cancel
Showing results for 
Search instead for 
Did you mean: 

WHEA - CmciRestoreAttempts=1

ttesty_1
Level 9

Seem to be getting this message only on Windows 24H2 (latest) startup (on a Z890 Hero @ BIOS 1302) each time. Any idea why I see this (CmciRestoreAttempts=1)? I also disabled Native PCI Power Mgmt.

["A "Corrected Machine Check Interrupt Restart" refers to a situation where a computer system experiences a hardware error detected by the processor (known as a "Machine Check Exception" or MCE), but the system is able to automatically correct the issue and continue running without requiring a full reboot, essentially restarting itself at a lower level within the operating system. The CPU detects a potential hardware problem, like a memory corruption or cache issue, triggering a Machine Check Interrupt (MCI)."]

============================

{
"Header": {
"Signature": "WhLg",
"Version": 1,
"Length": 48,
"Type": "Informational",
"OwnerTag": 541868360,
"Id": "CmciRestart",
"Flags": "LogInternalEtw",
"PayloadLength": 16
},
"Event": {
"CmciRestoreAttempts": 1,
"MaxCmciRestoreLimit": 3000,
"MaxCorrectedErrorsFound": 0,
"MaxCorrectedErrorLimit": 20000
}
}

============================

151 Views
4 REPLIES 4

ttesty_1
Level 9

Windows 11 Version 10.0.26120.2705. Sometimes I get a sudden hard crash/instant reboot without a dmp file written. Temperatures are perfect in the system. Strange behavior. Uninstalled XTU as well.

You seem to know a lot about the situation, so I can only guess.  Could that be an indication of the firmware update needing to do a partial restart to adjust the Memory speed or configuration?

You also seem to be running an Insiders Windows build.  I will go try that and report back if I see anything there, but I do not see it on public 24H2.

Maximus Z890 Hero,
Intel Core Ultra 9 285K

Like I said, this happens each time I boot Windows (WHEA event appears after the amount of boot up time is reported). So I'm not clear why there would need to be a "firmware update needing to do a partial restart to adjust the Memory speed or configuration"  each time. It's also running on the 1302 BIOS. Memory is G.Skill F5-6400J3239F48G (96GB Kit of 2x48 sticks) under conservative settings (XMP2). Nothing strange in memtestx86. Thinking this is more of a hardware issue v.s. Windows itself, but we'll see.

Probably going to take a year or so of BIOS, Microcode updates to iron out issues. Also waiting for 9200MHZ+ 2x48GB CUDIMM's to release. Going to be a long wait!

ttesty_1
Level 9

I've also ran the system on Prime 95 for 10+ mins w/ no reboots. CPU/memory gets really prickly sometimes with MCE-style fast reboots when there's nowhere near the load of Prime 95, especially for basic overclocking (like running Intel XTU and AIDA64 benchmarks => fast reboot). So I've eliminated that variable.

One other suggestion: inform user that system will reboot about 10x (or whatever amount) in sequence when doing memory changes/training in BIOS. Hard to tell if there's something genuinely wrong. And make the number of reboot failure detection (i.e. bad memory settings) retry cycles be user-configurable before going to back to defaults. Gets annoying watching post fail 10 times and having to push a BIOS reset button to resolve.