In windows 11 enterprise edition, I have events 103 AESMService: Platform Services initialization failed due to DAL error. How to stop these messages?

Question

I'm running an hp all-in-one and am getting event id 103, AESMService: Platform Services initialization failed due to DAL error. I believe I have followed all the usual steps to resolve to no avail., including:

sfc /scannow

DISM /Online /Cleanup-Image /RestoreHealth

chkdsk

I still have the same error this morning,

<Event xmlns="**http://schemas.microsoft.com/win/2004/08/events/event**">

• <System> <Provider Name="AESMService" Guid="{ce6e83d3-a7d9-4a91-96e0-e018ad574610}" /><EventID>103</EventID><Version>0</Version><Level>3</Level><Task>0</Task><Opcode>0</Opcode><Keywords>0x8000000000000000</Keywords><TimeCreated SystemTime="2024-03-15T14:09:00.0176140Z" /><EventRecordID>190</EventRecordID><Correlation /><Execution ProcessID="1864" ThreadID="7900" /><Channel>SGX/Admin</Channel><Computer>DESKTOP-ENSLEYS</Computer><Security UserID="S-1-5-18" /></System>
• <EventData> <Data Name="attrUnicodeString">Platform Services initialization failed due to DAL error</Data></EventData> </Event>

I also have an error 28,

<Event xmlns="**http://schemas.microsoft.com/win/2004/08/events/event**">

• <System> <Provider Name="Microsoft-Windows-Kernel-EventTracing" Guid="{b675ec37-bdb6-4648-bc92-f3fdc74d3ca2}" /><EventID>28</EventID><Version>0</Version><Level>2</Level><Task>3</Task><Opcode>25</Opcode><Keywords>0x8000000000000a20</Keywords><TimeCreated SystemTime="2024-03-15T14:06:50.4097255Z" /><EventRecordID>30</EventRecordID><Correlation /><Execution ProcessID="968" ThreadID="972" /><Channel>Microsoft-Windows-Kernel-EventTracing/Admin</Channel><Computer>DESKTOP-ENSLEYS</Computer><Security UserID="S-1-5-18" /></System>
• <EventData> <Data Name="ProviderGuid">{77811378-e885-4ac2-a580-bc86e4f1bc93}</Data><Data Name="ErrorCode">3221225477</Data></EventData> </Event>

This has been persistent and I have tried everything to fix it.

Any help on steps to solve these issues would be greatly appreciated.

Thank you!

***moved from Windows / Windows 11 / Performance and system failures***

Answer 1

I seen you have posted your issue in HP community forum for asking sgx drivers.

Setup file is worked for 64 bit processor but it in your computer not installed,because system SKU might be corporate series. Consumer series system sku worked fine.

I am in hospital now .

From my side no other help. But afternoon I wil find dal inf file ,and I will give the path Location.

Moreover I am not Processor.Just I am teacher .

Answer 2

C:\Users\yourusername\Downloads\intel_mei_2401.5.50.0(station-drivers.com)\0\Drivers\DAL.inf

dal.inf is used for installing the driver manually.

Answer 3

Dear Raju:

Installed intel_mei_2401.5.50.0(station-drivers.com).zip successfully. Here is a screen shot of the driver manager showing the version of the driver, the newest one.

My issue has now morphed into the following event viewer event (critical AESM event 108):

The error springs from the SGX problem of the AESM driver not being able to bind to the TCPIP:

Please let me know how to remove this critical error, which is causing the AESM to stop doing this:

AESMService: SGX is Disabled at AESM Service startup

Here is a shot of the aesm_service (Intel SGX) details:

After two reboots, the SGX diagnostics events are:

Where you can see that it is not able to bind properly both times.

The corresponding administrative SGX events:

I have not seen a DAL warnings pop up for some time. I think installing the Intel ME Components did the trick for those errors. Now this has morphed into a more serious critical error 108, from aesm_service.

There should be no need to adjust or install anymore the DAL drivers. It should not be a problem with the Intel stuff, now maybe the OS or the BIOS are not configured right. That is all I could glean off the intenet about this critical error 108.

It must be that the aesm_service in order to start MUST first bind to a socket, tcp/ip in this case. Since it can not even bind, it can 't start (WHICH IS WHAT THE NEXT EVENT EXPECTS). It should have started, and then the error would not have come up. The root of the issue seems to be that something is preventing it from binding to the tcpip, so that the service can not even start. Maybe the timing is such that the tcpip has not been started yet? Maybe the tcpip is too slippery?

I can only guess, but there must be tools out there to help diagnose this sort of problem.

I was just sharing a few thoughts on this. At least I have the right Intel drivers installed now! Just left with a critical error telling me that it can't start itself because it can't bind to a tcpip causing the critical 108 error from aesm_service.

Hopefully, the solution will soon become evident!

Cheers,

Don Ensley

Answer 4

Today I made the progress of setting the BIOS on my PC to indicate enable for the SGX, instead of under software control. This changed my errors substantially, so that now, the AESM service is no longer shut down. It is kept open! There are no more errors in the sgx_admin, and only the 2 errors in the sgx_diag that were previously noted.

The SGX diagnostics show the 2 small errors, not admin level:

Error 205: AESMService: Failed to load QE3: 0x4004

Error 205:AESMService: [load_qe X:\windows-trunk\dcap-trunk\dcap_source\QuoteGeneration\quote_wrapper\quote\qe_logic.cpp:721] Error, call sgx_create_enclave QE fail [load_qe], SGXError:4004.

The new SGX admin events (no errors):

I heard some say that a newer BIOS would clear these errors, but I have the latest I could find installed BIOS Version/Date AMI F.33, 8/23/2023. That is pretty recent!

At least the service does not hang, or have a critical error any more. It is safe to say that the SGX needs to be set to "enabled" instead of "controlled by software".

Consider this case closed, since the original DAL messages no longer occur and since I can theoretically eliminate down to 0 errors, as I only have that one big error remaining, the Kernel-EventTracing sourced error id 28. I don't think the errors under the hood in the SGX diagnostic log (perhaps from lack of an updated BIOS) will hinder the AESM service, since it now appears to run right around those things.

Error setting traits on Provider {77811378-e885-4ac2-a580-bc86e4f1bc93}. Error: 0xC0000005

That is my last one to tackle.

Thanks for all your help!

I hope you are well or recovering if not.

Don Ensley

Answer 5

Today (last night) had PC crash, then had to reinstall everything. The blasted warning 103 came back again.

At this juncture, I am not really sure why I am getting the bizarre errors in the SGX diagnostics that indicate 205 errors, I have been told I need a newer BIOS, but mine is new from 2023. So I do not think it is the BIOS. I will play around with the settings and see if I can get this thing working. It is most frustrating!

We are back to square one, but with newer drivers and a DAL loader.

INFORMATION FROM AESMService:

http://schemas.microsoft.com/win/2004/08/events/event"> - 202 0 4 0 0 0x4000000000000000 156 SGX/Diagnostic DESKTOP-DENSLEY - AESM will not bind to a TCP socket. Attempting to open the AESM_SOCKETS REGISTRY_SETTING_BIND_ADDR registry entry returned = 6

ERRORS IN FROM THE SGX/Diagnostic FOLDER:

http://schemas.microsoft.com/win/2004/08/events/event"> - 205 0 2 0 0 0x4000000000000000 154 SGX/Diagnostic DESKTOP-DENSLEY - [load_qe X:\windows-trunk\dcap-trunk\dcap_source\QuoteGeneration\quote_wrapper\quote\qe_logic.cpp:721] Error, call sgx_create_enclave QE fail [load_qe], SGXError:4004.

http://schemas.microsoft.com/win/2004/08/events/event"> - 205 0 2 0 0 0x4000000000000000 155 SGX/Diagnostic DESKTOP-DENSLEY - Failed to load QE3: 0x4004

That is all I have for now. I hope to get it working again.

Hope all is well,

Don Ensley