[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Obscure error messages from ACPI BERT in dmesg



Hi,

i don't get to cursing at vim and sshd of Debian 10 because i am still
stuck with hardware issues.

My newest problem is a message at boot time about some problem that was
recorded in the ACPI Boot Error Record Table during the previous uptime.
Google and uefi.org tell me that BERT is a facility where firmware can
store error messages which it deems at risk of being not delivered via
the normal error message channels of a running OS.

[    0.006446] ACPI: BERT 0x0000000089F05608 000030 (v01 INTEL  EDK2     00000001 INTL 00000001)
...
[    0.893410] BERT: Error records from previous boot:
[    0.893410] [Hardware Error]: event severity: fatal
[    0.893411] [Hardware Error]:  Error 0, type: fatal
[    0.893412] [Hardware Error]:   section type: unknown, 81212a96-09ed-4996-9471-8d729c8e69ed
[    0.893412] [Hardware Error]:   section length: 0xc20
[    0.893413] [Hardware Error]:   00000000: 00000001 00000000 00000000 ffffffff  ................
[    0.893414] [Hardware Error]:   00000010: 00000000 00000000 00000000 00000000  ................
[    0.893414] [Hardware Error]:   00000020: ffffffff ffffffff ffffffff ffffffff  ................

... repeated in steps of 0x10 and all words being ffffffff until:

[    0.893505] [Hardware Error]:   00000c10: ffffffff ffffffff ffffffff ffffffff  ................

Google search of 81212a96-09ed-4996-9471-8d729c8e69ed gives the impression
that it is like plague and cholera united. (Computers freeze or reboot ...)

I do not see such dramatic effects, though.
Meanwhile i outruled that it is the knock-happy HDD (because unplugging it
does not prevent or change the message at next boot) or the missing i915
firmware (i installed it).

In the hope to learn more about that UUID or the format of BERT i even
downloaded
  https://uefi.org/sites/default/files/resources/ACPI_6_2_B_final_Jan30.pdf
But my hope to see UUID lists for error types was disappointed.


Does anybody have an idea how to identify the offending hardware part ?
(I cannot unplug CPU or SSD, of course ...)


Have a nice day :)

Thomas


Reply to: