system does not boot after regular shutdown and power off. Kernel panic.

[jens]

System: u12-500 plus,
- 16GB RAM
- 2* SATA SSD (Samsung 850 Pro) - System, Storage Pool 2 (RAID 1), Volume 2 (ext4)
- 2* NVMe (Samsung 990 Pro) - Storage Pool 3 (RAID 1), Volume 3 (BTRFS)
- 6* Seagate IronWolf Pro - Storage Pool 1 (TRAID+), Volume 1 (BTRFS)

System was initialised and newest TOS version installed (7.0.1422).

When I shut down the system it does not start the next time I power it on (kernel panic - not syncing: No working init found. Try passing intit= option to kernel. See Linux Documentation/admin-guide/init.rst for guidance).
Powering off and a new start results in the system coming up again.
As this happened the last three times I shut the system down, I guess it is a reproducible error.

Log center doesn't show any errors.


After the restart I can see in the console:
dmesg | grep -i error
[ 1.139411] RAS: Correctable Errors collector initialized.
[ 31.322661] EDAC igen6 MC1: HANDLING IBECC MEMORY ERROR
[ 31.322662] EDAC igen6 MC0: HANDLING IBECC MEMORY ERROR
[ 31.414073] proc_thermal_pci 0000:00:04.0: error: proc_thermal_add, will continue
[ 32.003155] i915 0000:00:02.0: Direct firmware load for i915/adlp_dmc.bin failed with error -2


dmesg | grep -i warning
[ 0.000000] x86/split lock detection: #AC: crashing the kernel on kernel split_locks and warning on user-space split_locks
[ 32.042265] [668]: dev_vdisk: ***WARNING***: Unable to find /var/lib/scst/vdev_mode_pages (err -2), saved mode pages disabled. You should create this directory manually or reinstall SCST


Apps can be installed and seem to function, but it doesn't feel ok, when I have to start, power off and start again every time the system is shut down.

Any advices? Any details you would need for further analysis?

Best, Jens

[MikeZhang]

Thank you for this comprehensive report; it has significantly narrowed down the scope of the issue.

Regarding the "Kernel panic - not syncing: No working init found" error: This occurs after the kernel has booted but before user-space services have started. When the kernel fails to open the root device, it falls back to an empty rootfs; the error appears because it cannot locate `/init`. This is an issue with the boot sequence—not a problem with your data or applications—which explains why there are no entries in the Log Center: the kernel panic occurs before the logging service itself has initialized.

Since the system consistently boots successfully on the second attempt, this appears to be a timing issue during a cold boot. Your root file system resides on a RAID 1 array consisting of two SATA SSDs. We need to verify whether both drives are correctly identified and the RAID array is fully assembled before the kernel attempts to locate the root device.

Please provide the following:

1.At the next failure, photograph the screen before powering off. The lines above the panic line matter most, especially anything mentioning a root device or a missing filesystem.
2.Does Restart from TOS also fail, or only cold boot? Did this start after upgrading to 7.0.1422?
3.From a successful boot, please send the full system report from Log Center, plus the complete kernel message log. The exact commands for this will follow by private message.
4.Run the short and long S.M.A.R.T. tests on both system SSDs and send both results.

[jens]

Thank you for the fast feedback and the hints on which data you will need for debugging.
This morning - as I was well prepared for taking the pictures of the startup - the system started without any complaint.
So I powered it down pressing the on/off button for a short moment and with the next start the system came up again.
I repeated this two more times with success restart. Next I shut the system down via 'Start' menu for two or three times before I could see the error again.
Now I'm even more confused, as todays behaviour is probabilistic, in contrast to yesterdays deterministic one.


Now some answers to your questions:
>> 1.At the next failure, photograph the screen before powering off. The lines above the panic line matter most, especially anything mentioning a root device or a missing filesystem.

I have recorded the messages from beginning, but as the output is faster than the updates of the display and the sampling of the camera, it isn't completely readable.
It takes about 60 seconds from the start button to the GRUB display.
After GRUB, 'booting from RAID system' is displayed and then it only takes a second or so for the kernel to panic.
[0.81...] pcieport .... PME: Signaling with IRQ 128
[0.81...] pcieport .... PME: Signaling with IRQ 129
...
...
[0.85...] rtc_cmos ...
...
...
[0.86...] Console: switching to colour
[0.87...] fb0: EFI VGA ...
[0.87...] NET: ...
...
...
[0.90...] Demotion targets for Node 0: null
...
[0.95...] ima: No architecture policies found
...
...
[0.95...] evm: ...
...0.99]
...
[0.97...] clk: Disabling unsused clocks
[0.99...] Run /init as init process
[0.99...] Failed to execute /init (error -2)
[0.99...] Run /sbin/init as init process
[0.99...] Run /etc/init as init process
[0.99...] Run /bin/init as init process
[0.99...] Run /bin/sh as init process
[1.00...] Kernel panic - ...

I couldn't see error messages in the output rushing by. And there weren't delays where the system was waiting for some drives. No messages about missing drives or filesystem.
A photo of the last messages will come with the other details via email.


>> 2.Does Restart from TOS also fail, or only cold boot? Did this start after upgrading to 7.0.1422?
Upgrading was part of the initial setup, so I can't report about the behaviour under the previous TOS versions.
Once I have restarted the system more than once, I will give feedback.


>> 3.From a successful boot, please send the full system report from Log Center, plus the complete kernel message log. The exact commands for this will follow by private message.
From the Log Center I don't see how to export a full system report. When I try to export, it only repeats one row many times. Therefore I took a screenshot and will provide it together with the SMART reports and the kernel log.

>> 4.Run the short and long S.M.A.R.T. tests on both system SSDs and send both results.
Short and long finished without messages. I will send them via email to support@

Best, Jens

[EriChan]

Thanks for your cooperation and for providing such detailed test results and updates.

Regarding the diagnostic logs and system report, please follow these steps:

1.How to generate and export the System Report:

• Go to TOS Desktop > Support & Help > Issue Report to generate and download the complete system report package.

• Send the package to our technical support email, and attach the forum link and your forum ID as well.

• If the report file is too large to attach or fails to send directly, please upload it to a cloud drive (e.g., Google Drive, OneDrive) and share the download link in your email.

2. Enable Debug Mode for detailed diagnostics:

• We strongly recommend going to TOS > Support & Help > Debug Mode to enable debug logging. When enabled, it captures low-level logs up to the exact moment an anomaly occurs.

• When the issue reproduces: First restart the device, then return to Debug Mode to download the debug logs and send them to our support email.

• Timestamp note: If possible, please note the exact time (date and time) when the failure occurred—this helps us locate the corresponding events in the logs much faster.

• Important download note: The debug package contains three specific files: kern.log, syslog, and atop_log. Due to their specific file extensions, some web browsers may flag or block them automatically. If you notice any of these files missing after downloading, please check your browser's download manager and allow/keep the download.

Once we receive your email containing screenshots and technical information, our engineering team will immediately begin analyzing them to narrow down the root cause. You can follow up with the additional logs whenever you have them ready.

Thank you again for your patience and support.