Solaris won't shutdown (Tyan S7012 LGA1366 mobo)

Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.

TheBloke

Active Member
Feb 23, 2017
200
40
28
44
Brighton, UK
Hi all

I'm running Solaris 11.3 on a Tyan S7012 LGA1366 motherboard with 2 x X5670 2.93ghz CPUs and 72GB of ECC RAM.

I just got the Tyan mobo in the last few days. It works fine except for one annoyance: when I shutdown the server (shutdown -i5), it won't power down. It gets to "Syncing file systems.. done" and then just stops there with power on. Normally that would be the last line before the server immediately powers down.

I am using the same OS installation that I installed and used on my previous (consumer) mobo and it worked fine there. (I don't think it should cause a problem that I installed the OS on a different system then moved the SSD over?)

So currently I have to issue a shutdown command, then physically go to the server and hold down the power button for 5 seconds to get it to power down. I have to hold down the power, just pressing it once doesn't work - so I believe the OS is still in control in some way (from my experience, when in BIOS or otherwise outside of an OS, a single press of power normally shuts down, but when inside an OS, it requires a long press.)

Has anyone experienced this before, or knows some way to debug it? There are no messages or errors related to shutdown in any log that I can see - and really there couldn't be, because the issue occurs after all file system activity has stopped.

I do have some ACPI messages, including one warning, that appear in /var/adm/messages during boot:
Code:
Mar 12 12:18:58 magrathea acpica: [ID 919695 kern.notice] ACPI: RSDP 0x       0   FA890 000024 (v2 ACPIAM)
Mar 12 12:18:58 magrathea acpica: [ID 719334 kern.notice] ACPI: XSDT 0x       0BF790100 000064 (v1 092311 XSDT1015 20110923 MSFT 00000097)
Mar 12 12:18:58 magrathea acpica: [ID 148879 kern.notice] ACPI: FACP 0x       0BF790290 0000F4 (v4 092311 FACP1015 20110923 MSFT 00000097)
Mar 12 12:18:58 magrathea acpica: [ID 980510 kern.notice] ACPI BIOS Warning (bug): 32/64X length mismatch in FADT/Gpe0Block: 128/64 (20140325/tbfadt-704)
Mar 12 12:18:58 magrathea acpica: [ID 149549 kern.notice] ACPI: DSDT 0x       0BF7904F0 0063AA (v2 7012_  7012_001 00000001 INTL 20051117)
Mar 12 12:18:58 magrathea acpica: [ID 595778 kern.notice] ACPI: FACS 0x       0BF79E000 000040
Mar 12 12:18:58 magrathea acpica: [ID 579592 kern.notice] ACPI: APIC 0x       0BF790390 00011E (v2 092311 APIC1015 20110923 MSFT 00000097)
Mar 12 12:18:58 magrathea acpica: [ID 851639 kern.notice] ACPI: MCFG 0x       0BF7904B0 00003C (v1 092311 OEMMCFG  20110923 MSFT 00000097)
Mar 12 12:18:58 magrathea acpica: [ID 283188 kern.notice] ACPI: OEMB 0x       0BF79E040 000082 (v1 092311 OEMB1015 20110923 MSFT 00000097)
Mar 12 12:18:58 magrathea acpica: [ID 974727 kern.notice] ACPI: SRAT 0x       0BF79A4F0 000250 (v2 092311 OEMSRAT  00000001 INTL 00000001)
Mar 12 12:18:58 magrathea acpica: [ID 970772 kern.notice] ACPI: HPET 0x       0BF79A740 000038 (v1 092311 OEMHPET  20110923 MSFT 00000097)
Mar 12 12:18:58 magrathea acpica: [ID 990140 kern.notice] ACPI: DMAR 0x       0BF79E0D0 000130 (v1 AMI    OEMDMAR  00000001 MSFT 00000097)
Mar 12 12:18:58 magrathea acpica: [ID 564946 kern.notice] ACPI: SSDT 0x       0BF7A58C0 000363 (v1 DpgPmm CpuPm    00000012 INTL 20051117)
Mar 12 12:18:58 magrathea acpica: [ID 905775 kern.notice] ACPI: All ACPI Tables successfully acquired
Mar 12 12:18:58 magrathea acpica: [ID 190582 kern.notice] ACPI: Executed 1 blocks of module-level executable AML code
Mar 12 12:18:58 magrathea acpica: [ID 293138 kern.notice] ACPI: Enabled 3 GPEs in block 00 to 3F
Mar 12 12:18:58 magrathea genunix: [ID 958256 kern.info] Processing DMAR ACPI table
Mar 12 12:18:58 magrathea genunix: [ID 921332 kern.info] DMAR ACPI table: skipping unsupported unit type ATSR
Mar 12 12:18:58 magrathea acpica: [ID 607584 kern.notice] ACPI Exception: AE_NOT_FOUND, While evaluating Sleep State [\_S2_] (20140325/hwxface-717)
Mar 12 12:18:58 magrathea acpica: [ID 607592 kern.notice] ACPI Exception: AE_NOT_FOUND, While evaluating Sleep State [\_S3_] (20140325/hwxface-717)
Mar 12 12:18:59 magrathea acpica: [ID 491069 kern.notice] ACPI: Dynamic OEM Table Load:
Mar 12 12:18:59 magrathea acpica: [ID 301283 kern.notice] ACPI: SSDT 0xFFFFA1C02A1B7008 00603C (v1 DpgPmm CPUIst   00000011 INTL 20051117)
Mar 12 12:18:59 magrathea acpica: [ID 491069 kern.notice] ACPI: Dynamic OEM Table Load:
Mar 12 12:18:59 magrathea acpica: [ID 935601 kern.notice] ACPI: SSDT 0xFFFFA1C029B7E008 000C69 (v1 PmRef  P001Cst  00003001 INTL 20051117)
Mar 12 12:18:59 magrathea acpica: [ID 491069 kern.notice] ACPI: Dynamic OEM Table Load:
Mar 12 12:18:59 magrathea acpica: [ID 320856 kern.notice] ACPI: SSDT 0xFFFFA1C02A1C2AC8 000A0A (v1 PmRef  Cpu0Tst  00003000 INTL 20051117)

That warning is:
Code:
Mar 12 12:18:58 magrathea acpica: [ID 980510 kern.notice] ACPI BIOS Warning (bug): 32/64X length mismatch in FADT/Gpe0Block: 128/64 (20140325/tbfadt-704)
I Googled that warning and could only find info relating to FreeBSD, and the discussion I found indicated it was probably not majorly important. I have no idea if it relates to the shutdown issue I have.

It's not a huge issue because once the server is running properly, I won't shut it down often at all. And reboot works fine. But it's an annoyance I would prefer to fix.

Thanks in advance for any help
 

TheBloke

Active Member
Feb 23, 2017
200
40
28
44
Brighton, UK
What happens on the shutdown/ poweroff command
init 5
Exactly the same - I think shutdown -i5 is identical to init 5, isn't it? Except shutdown can have a schedule and sends a wall message first.

These are the messages visible on console after an init 5:
Code:
svc.startd: The system is coming down. Please wait.
svc.startd: 141 system services are now being stopped.
syncing file systems... done
Then it just stays there. I've tried leaving it a few minutes and nothing changed (except a couple of times the chassis fans audibly changed speed)
 

gea

Well-Known Member
Dec 31, 2010
3,141
1,184
113
DE
Maybe a mainbord setting/ bios problem.
If there is a newer bios available you may try an update
 

TheBloke

Active Member
Feb 23, 2017
200
40
28
44
Brighton, UK
Maybe a mainbord setting/ bios problem.
If there is a newer bios available you may try an update
Yeah I guess so. Unfortunately no new BIOS is available, the latest one is already installed (from 2012 or so)

It is not the end of the world. I have IPMI / DRAC, so I can shut the power down through IPMI after issuing init 5. It's just a minor annoyance, and makes me worry that something else might also be broken regarding ACPI - but no other symptoms visible yet.

Thanks anyway.