unable to shutdown or poweroff NUC10i7 / ubuntu 22.04.1

[Pura-Vida]

Hi,

I am really disappointed, I own 2 x NUC 10i7FNH 64GB/2TB 970 EVO plus.

Both are fresh install of Ubuntu 22.04.1 (workstation for one and Server for the other), and both got issue to shutdown, as 4 shutdown out of 5,  the system is rebooting rather than shutting down as requested.

Command used:

- poweroff

- shutdown -P now 

- sudo poweroff

- sudo shutdown -P now

My user is with admin credential 

BIOS and firmware are the newest versions.

sudo fwupdmgr get-devices (UUID, GUID are removed from this log)

NUC10i7FNH
│
├─SSD 970 EVO Plus 2TB:
│ │ Summary: NVM Express solid state drive
│ Current version: 2B2QEXM7
│ Vendor: Samsung (NVME:0x144D)
│ 
├─System Firmware:
│ │ │ Summary: UEFI ESRT device
│ │ Current version: 58
│ │ Minimum Version: 58
│ │ Vendor: Intel(R) Client Systems (DMI:Intel Corp.)
│ │ Update State: Success
│ │ └─UEFI dbx:
│ │ Summary: UEFI revocation database
│ Current version: 267
│ Minimum Version: 267
│ Vendor: UEFI:Linux Foundation
│ Install Duration: 1 second
│ 
└─Thunderbolt host controller:
Summary: Unmatched performance for high-speed I/O
Current version: 42.00
Vendor: Intel Corporation (THUNDERBOLT:0x8086, TBT:0x8086)

[LeonWaksman]

I got in past similar problem in NUC FN. I solved it by reconnecting the CMOS battery connector and reconnecting it after 15 minutes.

In the same time I have to add that my system was having this problem running on Windows 10 Pro, while Linux wasn't validated on NUC FN.

Your NUC BIOS is not updated. The latest BIOS is version 059.  In the Release Notes for version 059 there is a following comment: "Fixed issue where Grub Bootloader Vulnerability." So, I would suggest to update your BIOS as well. On BIOS Download page, you may find link to Read me File with the BIOS update instructions.

Leon

[Pura-Vida]

Hi Leon, 

Thanks a lot for your information.

I did it this morning (Bios Upgrade to 59 and also UEFI dbx upgrade (old 77, target 267).

pros: shutdowns seem to be a shutdown and not a reboot but this upgrade has killed my Ubuntu server 22.04.1 setting.

The workstation runs fine, but the server 22.04.1 is going into kernel panic, impossible de reboot.

I've tried to reinstall from scratch ==> impossible: Ubuntu 22.04.1 server IS NOT COMPATIBLE WITH NUC10i7FNH with last BIOS/ UEFI Signature, kernel turns to panic, memory faults (64 GB, and no memory available).

So I've tried to reinstall 20.04 LTS Server: Impossible GRUB signature used by this version is listed into the blacklist carried by UEFI dbx upgrade!

I really do not appreciate the Job Intel is doing. 

40+ y.o. architecture and Linux is not working on their 100% internal system!

I gonna install workstation 22.04.1 to get this NUC run!

Amazing. I hope that Canonical and Intel will hand check very soon or next architecture will be from RISC V or ARM. 

[Pura-Vida · Accepted solution]

Hi Leon, 

I've preferred to let some days to confirm changes:

The issue is not full fixed, but it is a major step forward as 90 of poweroff are now  poweroff and not a reboot.

There is some case that poweroff still continue to run a reboot instead. but they do not represent more than 5% of the previous situation.

downgraded features and critical isssues:
59 firmaware and dbx UEFI database upgrades cancel 20.04 installation with secure boot feature enabled as grub version of the Ubuntu 20.04 is into the black listed entries of the dbx AND UBUNTU 22.04.1 Server edition is looping on kernel panic (memory violation) after firmware upgrade and reinstallation from scratch is doing the same.

NUC ARE NOT the best friend of the Open IT!
who can imagine a worst feature for future marketing communication: Open minded people will have to move to RISC V and Linux to get Open system!