Faulty Motherboard?

[WillMcDade]

Hey! I'm new to Gentoo and the Forums!

After a while of having my HP dx5150, that I bought for £30 on eBay, and doing quite a few upgrades, the system was working fine. But recently I decided to buy new RAM (DDR-400, Non-ECC, unbuffered, PC3200R). I went onto Crucial, as I wanted somewhere dependable. (I managed to get a £3 8GB Flash Drive aswell ) I installed the modules, booted the system, all was fine...

After a while I ended up with crashes all over the place, and sometimes resulting in a freeze at POST, only showing the HP logo on a blue background. I couldn't even access the BIOS. I managed to find a problem with my PATA hard drive, which I then took out and reinstalled Arch, as that was the distro I was using then. I then found that my SATA hard drive was filled with bad sectors today, but then, both the PATA and the SATA 250GB hard drives were both quite old. I then installed a 160GB hard drive, which I am just in the process of installing Gentoo on, BUT my system ONLY boots with 2 modules out of 4 installed. The memtest on the modules passed, so it is NOT the modules. When I do try and boot all 4 modules, the system fails to boot, resulting in 5 beeps, with corresponding 5 flashes, with a 2 second pause, and after the beeps, which only happen twice round, only the flashes remain.

The system is:

HP dx5150 Microtower
2GB RAM (the whole 4GB cant be inserted at once?!)
160GB SATA Seagate Momentus 5400RPM
MSi GeForce GT N210 with 1GB dedicated memory
AMD Athlon 64 X2 @ 2.0GHz, 64 bit enabled.



Is my motherboard faulty? If so, I will sell all the parts that work, and build a PC with the money left over (also a bit more money on top of that )

Thanks,

Will
_________________
The Penguin has arrived, and he will not leave your hard drive.

[creaker]

[WillMcDade]

The modules are the the type that are recommended by HP for this system, being the only type it takes. The beeps, from the manual, say:

-Reinsert the DIMMs to make sure they are inserted correctly
-Replace the DIMMs (I did this)
-Check the DIMMs for errors
-Replace the system board

I have done all of these and I am guessing it is the motherboard, so I guess the machine is on its way out.

EDIT: The Gentoo doesn't freeze as it did with the other 2 hard drives, so I managed to get Gentoo installed on this machine for the first time I will also try booting with all 4 modules inserted to see if it was just a faulty hard drive.
_________________
The Penguin has arrived, and he will not leave your hard drive.

[creaker]

Reinsertion in most cases doesn't helps. I would suggest to:
1. clean DIMM's contacts with eraser.
2. remove all the DIMMs and insert old DIMMs into the slots that was occupied by new DIMMs. Don't use new DIMMs, It will give some info about slots, whether these slots OK or not. If you can't boot this means mobo fault.
3. If you can boot successfully, turn the box off, remove old DIMMs and insert new DIMMs into the slots that initially was occupied by old DIMMs and try to start. If you can't start, it is DIMMs fault.

[WillMcDade]

[NeddySeagoon]

WillMcDade,

Welcome to Gentoo

How did you detect the bad sectors on your HDD?

memtest tests a lot mere than your RAM.
Fails every cycle with the same data errors at the same addresses probably indicate RAM.
If you run with tow sticks and see some of these, swap the sticks over and see if the failing address changes.
If so, its probably the RAM, if not, its something slot related.

Your motherboard will try to use tho RAM sticks in pairs, probably indicated by coloured sockets but the CPU can use an odd number of RAM sticks.
Its possible that only one socket is affected or the same socket (address range) in each pair.

A long time ago there was a kernel patch called badram so the kernel would avoid allocating bad areas of RAM.

As your motherboard is quite old, it could be suffering from failing capacitors in the on board voltage regulator.
You need to eyeball them. Domed tops, gel leaking out the bottom, rubber bungs bulging out of the bottom are all bad signs.
You need soldering skills to replace them.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.

[WillMcDade]

Thanks for the reply, NeddySeagoon!

I have been told about the capacitors on the board, and I did check them. They all look ok to me

2 slots are black (the first 2) and the other 2 slots are blue (the last 2)

I have tried swapping RAM in ALL different directions, and the RAM I had before. Ran memtest in all the configurations... It turned out fine. But I think because of the system's age, I think it might just be ready to get replaced.

I would like to know if there was some cheap temporary fix because I like the system and would like to keep it going as long as I can (I'm a kind of guy who likes to max out on things to keep them going for as long as I can )
_________________
The Penguin has arrived, and he will not leave your hard drive.

[Anon-E-moose]

How old is the motherboard? The power supply? And are you overclocking at all?
_________________
PRIME x570-pro, 3700x, 6.1 zen kernel
gcc 13, profile 17.0 (custom bare multilib), openrc, wayland

[NeddySeagoon]

WillMcDade,

Post a photo of the area of the motherboard around the CPU.
You can try underclocking the CPU. That reduces stresses in all sorts of ways.

There is a more subtle RAM issue too. Not all RAM chips are the same. They have different geometerys and speeds.
A few brain dead BIOSes only check the first RAM found and assume all the rest are the same.
This leads to problems when the other sticks are not the same as the first.
Are all the part numbers on the RAM chips the same?

Again, a photo that shows the writing works.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.

[WillMcDade]