I'm an experienced builder and software developer, having built all my own workstations since 1990.
I recently built a new workstation running Windows 10 Enterprise Edition using the parts documented here:
PCPartPicker part list: https://pcpartpicker.com/list/sKJfWD
CPU: Intel - Core i9-9900K 3.6 GHz 8-Core Processor
CPU Cooler: be quiet! - Dark Rock 4 CPU Cooler
Motherboard: MSI - MPG Z390 GAMING PRO CARBON AC ATX LGA1151 Motherboard
Memory: G.Skill Ripjaws V Series
Storage: Samsung - 970 Evo 1 TB M.2-2280 Solid State Drive
Case: NZXT - H500i
Power Supply: CORSAIR RMX Series, RM750x, 750 Watt, 80+ Gold Certified, Fully Modular Power Supply
I re-used some components from a prior build:
- A GeForce GT710 GPU for multi-monitor
- A USB 4-port extender card
- A discrete Intel Ethernet card.
I am not overclocking. I am running the memory at the designated 3200 speed with SPD values.
I have been experiencing random reboots. Several a day.
Event Log contains the following:
Log Name: System
Date: 4/10/2019 1:14:53 PM
Event ID: 19
Task Category: None
User: LOCAL SERVICE
A corrected hardware error has occurred.
Reported by component: Processor Core
Error Source: Corrected Machine Check
Error Type: Cache Hierarchy Error
Processor APIC ID: 2 (varies - sometimes 6, sometimes 2. No other numbers seen.)
Shortly after these events (within a minute or two), the machine either freezes or quickly reboots. No BSOD, no prompt, no memory dump.
The CPU runs 33C when idle and 75C max running prime95 small FFTs.
Things I've Tried
- Ran all Windows Updates
- Ran sfc /scannow and dsim commands. No problems found.
- Upgraded motherboard BIOS to latest version.
- Ran the Intel Driver & Support Assistant and installed new drivers for WiFi and Bluetooth components.
- Swapped out the RAM for Corsair - Vengeance LPX 64 GB. memtest86 - passed
- prime95 small FFTs for 30 minutes - didn't reboot while running.
- Removed secondary discrete Ethernet card
- Removed secondary USB expansion card.
- Removed discrete GPU card GeForce GT710, now using only integrated graphics
- In Windows Advanced Power Settings, disabled the default of allowing the hard drive (SSD in this case) to sleep after 20 minutes idle.
- Logged CPU heat over time with hwinfo. Had a freeze-up while logging. The CPU temps were nominal at the time just prior to reboot.
- Removed PC from UPS (CyberPower CP1500PFCLCD PFC Sinewave UPS System, 1500VA/900W) in case UPS was not sending clean power (although I've used the same UPS with another build for 4 years).
- Installed April Microsoft patches.
- Disassembled all components, checked for shorts, cleaned off thermal paste and re-applied.
- Ran the motherboard outside the case.
- Installed the Intel Processor Diagnostics Tool 64Bit 4.1.33.W.MP, passed all tests.
At this point, there is nothing left to update. All BIOS and drivers are running the latest version.
I've spent hours troubleshooting this build. I'm in need of advice.
David, I don't have another motherboard that accepts this CPU. Are you suggesting I purchase another one to test? I feel I'd be better off seeking an RMA replacement from Intel at this point.