FreeBSD 13.0 freezes

… bectl was sometimes failing to destroy no longer needed boot environments …

If I understand correctly, this is by design for some situations … for no longer needed snapshots.

I previously misunderstood the manual page. After (hopefully) learning: <https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=254428#c2>. tl;dr when I most recently encountered the warning, I knew to use something other than bectl to (if necessary) destroy the remaining snapshot(s); and I recall thinking that the snapshots might have been a product of beadm.

Notable bugs include:
Postscript (afterthought): long ago I did find that some boot environments were indestructible. It puzzled me but not enough to raise a bug, I might have found the BEs indestructible with both beadm and bectl. Whatever the problem was – maybe (?) the result of me creating a boot environment whilst not running the 'most recent' environment – eventually it disappeared. I found it possible to destroy the environments.
 
I got freezes on FreeBSD 13.0 too. Then, I simplified my settings, and reverted to the kernel that came with the install for it to work for now.

On my running system, this shows up:
dmesg | grep -i acpi | egrep -i 'warning|error'
Code:
Firmware Warning (ACPI): Optional FADT field Pm2ControlBlock has valid Length but zero Address: 0x0000000000000000/0x1 (20201113/tbfadt-796)
Firmware Error (ACPI): Could not resolve symbol [\134_SB.ALIB], AE_NOT_FOUND (20201113/psargs-503)
ACPI Error: Aborting method \134_SB.PCI0.VGA.ATC0 due to previous error (AE_NOT_FOUND) (20201113/psparse-689)
ACPI Error: Aborting method \134_SB.PCI0.VGA.ATCS due to previous error (AE_NOT_FOUND) (20201113/psparse-689)
These error messages or similar ones also showed up before more fatal error warnings and system freeze.

It's likely the power settings or the video driver causing overheating or another fatal issue. My CPU indication was a little hot when looking at BIOS.

* Edit: this is a desktop with an AMD CPU/GPU.
 
Does anyone know if this issue has been reported? Is anyone
working on it?
Just FYI but this problem seems like some specific race condition with P-states and Lenovo systems. (and I also think only U series processors).


The more people subscribe to the bug(s), the more attention they well get :).
 
Last edited:
Just FYI but this problem seems like some specific race condition with P-states and Lenovo systems. (and I also think only U series processors.


The more people subscribe to the bug(s), the more attention they well get :)
I'll subscribe to both of these bugs, thank you very much for this information!
 
Back
Top