LabVIEW
windows xp crash bccode 1000008e
Hello everyone,
guess what happens on a production line PC: Windows crashes.
There's Win XP and LV 2010 sp1
Dump info reports BCCode 1000008e BCP1 10000005 BCP2 8053A720 and some other infos
Of course googling that yelds little infos, such as check your RAM, change your video card, ecc...
The PC+LabView + my application has run without any crash for some months.
Then what I guess is that a windows automatic update changed some driver or some "who knows what" ?
Can you point me to some meaningful check or action to carry out ?
Thanks
@Norbert_B wrote:
The information you provide is insufficient.
What is the OS process throwing the BSOD (blue screen of death)?
What are you doing in this process?
Is it possible that the crash occurs only with certain combination of processes?
What is CPU load, RAM usage and other system parameters short before the crash occurs?
Which devices are present in the system (MAX report)? Which devices are you using during the crash?
Production line:
What is the noise level of power supply? Are there shocks/vibrations? Humidity? Temperature? Are you using a PC, IPC or PXI?
Have you checked cooling capabilities of the system (e.g. fans are defective, filters stuck, ...)?
These are the very basic questions/parameters you have to look into.... there are of course still many more.
Norbert
The OS process is ntoskrnl.dll called by ser2pl.sys which is a driver for a USB-RS232 bridge for an external hardware.
The MInidump files reports all the same.
In that process I am sending and receiving few data from an external device (homemade and not 100% reliable).
Many time the moment it crashes makes no sens at all. But it's surely linked to the use of the USB-RS232 bridge.
Sometimes I've seen the whole Windows crashing when I started the Windows watch (Date and Time) at the right bottom corner !!!
CPU, RAM load pretty low. CPU <15% RAM 0.5GB over 4GB
Attached there's the max_report.zip
During the crash basically anyhthing is used, unfortunatly there's not a unique point where the crash occours.
Noise of power supply ? Are you talking about the main 220V Hz (I'm in Europe) ? N.A. there are many others PCs running in that dept.
Shocks/vibrations: None at all.
Humidity/ Temp: it's an A/C controlled environment for PCB assembling, I think it's 26°C temp and about 50% relative humidity (anyway, if those are sensible parameters, than something is really wrong 🙂 ).
It's desktop PC 2core intel 3.0GHz, a pretty commercial household type device.
Fans are running and internal PC temps seems ok.
Thanks for your attention, hope all this can give some insights.
Davide
@Norbert_B wrote:
Davide,
so i understand that the crash is not related to a LV base application. Is that correct?
Since the crash refers to the ntoskrnl.dll (NT based OS kernel, a system component) i would approach the vendor of the USB-RS232 bridge for support. It seems to me that their driver is creating access violations into the address space of the OS or otherwise corrupting memory of the OS.
Are you sure that this ser2pl.sys driver version is compatibel to Windows XP? I've seen in Google that there are vendors supplying it for Vista based Windows only (Vista, Win 7, Win 8).
Norbert
This is the related page.
http://www.prolific.com.tw/US/ShowProduct.aspx?p_id=225&pcid=41
The vendor is Prolific and the chip is PL2303.
Unfotunatly the outside RS232 lines can see very dirty signals e.g.: RX always at ground, fast spikes, a square wave, etc.. because it is a OR wired line which is shared with a PWM input control. The external device supply may go off and on during the test.
It's an application that gave me endless headaches and this is only the last. 
It seems that some signal pattern (maybe random) can lock the PL2303 and therefore lock the USB controller.
I will try to: install an addition USB controller on the PCI bus, and in theis way the PL2303 will be alone on its USB controller. Now the PL2303 shares the USB controller with mouse and keyboard, and with the NI DAQmx 9178 (USB controlled). Even sometimes the USB pen drive.
I hope that a separated USB controller will help.
