS Kernel:SLE15-SP4 repo, and install them with "zypper install" with --oldpackage option. Then boot with this kernel, and verify that the problem doesn't happen with it.

> The symptom is still not clear.  It happens occasionally after a few
> reboots, even if you don't touch systems (no update, etc)?

yes, exactly

> For verifying whether it's a kernel problem, you can try to install the old
> Leap 15.4 kernel on the Leap 15.5 system.  Grab kernel-default.rpm,
> kernel-default-extra.rpm and kernel-default-optional.rpm from OBS
> Kernel:SLE15-SP4 repo, and install them with "zypper install" with
> --oldpackage option. Then boot with this kernel, and verify that the problem
> doesn't happen with it.

Oh... ok, could you please give more detailed commands for repros and packages?

You can get the SLE15-SP4 kernel from the latest git branch in OBS Kernel:SLE15-SP4 repo:
  http://download.opensuse.org/repositories/Kernel:/SLE15-SP4/pool/

Download kernel-default.rpm, kernel-default-extra.rpm and kernel-default-optional.rpm from x86_64 directory (the actual file names are with version numbers). Then install those files via
  % zypper in --oldpackage kernel-default*.rpm

But, maybe it's better to increase the number of installable kernels before the installation. Edit /etc/zypp/zypp.conf, and add more entries to the line defining "multiversion.kernels", e.g.
  multiversion.kernels = latest,latest-1,latest-2,latest-3,running

Then reboot and retest with that kernel.
Once after testing, you can remove the kernel simply by "zypper rm kernel-default-xxx" where specifying the kernel package you'd like to remove.

Ok, the new config is active.
After 20 trials 15.4 I will start up with 15.5 and I resume.

Ok, today:
After 10 start ups the last one failed

1 fail of 17 trials

1/3, 1 fail out of 3 trials

Unfortunately it's not helpful if you just write "it doesn't work"...

Basically we need:
- A more clear description of the symptom;
  does it happen only with the graphics UI?  e.g. can you remote login and test whether it equally slows down?

- Whether any relevant errors are seen;
  kernel log, journal, or other logs.  Do you see any relevant errors?

- What *exactly* did you test?  How?
  Please describe, so that other people can reproduce the very same problem with his machine.

And about the test with SLE15-SP4 kernel:

(In reply to Peter Thoms from comment #9)
> Ok, today:
> After 10 start ups the last one failed
(In reply to Peter Thoms from comment #10)
> 1 fail of 17 trials
(In reply to Peter Thoms from comment #11)
> 1/3, 1 fail out of 3 trials

So all those are the results with SLE15-SP4 kernel on top of Leap 15.5 system, right?  Or what were those?

If SP4 kernel also fails, you can try to install older SLE15-SP4 kernel instead -- the version you had and worked in the past.  It must be found in the standard distribution repo in OBS.

Ok,
I am always using the Power Button to start up to KDE.

Sorry, there is no remote possible, the ethernet via wlan is down.
By the way, even the internal speaker seem to be off, indicated by a small icon.

What about a command to serve your wishes and intercept the fault corresponding to the uploaded journal?

As a start, please give hwinfo output from the good running system (with both Leap 15.4 and Leap 15.5 kernels).  Then give the dmesg outputs from both kernels.

Created attachment 868253 [details]
hwinfo, start trial succsessful

uname-r 5.14.21-150500.53-default

Summary of fail start:

Start a terminal possible
"Enter new line" possible
sudo hanging in the new line, no action
also
atop
sudo hwinfo > peter (only generating an empty file)

kinfocenter is starting
no kill and no term for sudo, atop etc. possible

kinfocenter/network is hanging, no quit possible

shut down runs into a black screen with cursor blinking on Pos 1
Power down with Power button

I got some screenshots.
Please ask for if might be helpful.

Hm, this doesn't sound like a usual graphics problem, then.

The best thing we can do for now would be to set up kdump, and trigger manually via alt-sysrq-c, and get the crash dump.

Ok, I got a capture.
How to transfer the dir?

Created attachment 868296 [details]
kdump README.txt.gz

Created attachment 868297 [details]
kdump System.map-5.14.21-150500.53-default.gz

Created attachment 868298 [details]
kdump vmlinux-5.14.21-150500.53-default.gz

Created attachment 868299 [details]
kdump dmesg.txt

vmcore is not transferable
Do you need vmcore? Then I will try to put it in the cloud

Thanks.  You can't upload the full crash contents as they are too big, and README, System.map and vmlinux are superfluous (they are as same as found in the kernel package).  The dmesg is helpful, and at least it doesn't show anything suspicious.

That is, we see no sign of a kernel crash or such at this moment.  You can check it with "crash" program (included in crash package) for further investigation in your side, too.  There one can see which processes are running, and examine the stack traces.

OTOH, I wonder whether this is rather related to desktop instead.  When the problem happens, can you try to switch to VT1 via ctrl-alt-F1, and see whether things work?  Also, can you try to kill the desktop forcibly?

But, above all, please retest with the freshly released Leap 15.5 update kernel (5.14.21-150500.55.7).

Many thanks!

Could it be that this Laptop only misses an real Handshake at the right place?

Maybe some of my hardware is straight malade.
When it appears both the wlan and voice output (speaker) are out of function
Can you see what both have collectively?

Please give command set for switch over to the newest kernel.

Created attachment 868326 [details]
pci info

Could it be helpful having a focus at PCI?

Now linux got a simple update command

(In reply to Takashi Iwai from comment #24)
...
> OTOH, I wonder whether this is rather related to desktop instead.  When the
> problem happens, can you try to switch to VT1 via ctrl-alt-F1, and see
> whether things work?  Also, can you try to kill the desktop forcibly?

Ok,
to start VT1 I have to use Strg-Fn-Alt-F1, now I`m able to start it.

What about to kill the desktop, what is the name of via VT1?

At first, check whether you can work normally on VT1.  For example, the command you tried beforehand and stalled (sudo, top, whatever).  Possibly you can face the same problem on VT1, too.

If everything seems working on VT1, then simply kill the whole desktop session.
If you're running X, killing X.org process will do it.  (Or pressing ctrl-alt-backspace *quickly* twice on the desktop screen might work, too -- it's self-killing X server.)

Hello, here some news, quite short:
After starting VT1 I saw only the cursor blinking at Pos1, no commands possible

So I switched back to the gui and started a teminal and got there a prompt

Today a strange situation:
new:
bluetooth mouse inavtive all the time, only mousepad usable
But the bluetooth setup is working with my fujitsu-pc

Fail start: no VT1 possible, no crashdump possible

I think, all my external usb ports passed away

ok,
the usb Hardware on the motherboard passed away.
WIN7 runs all the time well, but newly without USB.

If you are more interested in the actual linux behavior, I will hold on debugging.

The future of this laptop will be selling it as replacement (with defect motherboard: USB inactive).

Hardware malade
Thank you for patience!

OK, thank you for the information update.