FreeBSD 11.0-RELEASE-p15 won't reboot

[Dave_Berg]

Hi,
I've got a new PC that needs to run FreeBSD 11.0-RELEASE-p15. The OS installs and runs fine, but when I use reboot or halt -p, the system will shutdown to the point of telling me all buffers are sync'd and the uptime, but never reboots or powers down. Using shutdown -r reboots the system as expected.

After reading a number of posts to the forums, I've tried both PS/2 and USB keyboards. There's only the onboard SATA controller, no hardware RAID.

System specs:
Filesystem - mirrored ZFS on SSDs
Processor - Intel® Core™ i5 -8400 Processor
OS - FreeBSD 11-RELEASE-p15 (userland is p16)
System Board - ASRock H310M-HDV/M.2 System Board
System Memory - 16GB (2- 8192MB) DDR4-2666 288-pin DIMM
Hard Drives - 250GB M.2 SSD, 250GB Samsung EVO 860 SSD
Video - Intel® UHD Graphics 630
LAN - Intel® Integrated 10/100/1000 Ethernet

dmesg output:
acpi0: <ALASKA A M I > on motherboard
ACPI Error: [_SB_.PCI0.RP05.PXSX] Namespace lookup failure, AE_NOT_FOUND (201605
27/dswload2-202)
ACPI Exception: AE_NOT_FOUND, During name lookup/catalog (20160527/psobject-238)
ACPI Error: Method parse/execution failed [\134_SB.PCI0.RP04.PXSX] (Node 0xfffff
800094fb500), AE_NOT_FOUND (20160527/psparse-559)
ACPI Error: [_SB_.PCI0.RP09.PXSX] Namespace lookup failure, AE_NOT_FOUND (201605
27/dswload2-202)
ACPI Exception: AE_NOT_FOUND, During name lookup/catalog (20160527/psobject-238)
ACPI Error: Method parse/execution failed [\134_SB.PCI0.RP08.PXSX] (Node 0xfffff80009501c00), AE_NOT_FOUND (20160527/psparse-559)
ACPI Error: [_SB_.PCI0.RP13.PXSX] Namespace lookup failure, AE_NOT_FOUND (20160527/dswload2-202)
ACPI Exception: AE_NOT_FOUND, During name lookup/catalog (20160527/psobject-238)
ACPI Error: Method parse/execution failed [\134_SB.PCI0.RP12.PXSX] (Node 0xfffff80009500100), AE_NOT_FOUND (20160527/psparse-559)
ACPI Error: [\134_SB_.PCI0.RP05.PXSX.WIST] Namespace lookup failure, AE_NOT_FOUND (20160527/psargs-391)
ACPI Error: Method parse/execution failed [\134CNDP] (Node 0xfffff800095c09c0), AE_NOT_FOUND (20160527/psparse-559)
ACPI Error: Method parse/execution failed [\134] (Node 0xffffffff81ddf768), AE_NOT_FOUND (20160527/psparse-559)
acpi0: Power Button (fixed)
cpu0: <ACPI CPU> on acpi0
ACPI Error: Mutex [0x0] is not acquired, cannot release (20160527/utmutex-386)
ACPI Error: Could not release AML Interpreter mutex (20160527/exutils-147)
ACPI Error: Mutex [0x0] is not acquired, cannot release (20160527/utmutex-386)
ACPI Error: Could not release AML Interpreter mutex (20160527/exutils-147)
cpu1: <ACPI CPU> on acpi0
ACPI Error: Mutex [0x0] is not acquired, cannot release (20160527/utmutex-386)
ACPI Error: Could not release AML Interpreter mutex (20160527/exutils-147)
ACPI Error: Mutex [0x0] is not acquired, cannot release (20160527/utmutex-386)
ACPI Error: Could not release AML Interpreter mutex (20160527/exutils-147)
cpu2: <ACPI CPU> on acpi0
cpu3: <ACPI CPU> on acpi0
cpu4: <ACPI CPU> on acpi0
cpu5: <ACPI CPU> on acpi0

Anyone have ideas?

Thanks!

 

[SirDice]

I've got a new PC that needs to run FreeBSD 11.0-RELEASE-p15.

This begs the question, why? Why do you need to run an outdated version on a new PC? Why this specific version?

Topics about unsupported FreeBSD versions

Looking at the bunch of ACPI errors my first guess is that your PC has hardware that's not supported by an old version of FreeBSD. ACPI is used to power off the machine. So try 11.2-RELEASE or the latest release candidate of 12.0.

 

[Dave_Berg]

This begs the question, why? Why do you need to run an outdated version on a new PC? Why this specific version?

It's one of my production servers. Once 12.0-RELEASE comes out, they'll all get upgraded, but for now, running this version is mandated.

Topics about unsupported FreeBSD versions

Looking at the bunch of ACPI errors my first guess is that your PC has hardware that's not supported by an old version of FreeBSD. ACPI is used to power off the machine.

That's what I thought, but figured it couldn't hurt to ask.

Thanks!

 

[ShelLuser]

It's one of my production servers. Once 12.0-RELEASE comes out, they'll all get upgraded, but for now, running this version is mandated.

It's mandated to run an outdated version? That makes no sense what so ever. For the record: I know how production works, but we're talking 11.0 which went EOL one whole year earlier (November 2017, see here).

Pushing the upgrade cycle back due to production issues makes sense. But pushing it back one whole year without any planned upgrades is plain out unprofessional and a sign of severe negligence. You do realize that upgrades can be done within 20 - 30 minutes on older hardware and that minor versions don't pose as much risk as major ones?

Sorry, but this is just stupid.

 

[Datapanic]

It's not that unusual to have software freezes on a project.

 

[SirDice]

It's not that unusual to have software freezes on a project.

Lots of companies I worked for had an "end of the year change freeze". But that only lasted a month, during December, not two or more years. And if there was a major security bug during that time the change freeze went out the window.