I’m having a major issue with my Gigabyte MZ72-HB0 Rev 1.0 motherboard and would really appreciate any help or insight.
Short Background:
I bought this motherboard used from another seller and built a workstation around it. Everything worked fine initially — it powered on, video output was working.
I left the system running (mostly idle) for a few weeks, with little activity due to busy workdays. Later, I started using it lightly (30–40% CPU load) (Dual 2x EPYC 7742), and one day when I returned home from work, I found the screen completely black. Since then, the system has failed to boot or show any signs of life beyond fan spin.
Current Problem:
The system powers on (fans spin, standby power active), but:
No video output
No serial output over COM1
No IPMI access via the MGMT port
No POST beeps
What I’ve Tried So Far: 1. Serial Console (COM1)
USB-to-serial with proper null modem cable (confirmed working via loopback test)
Tried PuTTY and Tera Term at 115200 / 8N1 / no flow control
No output from the board at any point during boot
2. IPMI/MGMT Port
Connected directly to MGMT port with static IP (192.168.1.100)
Tried known default IPs: 192.168.1.1, 192.168.1.10, etc.
Scanned network using Advanced IP Scanner — nothing detected
Also tried direct PC-to-MGMT cable with static routes — still nothing
3. Cleared CMOS
Used the JCMOS1 jumper
Held short for 10+ seconds with power removed
No change in behavior
Questions:
Is the motherboard likely already bricked or the BMC flash corrupted?
Has anyone here recovered a Gigabyte server board like this using the jumpers?
Is there any known way to force output or use a hidden debug header?
Any advice or experiences would be really appreciated.
If anyone here can help me figure out what’s going on and actually get it working, I’m more than happy to pay for your time and assistance. I’ve already spent over $1,000 on this and really don’t want to drop another grand on a new motherboard.
I don’t think so Wendell, as it would power up when supplied via the PSU.
@Trizocy you may be able to force a PSU to apply power by shorting the appropriate pins. If that works, something’s wrong with the board (cut trace perhaps) that prevents the PSU from switching on. In which case you either have to source a new board or frankenstein a solution with a soldering iron and some wire. Not recommended for newbees though.
The few times I have had a motherboard flat out refuse to respond, but ended up not being dead, I left it without the battery or power for several hours. Even when doing the same exact thing for several minutes didn’t work. I had exhausted all other hopes and somehow this has worked a few times.
If that doesn’t work, heating expansion and contration may have caused a connection issue somewhere. You only mention clearing CMOS on the physical side of things. It may need everything reseated and attempt to boot with the bare minimum hardware. It might be worth looking at the CPU’s for signs of heat discoloration or burning, although a simple reseat in the socket might be all it needs.
Didn’t @wendell just mention in one of his videos that some motherboards/systems won’t power up at all if the CMOS battery is dead? I’d try replacing it, as that is an easy and inexpensive troubleshooting method. The other thing I’d try is seeing of the blinking pattern you’re seeing on the motherboard LEDs corresponds to any troubleshooting codes in the motherboard manual. A few years back I somehow managed to fry the BMC chip on my asrock Rack board and that killed the board dead until I could get a replacement BMC from Asrock. But if the blinking BMC light means that the BMC is functioning, you should at least be able to connect via the IP as @ozlay mentioned.
loads of boards won’t power up with a dead cmos battery. just the other day my beloved evga sriii dark I thought has died. nope just dead cmos battery.
I was going to suggest to jumper the CMOS but I see you have already attempted this potential fix. I was also going to suggest to swap out the battery with a fresh one but Wendell beat me to that one. If you have the time to try this it might be worth a shot:
If you haven’t already done so, completely remove the board from the case and all peripherals. Place it on a box (or a test bench if you have one). Make sure the power supply is completely disconnected. Now do the fresh battery thing, and jumper the CMOS. Leave it jumpered for at least 15 minutes. Do NOT plug any power to the board during this time. ZERO. Nada juice. After the 15 minutes is up connect PSU to mainboard and your CPU juice and PCIe. After this move your jumper back. Make sure PSU is putting power to the board at this point. Try the power switch.
I know this method may not seem any different but it has worked for me at least three times with older, stubborn system boards. I’m guessing perhaps the moving of the board changes the flex in the board and reconnects a weak contact. I really don’t know why it works but it just does — some times.
As a last resort you might also try flashing the BIOS if it will let you.