NUC8i5BEH is registering iTCO_wd watchdog wronlgy and so watchdog is unusable!

[IFisc]

I upgrade from NUC5PPYU because I wanted more RAM and now I end up in having a system with an unusable watchdog :-(

dmesg under Linux show:

[   6.929584] iTCO_vendor_support: vendor-support=0

[   6.930178] iTCO_wdt: Intel TCO WatchDog Timer Driver v1.11

[   6.930251] iTCO_wdt iTCO_wdt: can't request region for resource [mem 0x00c5fffc-0x00c5ffff]

[   6.930259] iTCO_wdt: probe of iTCO_wdt failed with error -16

this matches to infos that can be found at https://www.spinics.net/lists/linux-i2c/msg41504.html where Linux Kernel devs analyzed the tpic and said that it needs to be fixed in BIOS from NUC.

Please forward this as bug to your BIOS developers to get that fixed. Thank you!

[AlHill]

As you know, only Windows 10 is supported on your NUC.

https://www.intel.com/content/www/us/en/support/articles/000005628/mini-pcs.html

The latest bios is 0074 and can be found here:

https://downloadcenter.intel.com/download/29077/BIOS-Update-BECFL357-86A-?product=126148

As for issues with Linux, I suggest you contact the linux community. The Intel support engineers monitor this forum (and thread) and will look at the issue.

Doc

[IFisc]

The iTCO_wd is a feature of Intel chipsets since 15 years and it should be supported correctly. And when it is exposed wrongly by BIOS for Linux then for Windows too I expect. Thats why the forum name here sounded ideal :-)

I'm currently on BIOS 0073 and will try 0074 tomorrow, but I expect it only means the NMI watchdog, but we will see.

But Ok I will do a cross post in linux community :-)

Ingo

[IFisc]

I wasn't able to find " linux Community area" here for Nucs ... so If you mean the Linux COmmunity itself then my Link is exctly coming from them ... But They are our if the BIOS is doing things wrong

[AlHill]

I was referring to linux.org.

Doc

[AlHill]

I just read the release notes on 0074, and there is a fix for the watchdog, but may not affect you.

https://downloadmirror.intel.com/29077/eng/BE_0074_ReleaseNotes.pdf

Doc

[IFisc]

As expected 0074 do not change anything here ... :-(

I also Googled a bit more and some discussions in Linux and other communities (also for some Lenovo Laptops and such) ... but the link in my first post referencing on BIOS is the detailed analysis so far (from my "not kernel or bios developer understanding at least ;-) )

Hopefully an Intel BIOS experts also checks and reads this and maybe we get a fix ;-)

[AlHill]

Sorry it still does not correct the issue. One of the Intel support engineers will assist you.

Doc

[IFisc]

Hopefully ;-) And thank you also for your time and answers!

[David_V_Intel]

Hello IFisc,

Thank you for posting on the Intel ® communities.

I would like to know the model and part number of RAM you used to upgrade your system, does the issue stop if you insert the old RAM modules?

Let me know so I can analyze further.

David V

Intel Customer Support Technician

A Contingent Worker at Intel

[IFisc]

Hi,

It's Crucial RAM 2x8GB, details:

# dmidecode 3.0 Getting SMBIOS data from sysfs. SMBIOS 3.2.1 present. # SMBIOS implementations newer than version 3.0 are not # fully supported by this version of dmidecode.   Handle 0x003C, DMI type 17, 40 bytes Memory Device Array Handle: 0x003B Error Information Handle: Not Provided Total Width: 64 bits Data Width: 64 bits Size: 8192 MB Form Factor: SODIMM Set: None Locator: SODIMM1 Bank Locator: Memory Channel A Type: DDR4 Type Detail: Synchronous Speed: 2400 MHz Manufacturer: 859B Serial Number: E0FE8E38 Asset Tag: 9876543210 Part Number: CT8G4SFD824AC16FBD1 Rank: 2 Configured Clock Speed: 2400 MHz Minimum Voltage: 1.2 V Maximum Voltage: 1.2 V Configured Voltage: 1.2 V   Handle 0x003D, DMI type 17, 40 bytes Memory Device Array Handle: 0x003B Error Information Handle: Not Provided Total Width: 64 bits Data Width: 64 bits Size: 8192 MB Form Factor: SODIMM Set: None Locator: SODIMM2 Bank Locator: Memory Channel B Type: DDR4 Type Detail: Synchronous Speed: 2400 MHz Manufacturer: 859B Serial Number: E0FE7776 Asset Tag: 9876543210 Part Number: CT8G4SFD824AC16FBD1 Rank: 2 Configured Clock Speed: 2400 MHz Minimum Voltage: 1.2 V Maximum Voltage: 1.2 V Configured Voltage: 1.2 V

I do not have other RAM, no RAM was included. And I can not use the RAM from the "old" NUC5PPYH because they are not compatible regarding voltages.

Can I provide other details to support the investigation?

[IFisc]

Hi David,

I now saw that BIOS 75 was released, but the changelog do not sound like it contains a fix for this, or is it in one of teh changelog topics?

1. New Fixes/Features:
2. Fixed issue where changes to “Native ACPI OS PCIe Support setting in BIOS doesn't stay.
3. Implemented security fixes
4. Updated processor microcode to MC0806EA_000000C6.pdb

[YanaT_Intel]

Hello  IFisc,

We are still waiting for the BIOS containing the fix. We apologize for the delay and your inconvenience.

Thank you

Yana

[IFisc]

Thank you, great to know that it reached the correct persons and it being worked on

[IFisc]

@Yana_Intel​ Happy new year as first :-)

In the meantime we have BIOS 0077 ... If not hidden in one of the other changes still no Watchdog fix in there, or?!

• Updated ME to 12.0.47.1524.
• Fixed issue where “Chassis value type changed during

flash by BIO” for Microsoft request

• Fixed issue with auto detect Thunderbolt device.
• Fixed issue with basic standard speed performance.
• Fixed issue where “SMBIOS Field update BIOS” for Microsoft request.
• Fixed issue with Thunderbolt controller.
• Fixed issue finding Bluetooth device.
• Fixed issue waking up Bluetooth in Legacy mode.

There also was an "update in "Memory Reference Code" (Based on 7.0.44.23 --> Based on 7.0.51.41). Could this be related?

[Alberto_R_Intel]

Hello IFisc, I just received an update on this matter.

The Ubuntu is not supported by the Intel® NUC Kit NUC8i5BEH

and the ROI is not enough for the feature implementation.

The feature will not be implemented due to the Operating System support limitations.

Any questions, please let me know.

Regards,

Alberto R.

Intel Customer Support Technician

A Contingent Worker at Intel

[IFisc]

Thank you Alberto for the answer,

while 100% NOT agreeing with the view of Intel on this topic it seems I only can accept it and need to switch the hardware platform.

In my eyes it is "just" an incorrect mapping on memory definition in Bios mapping the device RAM. The watchdog on intel motherboards is there since >10 years and was working perfectly e.g. on NUC5PPYH that I used so far. But this models is EOL and so I wanted to stay within a proofen hardware platform, but this decision makes this very problematic.

I also do not say anything about the "Linux (in my case Debian not Ubuntu) is not supported" topic - here Intel should change his strategy because saying it that way is simply too easy ... WIndows is simply not a good server OS (my opinion still).

Also saying that it will not be done due to "os limitations" is simply wrong - please see the details of Linux kernel pro's in the list in my first post. This is nothing that can be fixed by kernel :-( So it is a bug in the Intel BIOS and has nothing to do with the OS.

I'm saying a very unhappy goodby ...