BSOD Memory_management

[Bill_Bright]

Wow. I don't see where you stated the brand and model of this cooler - so we can avoid it! No doubt, ensuring the mounting mechanism is securely fastened is important to ensure sufficient and even pressure is applied atop the CPU (and socket). But it seems some technology to prevent over tightening (and/or stripping of the screw hole too) would be easy to implement.

 

[IceScreamer]

Wow. I don't see where you stated the brand and model of this cooler - so we can avoid it! No doubt, ensuring the mounting mechanism is securely fastened is important to ensure sufficient and even pressure is applied atop the CPU (and socket). But it seems some technology to prevent over tightening (and/or stripping of the screw hole too) would be easy to implement.

I figured since it's in my system specs I did not need to explicitly say it. The cooler itself is rather good, but yea, the mounting system is leaving a lot to be desired.

 

[Bill_Bright]

I figured since it's in my system specs I did not need to explicitly say it.

Sorry - I don't normally pay attention to those system specs since many people have more than one computer, or are helping someone else with their computers.

In looking at the picture now, my worries are relieved because I see the ring you mention mounts to the motherboard with 1/4 turn screws. So it is not likely uneven or excess pressure will be applied to the motherboard, processor or socket.

 

[IceScreamer]

Sorry - I don't normally pay attention to those system specs since many people have more than one computer, or are helping someone else with their computers.

In looking at the picture now, my worries are relieved because I see the ring you mention mounts to the motherboard with 1/4 turn screws. So it is not likely uneven or excess pressure will be applied to the motherboard, processor or socket.

Actually, those are not screws. The way this mounting system works is as follows:
- you place the plastic ring on the motherboard and push through the hollow plastic sockets in the motherboard holes
- after that, you insert and push 4 pins through the plastic sockets on the mounting ring, securing the ring in place
- you mount the cooler and place the metal cooler clips over the plastic notches on the plastic ring that is mounted on the board
- this is the step that I might have messed up, after mounting the clips you tighten 2 phillips screws on each side of the cooler, tightening the cooler in the process, from what I've seen there is no limit on how much you can tighten them (apart from the thread length)

My guess, as @eidairaman1 pointed out, is that the board might have warped enough and contact was made where it shouldn't have.

In any case, it appears that the problem is solved and I learned my lesson, tight enough.

 

[eidairaman1]

Wow. I don't see where you stated the brand and model of this cooler - so we can avoid it! No doubt, ensuring the mounting mechanism is securely fastened is important to ensure sufficient and even pressure is applied atop the CPU (and socket). But it seems some technology to prevent over tightening (and/or stripping of the screw hole too) would be easy to implement.

Bill you could say the same about my cooler too, when I first put it on I warped my board initially but before putting in the case I backed it off to unwarp the board, im so glad i did not crack the socket, contacts, grounds, traces or broke some smds when I did that.

 

[IceScreamer]

Scratch everything and anything I said, it happened again.

I'll just take it outside and burn it.

 

[er557]

i would try to trade in each hardware piece in that pc, building another from scratch, as the only route to give you piece of mind

 

[IceScreamer]

i would try to trade in each hardware piece in that pc, building another from scratch, as the only route to give you piece of mind

I would, gladly, but not knowing what component is faulty is like selling mines wrapped as candy.

I really don't know what I'll do, I guess I'll just splurge on a budget laptop and be done with it.

 

[er557]

did you google web threads with this error and possible workarounds? it could be a faulty driver

 

[IceScreamer]

did you google web threads with this error and possible workarounds? it could be a faulty driver

It was the first thing I did, all of them point to the kernel and that means that the driver is not recognized. It was supposed to be something to do with tcpipsys but it did nothing. All the codes I googled return nothing, like no results at all.

 

[er557]

what about trying windows 19h2 skip ahead as a radical solution? worth a shot and sometimes more stable than retail, it should release in a few days

 

[IceScreamer]

what about trying windows 19h2 skip ahead as a radical solution? worth a shot and sometimes more stable than retail, it should release in a few days

Already did, on a new drive, also got a blue screen.

I did every single test that I could think of, new windows on a new drive, every single driver install/reinstall, HDD surface tests, PSU change, memtests with one/two sticks and in different slot combinations, driver verifier check, with and without modifiers, IDE to AHCI switch. I don't know if I missed something but this thing is really giving me a headache.

 

[er557]

any auto optimization/ subtle overclocking in bios disabled?

 

[IceScreamer]

any auto optimization/ subtle overclocking in bios disabled?

That too, CMOS reset, defaults, any and all startup programs removed.

 

[Regeneration]

What error codes in the BSODs are you getting?

 

[IceScreamer]

I already posted the dumps in the OP but this is the most common one:

On Wed 13.2.2019. 15:44:19 your computer crashed or a problem was reported
crash dump file: C:\Windows\MEMORY.DMP
This was probably caused by the following module: ntkrnlmp.exe (nt!PoUnregisterSystemState+0x22FD)
Bugcheck code: 0x1A (0x41790, 0xFFFFFA800051A6E0, 0xFFFF, 0x0)
Error: MEMORY_MANAGEMENT
Bug check description: This indicates that a severe memory management error occurred. A page table page has been corrupted.
This might be a case of memory corruption. This may be because of a hardware issue such as faulty RAM, overheating (thermal issue) or because of a buggy driver. This problem might also be caused because of overheating (thermal issue).
The crash took place in the Windows kernel. Possibly this problem is caused by another driver that cannot be identified at this time.

 

[Regeneration]

You might have some rogue device on the 3.3v rail causing interference. Some fan perhaps. Disable automatic fan speed.

It's possible your motherboard sometimes sets aggressive B2B/RTL/channel timing. Plug a USB/DVD with MemTest86, set it as 1st boot device, run for a minute and then reboot/shutdown. Repeat 20 times.

 

[Bill_Bright]

Bill you could say the same about my cooler too, when I first put it on I warped my board initially

Yeah, uneven stress is always applied during initial mounting. It is like tightening the lug nuts on a car wheel. You should slowly tighten in a criss-cross pattern and not round in a circle.

I was talking about the use of screws to mount to the motherboard. Typically there are 4 pins you push through the board, then do a 1/4 turn to lock them in place. Yes, until all 4 are locked place, uneven stress occurs. But it should not be enough to cause an over-tightening situation, or permanent damage. That's what I was getting at.

I have seen where mounting plates are attached with small nuts and bolts (not screws). And care must be taken to avoid uneven stress or over tightening. But I have never found that to be a problem. What I have done in the past is apply a small drop of Loctite Purple to lock the nut in place, keep it from vibrating loose over time, but not bond it so strongly you cannot easily break it loose when needed. Note the "purple" series is the lowest strength in terms of bonding strenght. You can go blue, but I do not recommend red for these tiny nuts as it may take too much force to break the bond.

Another friend/tech just uses a touch of hot-glue. It works.

 

[IceScreamer]

You might have some rogue device on the 3.3v rail causing interference. Some fan perhaps. Disable automatic fan speed.

It's possible your motherboard sometimes sets aggressive B2B/RTL. Plug a USB/DVD with MemTest86, set it as 1st boot device, run for a minute and then reboot/shutdown. Repeat it 20 times.

Hmm, now that you mention it, something power related could be it, and not due to the PSU (which I changed). I mean, the crashes started when I changed the cooler, and now it's clear it's not due to the possible board warp but maybe the fan or the auto-control of it?

I'll have some free time soon so I'll try these out.

 

[eidairaman1]

Yeah, uneven stress is always applied during initial mounting. It is like tightening the lug nuts on a car wheel. You should slowly tighten in a criss-cross pattern and not round in a circle.

I was talking about the use of screws to mount to the motherboard. Typically there are 4 pins you push through the board, then do a 1/4 turn to lock them in place. Yes, until all 4 are locked place, uneven stress occurs. But it should not be enough to cause an over-tightening situation, or permanent damage. That's what I was getting at.

I have seen where mounting plates are attached with small nuts and bolts (not screws). And care must be taken to avoid uneven stress or over tightening. But I have never found that to be a problem. What I have done in the past is apply a small drop of Loctite Purple to lock the nut in place, keep it from vibrating loose over time, but not bond it so strongly you cannot easily break it loose when needed. Note the "purple" series is the lowest strength in terms of bonding strenght. You can go blue, but I do not recommend red for these tiny nuts as it may take too much force to break the bond.

Another friend/tech just uses a touch of hot-glue. It works.

Bill i know about x pattern mounting, that didnt stop it from happening, i was overzealous.

 

[Bill_Bright]

i was overzealous.

LOL! Been there, done that!

 

[Zahran14]

I have same same problems with x570 master and ryzen 9 5950x - i did almost every thing you did and nothing worked also tried 4 different ram brands. I am returning the motherboard now and i will test my ram in my friend pc if it works i get a new motherboard asus or msi - not dealing with gigabyte