F5F Stay Refreshed Hardware Desktop The workstation stops working yet remains switched on.

The workstation stops working yet remains switched on.

The workstation stops working yet remains switched on.

A
AmericanExpats
Junior Member
Find
20
11-14-2016, 10:46 PM
#1
This machine runs an HP Z440 CPU: Intel Xeon E5-1660 v3 (8C/16T). It’s not overclocked, just using regular RAM: 256GB ECC DDR4 at 2133 MHz. The GPU is a NVIDIA Quadro K2000, and it’s powered by an HP 700w PSU that came with the setup. I purchased an adapter to use a standard PSU when needed. The BIOS is Hewlett-Packard M60 v02.61, with SMBIOS 2.8 and Windows 10 Pro (64-bit). I’m using it to operate the AMP instance manager, which handles low-pop servers and a small 7D2D server. I also run CMD for Discord bots—one manages tickets, the other processes skin IDs for staff approval. I’m not using AMP in Docker containers. AIDA 64 Extreme is installed for monitoring via the sensor panel.

My problem is the system “crashes” but stays on. Specifically, I lose graphics card visibility and can’t access Windows Remote Desktop. The only way to restart it is to press the power button and then turn it back on. The biggest issue is unpredictable crashes—sometimes after hours, sometimes weeks. I’ve tried: running MEMS test, Intel Processor Diagnostics, and changed OS from Windows Server 2019 to Windows 10 thinking the OS was faulty.

Possible causes: CPU damage, BIOS fan settings, or overheating. At the time it started, I thought the CPU temperature should be around 74-84°C, but the max was only 66°C for a 140W unit. The PSU condition is unknown, and I lack the tools or funds for a proper check. I bought adapter cables to connect a standard PSU, but still prefer Windows for DayZ Original (server) since it’s not available on Linux. Any advice would be greatly appreciated.
Reply
Reply
A
AmericanExpats
11-14-2016, 10:46 PM #1

This machine runs an HP Z440 CPU: Intel Xeon E5-1660 v3 (8C/16T). It’s not overclocked, just using regular RAM: 256GB ECC DDR4 at 2133 MHz. The GPU is a NVIDIA Quadro K2000, and it’s powered by an HP 700w PSU that came with the setup. I purchased an adapter to use a standard PSU when needed. The BIOS is Hewlett-Packard M60 v02.61, with SMBIOS 2.8 and Windows 10 Pro (64-bit). I’m using it to operate the AMP instance manager, which handles low-pop servers and a small 7D2D server. I also run CMD for Discord bots—one manages tickets, the other processes skin IDs for staff approval. I’m not using AMP in Docker containers. AIDA 64 Extreme is installed for monitoring via the sensor panel.

My problem is the system “crashes” but stays on. Specifically, I lose graphics card visibility and can’t access Windows Remote Desktop. The only way to restart it is to press the power button and then turn it back on. The biggest issue is unpredictable crashes—sometimes after hours, sometimes weeks. I’ve tried: running MEMS test, Intel Processor Diagnostics, and changed OS from Windows Server 2019 to Windows 10 thinking the OS was faulty.

Possible causes: CPU damage, BIOS fan settings, or overheating. At the time it started, I thought the CPU temperature should be around 74-84°C, but the max was only 66°C for a 140W unit. The PSU condition is unknown, and I lack the tools or funds for a proper check. I bought adapter cables to connect a standard PSU, but still prefer Windows for DayZ Original (server) since it’s not available on Linux. Any advice would be greatly appreciated.

Reply
Reply