Cache Hiearchy Error

[TempletonPeck]

In the past 24 hours my machine has frozen completely twice. Basically the last screen the computer is on is displayed and nothing works for input and I have to physically power off the machine.

In event viewer I see a bunch of these:

Reported by component: Processor Core
Error Source: Corrected Machine Check
Error Type: Cache Hierarchy Error
Processor ID: 2

The processor ID changes at random.

The error also seems to show up even without the freezing.

I'm running a first gen i7 860. It's about 4 years old now.

Is the processor dying, or is it something else? No hardware changes have been made in some time to the machine.

 

[Smoove910]

Is it overclocked? Stuff I've been reading suggests a core going out on you

 

[Matt1970]

I have never seen that one personally but the first few hits on Google seem to confirm what Smoove910 found.

 

[denis280]

did you get any Bsod

 

[TempletonPeck]

It's not overclocked. I'm trying to confirm if it's is the processor,that's what I'm thinking it is, but it's kinda hard to know for sure.

No BSOD either, it just freezes on whatever screen it's on and I have to physically power it off.

 

[Matt1970]

If it makes you feel any better it is really rare for a processor to go bad.

 

[denis280]

OK! Wat are all the system spec.and wat OS are you running

 

[TempletonPeck]

The PSU is a Corsair TX750.

It's a Intel i7 860 processor as mentioned, the motherboard is a Gigabyte GA-P55A-UD4P, the RAM is Corsair CMX4GX3M2A1600C9, video card is EVGA GTX 560 Ti and i'm running Windows 7 Professional 64-bit.

 

[TempletonPeck]

A bit of an update. I went to change the memory over to 1.5v DDR3-1333 and in the BIOS the following message appears:

"The system has experienced boot failures because of overclocking or changes of voltages

Last settings in this page may not coincide with current H/W settings".

I changed the profile over to 1.5v, restarted, still getting the same Cache Hiearchy Error in event viewer.

 

[denis280]

try it in Safe Mode.and If the computer remains stable.you will have to go to device manager.and start to disable some device to find out the faulty one.

 

[denis280]

O! and set your mobo bios to default

 

[TempletonPeck]

The cache hierarchy error does not seem to occur while in Safe Mode.

I was in it for about an hour and it never showed, meanwhile a normal boot into Windows and it's pretty much non-stop.

 

[denis280]

The cache hierarchy error does not seem to occur while in Safe Mode.

HA! it's not the cpu.Now you have to find out wat is the cause. in safe mode run CMD and type in chkdsk.or you could get seatool iso and boot from it.then memtest.try another gpu.and try unpluging dvd cdrom.

 

[TempletonPeck]

I have a SSD which comes up clean when I run the diagnostic tool on it. I already did a Memtest earlier today which also passed fine.

The voltage issue in the BIOS also still concerns me.

Another thing I'm noticing is that my data drive a Western Digital Caviar Black drive is failing SMART tests. It passes an extended test with the WD diagnostic tool though.

The Intel Processor Diagnostic tool also cleared my processor as fine as it passed all its tests.

Is it the PSU maybe? the motherboard?

 

[denis280]

Western Digital Caviar Black drive is failing SMART tests.

try with this drive disconnected

 

[Ketchup]

Which parts of the SMART test is it failing on?

 

[TempletonPeck]

I disconnected the hard drive, still getting the same Cache Hierarchy Errors non-stop. Doesn't seem to be the issue.

 

[Smoove910]

According to Newegg, your ram requires 1.65volts, not 1.5. Up your voltage to what is required and see if any changes:

http://www.newegg.com/Product/Produc...82E16820145260

 

[Steltek]

According to Newegg, your ram requires 1.65volts, not 1.5. Up your voltage to what is required and see if any changes:

http://www.newegg.com/Product/Produc...82E16820145260

Good catch - the Corsair product support site for those modules supports that as well. It also might be worth replacing your CMOS battery, resetting to defaults, then rekeying your BIOS settings.

Are you running any type of monitoring software you could use to check temps and voltages?

 

[TempletonPeck]

it runs at 1.65v for DDR3-1600 and 1.5v for DDR3-1333.

It defaults to the 1.5v.

 

[denis280]

it runs at 1.65v for DDR3-1600 and 1.5v for DDR3-1333.

It defaults to the 1.5v.

exactly.

 

[Steltek]

exactly.

What OP is saying is that running the modules at default SPD (i.e. 1333mhz) they are supposed to run at 1.5v. He hasn't been running the modules as DDR3-1600 but rather as DDR3-1333 (which requires 1.5v).

However, OP have you tried to run them at 1600mhz with manual voltage at 1.65v and manual timings set at 9-9-9-24?

http://www.corsair.com/us/cmx4gx3m2a1600c9.html

Since the processor seems to check out, I'm thinking you have either a PSU or motherboard problem. Were it my system, I'd swap the power supply before going any further to see if it resolves the issue. If it isn't the PSU and is the motherboard, replacing a socket 1156 board isn't going to be easy or cheap - you'd pretty much be looking at eBay and it would probably be more cost effective long term to move to a last gen socket LGA1155 MB/CPU or 1150 Haswell MB/CPU.

Out of curiosity, what happens if you boot the system from a Linux live CD? Does it work properly, or does the system crash?

 

[TempletonPeck]

I was running them at 1.65v previously, I just recently switched it back to 1.5v to test to see if it fixed my problem.

I'm thinking PSU or motherboard as well. I've actually ordered a new PSU just today, the only thing I could really do as I don't have a spare around to test.

Basically if that doesn't fix the issue, i'm going for a semi-rebuild then with a new CPU and MOBO. Like you mentioned good luck finding a LGA1156 board outside of eBay and then it's probably just not cost effective at that point and i'll end up with something that might be questionable quality.

The system basically crashes every 10 minutes at this point or thereabouts. I've pulled the PSU out now and will see how the replacement goes.

 

[Ketchup]

I was running them at 1.65v previously, I just recently switched it back to 1.5v to test to see if it fixed my problem.

I'm thinking PSU or motherboard as well. I've actually ordered a new PSU just today, the only thing I could really do as I don't have a spare around to test.

Basically if that doesn't fix the issue, i'm going for a semi-rebuild then with a new CPU and MOBO. Like you mentioned good luck finding a LGA1156 board outside of eBay and then it's probably just not cost effective at that point and i'll end up with something that might be questionable quality.

The system basically crashes every 10 minutes at this point or thereabouts. I've pulled the PSU out now and will see how the replacement goes.

Well, even if that doesn't fix the problem (and I hope it does) you will have a spare PSU on your hands, which is a good thing to have IMO.

 

[TempletonPeck]

So I put the new PSU in and... same problem. Time for a new motherboard and processor it looks like. Just my luck.

Thanks for all the help/advice guys, appreciate it.