What is the best way to identify what is causing a PC to hang

[gandalf72]

I have just left it running currently on pass 2 with top section showing pass 46% test 10%
Test #6

I am not sure on how to configure this application. but will leave it running over night.

Cheers for info Tottaly Random much appreciated.

[gandalf72]

Morning all,

The system has been running the memtest 486+ v4.10 now for 13 hours and it has not produced any errors. Its current status is 13 passes with non errors found. I will give it another hour of testing and then shut it off.

Any other suggestions welcome.

[gandalf72]

Sorry for delay in reply.

System has been up for 19hrs and 25 mins 19 passes with no errors.

I am going to switch it off now and let it cool down as it has been on constant now for past 3-4 days.

I dont have any spare mouse or keyboard to try unfortunetly.

I have disabled all powersaving that I can find and there is no screensaver or hibernate.

I will try the different ports tomorrow to give the system time to cool down before plugging it back in.

Conclusion so far:
I think we can rule out hard drive due to hard disk sentinal not finding any errors on short or long test.

I think we can rule out memory due to 19hrs 25 mins now and it not showing any errors on it.

When system has hung I have gone into bios and temp has been in the 40 degrees so dont think over heat is the issue.

bad driver potentially - doubt its driver related as it would blue screen and it does not do this.

This leaves me thinking either board or psu.

The bizarre thing for me is that I have really thrown everything but the kitchen sink at it and its still standing tall but the system has since I got it back hung no fewer than about 20 times (conservative estimate).

Many thanks for the help.

[DejaVu]

If it was the mainboard, you'd have alsorts of problems with it and it would not be on for 19 hours straight.

PSU needs replacing mate. I reckon it's a cheapo they've put in it and it's struggling to cope.

[gandalf72]

Yes more thinking that myself Deja Vu,

Is there any way that anyone can think of that will get the psu to trip and make the system impact - By this I am not meaning something bad I am just meaning if I run an extensive game or something like that or is it a case that the system would need to be opened and the psu tested?

Any diagnostic PSU software that can tell how it is running etc.

Thanks in advance.

[gandalf72]

Sorry forgot to add this in an earlier post.


There are 8 event id's of 41 that are all critical. Please note that around about the 21st I formated the drive and done a fresh install of ultimate windows 7 which was the second time I had done that in May.

There are no other critical event id's

Other warnings event ID's

Hard Disk Controller error Event ID 11, on dates : 22 (*4),23 (*2) May, 2nd June

ATIKMDAG display not active event id 432029 - I resolved that issue happened on 21st when I had just rebuild operating system.

ATIKMDAG CPLIB:: General Invalid Parameter event id 52236 - see above.

Event ID 36 the shadow copies of voume c: were aborted because the shadow copy storage could not grow due to a user imposed limit -Dates -: 28th May 3rd, 5th June

Previous shutdown unexpected event id 6008 - Dates :23, 24 (*2),25, 28 May,1,2, 3 June

The machine-default permission settings do not grant local activation permission for the com servier application with clsid event id 10016 Dates 21 (*3) May

The server did not register with DCOM within the required timeout event id 10010 - Date: 5th June

These are the critical and errors in the event viewer there are further warnings but the above is the main jist.

[Over Carl]

Have you tried hammering your cpu with prime95? I've used it when overclocking - sometimes I've had pc's that seem 100%, but this finds problems I often fixed by increasing power which is common when overclocking.

[DejaVu]

Failed to mention the Kernel-Power statements you found there gandalf72.

I think they are the ones that making it suspect, especially being linked to the Hard Drive Controller.

The rest, there are reasons for them and a minor. Everyone will have DCOM errors.

[gandalf72]

Over Carl, not done but am currently running this on the pc as we speek. Will see how it gets on.

Deja Vu - I am in agreement there seems to be too much of a coincedence between the two. I just need to ensure that Mesh think the same - They are a master at not believing unless they can see with their own eyes and it cant be denied that its not faulty etc.

Thanks again for all the help it is very much appreciated!!

[gandalf72]

Houston we may have success - Prime 95 crashed system at 8:43 I just noticed it now.

Not sure what that indicates though but first hang in about 3 or 4 days.

I had selected the in-place large ffts option numer of tortue tests to run 4.

Bad news though is that event viewer only shows the 2 generic messages 1 critical Kernal - Power and the second in warning saying about invalid shutdown. I have 3 logs belonging to prime on the desktop but they were saved pre crash going by the log time of pc stoping.

[gandalf72]

Morning all,

Update for you. I wanted to check that the hang last night was not a fluke so re ran the prime95 using same settings and I have just checked the pc and its still working.

So I am guessing from running that test from 10:16 till 8:47 that rules out those components?

[gandalf72]

Hi All,

just to let you all know that I contacted Mesh and arranged for Friday uplift of my system to get checked out.

Brilliant call with their customer care lady but their first chap I spoke with in the repair department well......... lets just say his first troubleshooting step was to tell me to turn off the computer and turn on, then to tell him if issue happened despite me telling him that it was an intermittant fault.

Next step after that was to remove wireless devices and see if that made any difference.

Next step turn off router and back on again and see if that resolved the issue!!!

All I can say is thank god for the lady I spoke too

I will keep everyone up to date with what is happening and thanks once again for all your help.

[TotallyRandom]

nice one hope they sort it this time! Stick a letter in the box telling them what you have done and tell them what parts you want replaced or you will be speaking too trading standards. So they do replace something

[gandalf72]

Thanks everyone for alll your help. This is the letter I have been composing for past wee while. This will be going in the box along with the pc. If there is anything else you think I should add or remove let me know.

I have edited some bits just in case they take offence if they browse this forum. Looking back at the help and offers of help you have given I am very greatful you have helped me and if ever I can help you let me know.

Respect to you all!!!



My Address is:
My phone number is:
RMA:
Serial Number:
Windows Ultimate 7 Licence Key:
Windows Ultimate 7 Image Location:
Image downloaded directly from Microsoft web site.


Given the nature of the problem being intermittent I would like to ask you phone me please prior to any work being done to discuss what is on this page and to offer you any assistance and log files you may wish to have. The system has no pattern for this hang and can go many days with heavy disk activity and music on and not hang other times it will hang once a day on consecutive days. It is very rare for the system to hang twice on a day.

I think that the power supply unit is not powerful enough to support the power required from the system so when the power usage gets too high this is when it hangs or is faulty. This links in to the errors being logged in event viewer re power-kernel. I would like to have that power unit replaced please.

When the system came back from repair I left it running for 1 week. It hung within 4 days with it being left on and no software installed. This was reported to x at customer services. Since that point the PC has constantly crashed every week either once or multiple timesa day. This is not dependant on operating system – power issue.

As I am sure you can see I have been working very hard to do diagnostic to aid Mesh but am unable to get a pattern of when this happens. I have numerous perfmon report files that I can send if this makes life any easier for you.

I have done the following diagnostics since the PC was returned to me:
· Please note that it is not dependant on just windows open and no other application it can be with just windows or with multiple applications. NO CLEAR PATTERN FOR THIS ISSUE
· Dual booted with Vista and Windows 7 32 bit
· Formatted the drive and installed Windows 7 twice
· Formatted the drive and installed Vista once
· Ran a program called hard disk sentinel – passed short and long test with no indication of any errors
· I have ensured that all drivers are up to date – pc returned to me with vista without latest service pack installed or dvd rom and bios firmware updated.
· Ran the windows 7 memory check and it passed
· Ran the windows 7 repair option
· Ran memtest 486 486 v4.10 for 19hrs and 26 mins and it passed.
· Ran a program called Prime95 it hung after 45 mins of running that first time but then I shut down pc and started test again and after 10 hours of further tests it did not hang. Test option on this was (in large ffts option number of torture tests to run 4)
· I also transferred 159GB of data from local hard drive to external hard drive and system did not hang.
· Looking at the critical event dates and then having a look at the error dates there seems to be a link between :

Hard Disk Controller error Event ID 11, on dates : 22 (*4),23 (*2) May, 2nd June

Previous shutdown unexpected event id 6008 - Dates :23, 24 (*2),25, 28 May,1,2, 3 June

Just seems it may be too much of a bit of a coincidence that the hard drive controller and shutdown at similar times.
· On 3 occasions after it has hung it has black screened prior to bios splash screen. This has been when I have held the power button down for 10 seconds and then immediately after it shuts off I press power button again.
· Please also note that the memory card that is built into the pc is detected in disk management HOWEVER IT IS NOT DETECTED IN WINDOWS EXPLORER and I think this may be relating to those controller errors?

Fault:
This fault is the exact same fault that was with the system when I first raised a ticket back in December. The system will intermittently hang with either no program running or with multiple windows open.Also windows explorer will not display the memory card reader.

· Intermittent hang – No response from keyboard, caps lock, num lock when it hangs.
· No blue screen when it hangs
· No error message when it hangs
· No memory dump file when it hangs.
· If listening to music when it hangs it will distort the sound then hang
· If no music on it will hang with no indication that it has hung till you try to move mouse and it won’t move.
· System could hang on a Tuesday then not hang again for 4 days then hang the day after.
· Not dependant on software being installed.
· Not dependant on operating system will hang in Vista and Windows 7
· Can hang with no programs running other than operating system or will hang with applications opened.
· There are 8 critical event ids and 41 errors. Please note that around about the 21st I formatted the drive and done a fresh install of ultimate windows 7 which was the second time I had done that in May.
Event Log
Event logs show that generic shutdown message with no memory dump file being created.

Performance log
Shows that the system has been shutting down at random time periods incorrectly as I had to hold the power button down for 10 seconds to get it too boot back up.
Recently the system has taken to not boot up when it has hung first time resulting in just a black screen and me having to turn the system back on and off again. This though has only happened three times.

Device Manager
There is nothing in Device manager showing any errors or exclamation marks.

External Devices

There is a Logitech Bluetooth dongle for a MX550 keyboard and mouse combo. There is also a Linksys usb dongle for connecting to the internet (wusb100 I think) . The dongle has only recently been purchased and the pc had the hang issues prior to the device being attached. The Logitech keyboard is also a fairly new introduction and issue also happened pre install of the keyboard. I checked both are windows 7 compatible. Issue happens in vista as well as windows 7. Just to confirm both devices are confirmed by Microsoft to work under windows 7. Just to confirm previous mouse and keyboard and internet dongle attached still had the same result with PC hanging.

There is a wireless printer also and occasionally I have use of an external powered Seagate hard drive again issue happened pre and after they are used.


What I think

I believe that either the PSU is not able to handle the power being drawn from HDD, GPU, PSUor is not powerful enough to handle demands being placed on it. I would like this PSU replaced and if possible with an appropriate PSU wattage that will support the power.

System Spec

AMD Phenom™ II X4 920 Quad Core Processor AM2+ (2.80GHz, 8MB Cache)
ASUS M4A79T Deluxe Mainboard - AMD 790FX CrossFireX- AM3 Phenom™ II, DDR3 - ATX
4096MB 800MHz Dual Channel DDR2 SDRAM - (2x 2GB)
1TB (1x 1000GB) Serial ATA 2 Hard Drive with 32MB Buffer
Samsung 22x Dual Layer DVD Writer Super Format +R/-R/RW/RAM
1GB ATI HD4670 Graphics accelerator - DVI, DirectX 10.1 PCI Express 2
7.1 High Definition onboard sound card - for 8 Channel Cinema sound
12 USB 2.0 ports (12 USB 2.0 ports (6 on back I/O, 6 on board) - ASUS M2N68-VM
Network Ready Gigabit LAN (10/100/1000)
I added a wireless linksys wusb100 usb adapter
I dont know about the psu it came with the system case
Midi Tower XGS Xtreme Gaming Chassis - Black / Black Mesh + 550W PSU

[Over Carl]

Houston we may have success - Prime 95 crashed system at 8:43 I just noticed it now.

Not sure what that indicates though but first hang in about 3 or 4 days.

I had selected the in-place large ffts option numer of tortue tests to run 4.

Bad news though is that event viewer only shows the 2 generic messages 1 critical Kernal - Power and the second in warning saying about invalid shutdown. I have 3 logs belonging to prime on the desktop but they were saved pre crash going by the log time of pc stoping.

Sorry for my brief post earlier. Prime95 I've found is a very very demanding stability tester, but it doesn't help much with diagnosis. You can tweak the tests to put more strain on the cpu or memory, if you select small fft sizes like 8kb/8kb that seriously batters your processor, and I think running fft's in place also helped bring up errors. If you have time / can be arsed, I would try running p95 with the following settings:

Custom: 4 threads, 8kb/8kb, fft's in place, 15 mins - kills cpu/fpu
Custom: 4 threads, 2048kb/4096kb, fft's not in place, 15 mins - kills ram/chipset
Small FFT's
In-place large FFT's.

Before you start doing this, install speedfan and get it logging your temps and voltages to file. If any of the above tests generate errors, check event viewer and the speedfan logs to see if you can find anything interesting, then shutdown pc, start it after 30 secs and see if you can recreate the fault. Well, you may need to actually check after rebooting if it totally dies on you.

If possible, I like to run p95 at least 24hrs on any particular test, and once I think I've eliminated a particular problem, if possible I will leave it running a full 7 days on the test it previously would fail on.

If you have the resources and really want to hammer your system, pop in a spare hard drive and install a 64bit copy of windows, then run p95x64.

[DejaVu]

Please also note that the memory card that is built into the pc is detected in disk management HOWEVER IT IS NOT DETECTED IN WINDOWS EXPLORER and I think this may be relating to those controller errors?

Some systems, mine included, hide these drives in Windows Explorer until they are used.
Tools > Folder Options > View
Untick - Hide Empty Drives in the Computer folder

Should make it reappear.

Other than that I think it's a great letter and you should get the result you ask for. Hopefully, Mesh stay amicable about it and get it sorted.

[gandalf72]

Thanks Deja Vu,

That indeed was the problem with the drives, I had not seen that option before. Just in the door and my new external hard drive has appeared. I tried to transfer 310gb of files in one shot to the external and it crashed after about 30 mins of transfer.

Is there any size limits in windows 7 about the amount of data you can transfer at anyone point in time?

It was same generic shutdown message that I recieved re power and system shutdown unexpected.

[EvilBoB]

Thanks Deja Vu,

That indeed was the problem with the drives, I had not seen that option before. Just in the door and my new external hard drive has appeared. I tried to transfer 310gb of files in one shot to the external and it crashed after about 30 mins of transfer.

Is there any size limits in windows 7 about the amount of data you can transfer at anyone point in time?

It was same generic shutdown message that I recieved re power and system shutdown unexpected.

Sounds strange. I have had copy overs fail in the past due to the drives overheating (Transferring between cheap drives that got too hot with all the furious read/write operations.

[DejaVu]

I have 4TB drives connected to my Windows 7 Acer Revo and it's always transferring incredible amounts of files about. The largest I've transferred in one hit was 450Gb (all approximately 5Gb MKV films) and it had no problems.

[gandalf72]

Quick update :

Got a phone call from the engineer on Friday and he has tested it and found no fault with the system. I explained about the various tests run and had them all printed out and attached to system but he didnt seem interested unless he could find fault then he would do nothing.

Unfortuneltly this was the scenario that I feared the most, I needed to prove which part was likely to be faulty and prove that beyond any shadow of a doubt but the engineer said he has run for 2 days and it has behaved itself. I tried to explain about the essance of it being so intermitant and even to check out their own forums where I posted the information but he didnt seem that interested in that.

He is going to check it today but looks like its going to behave and he is not going to replace anything without the system hanging. I asked him what does he think would cause the system to hang without leaving dump files or blue screens but he seemed to want to put it down to my software rather than hardware.

I just wanted to say thanks to everyone who has very kindly helped me and will keep you all up to date if anything changes but looks likely now they will charge me for delivery and for not finding any issues.