Usbport.sys bsod driving me nuts

Status
Not open for further replies.
Hello and welcome to Techspot.

1 minidump crashes at hidusb.sys. USB Miniport Driver for Input Devices. It has a bugcheck of 8E.

2 minidumps crash with memory corruption. Faulty ram? They also have bugchecks of 8E.

0x0000008E: KERNEL_MODE_EXCEPTION_NOT_HANDLED
(Click to consult the online MSDN article.)
A kernel mode program generated an exception which the error handler didn’t catch. These are nearly always hardware compatibility issues (which sometimes means a driver issue or a need for a BIOS upgrade).

You may have a ram problem.

Go HERE and follow the instructions.

Regards Howard :wave: :wave:
 
howard

not sure how you read that .dmp but glad someone knows how. even though ive run literally 3 different memory tester for hours it still indicates a ram problem. guess im confused at how my testing passes.

dont get me wrong i believe you just wondering if my testing the memory is inaccurate then how can one be confident when running memory testers ???

i will order another chip this time through 4allmemory.com have ordered quite a lot of modules from them with no problems. just hope they have the right ones for my ecs pm800-m2 board.

when i built this system the manual said micron with the model number was tested to function with this board. so i guess the module i was sent was bad from the get go.

thank you. i have another machine a clients of mine that keeps getting bsod's i will make a zip file as soon as shes out of the hospital for you to look at.
 
Not so fast. I only said you may have a ram problem.

It could also be caused by one of your usb devices.

You might also want to try updating your bios.

Did you follow all the instructions in the link I gave you?

Regards Howard :)
 
Howard

Okay. I only have 4 usb devices connected - canon i960 printer, epson perfection 1650 scanner, apc ups xs1000 and logitech web cam but its hardy used just hocked up. Problems were happening before the web cam was installed. Having always been taught that unless the bios updates point to a specific problem one is having that your better off not updating the bios. I have an email to the board manufacturer to see if I missed some thing in their fixes that might address this issue. so far no response from them but its only been a couple days for that.

Like I said have tested the ram with 3 different testers and literally for hours and that always comes back okay.

System is not overclocked. Only have 1 512mb module on the board so cant test by removing and the micron module was part of the manufacturers compatible list. Havnet tried putting into different slot - will give that a try.
have receded all connections and periodically blow the system out with canned air. I will update the nvidia drivers and replace the m.s. drivers. is onboard sound. have run chkdsk with the switches as well as other h.d. diagnostics including western digitals. all okay there.

im usually pretty good at this stuff. have been doing since the first pc hit the market back in 81. have a couple certification but not in reading dump logs. guess I missed that class. only 2 bsod that I havent been able to figure out 1 being my machine and the other a clients where I traced all errors down and replaced parts that were tested as being bad but her machine is still blue screening her. would like to send .dmp for your review also.

the bios on my machine is pm800-m2 v1.1b 5/25/04. I looked at the updated bios on ecs web site and I couldnt find anything close to the problem I am having. but if you suggest updating I will give a try.

sorry about the length of this reply and I do appreciate your assistance and knowledge. wish I knew how to run that debug program.

going to push my luck with you and attach a couple .dmp from this other machine I was mentioning. maybe you could take a quick look and give me your opinion. again thanks.
 

Attachments

  • jennifer.zip
    74.3 KB · Views: 13
1 minidump crashes with memory corruption. It has a bugcheck of 8E.

1 minidump crashes at win32k.sys. Windows Multi-User Win32 Driver. It has a bugcheck of 0A.

2 minidumps crash at ntkrnlmp.exe. NT Kernel & System file. They have a bugcheck of 0A.

Faulty ram is a distinct possibility.

If you would like to read dump files. Go HERE and follow this short tutorial.

Regards Howard :)
 
howard - so that second zip file indicates bad memory. its brand new from 4allmemory.com so it doesnt make sense. is it possible that the boards memory is bad thereby meaning the system board is defective ??? again with her system as ive done with mine all tests indicate all is well. the memory is new as is the hard drive. these errors are really confusing. they indicate possible memory damage but the module in this machine is like i said brand new. now i suppose its possible the memory was sent as bad but test didnt indicate that.....greg
 
Faulty ram can be. Faulty physical ram/video ram/L2 cache ram of the cpu/page file(virtual ram) etc.

Some faulty ram can pass Memtest86+.

Ram incompatibility can cause lot`s of problems. Check that the ram is compatible with the mobo. The mobo manufacturers website will have info on this.

Sometimes downclocking the ram in bios may help, as may running the system with just one stick of ram. Also, try putting the ram into a different dimm slot.

An unstable psu can also cause all manner of problems. Check that the psu is putting out enough power for the system.

This probably isn`t the problem, but it might be worth checking the mobo`s capacitors, for bulging, or leaking. See HERE for further info.

This may still be a driver issue, but the minidumps are inconsistent, and don`t point to anything in particular. that is why I thought it may be a ram problem, as that is exactly what faulty ram would cause. As would overheating/psu/overclocking etc.

Regards Howard :)
 
Howard - in my case updated nvidia drivers with nvidia not m.s. and 10 minutes later it bsod me with the same usport.sys 8e error so guess not the video drivers. virtual ram redoes itself every time the machine is restarted is that not true. so the drive space it uses should be refreshed - true not true ?

have done extensive testing on the board and cpu with pc certifies diags. with no errors. the memory i have on this system is based on the manufacturers compatible list. psu is a 450w enermax. dont think overheating i have 2 case fans inside the box and they are working just fine. also temps by feel are very cool plus i have the windows open blowing cold air inside the box to keep cool.

my usbport.sys is version 5.1.2600.2180 dated 3/31/03 microsoft.

whom do you recommend getting memory from ??? i have used 4allmemory.com and usually with great results.

this is very confusing and frustrating.
 
virtual ram redoes itself every time the machine is restarted is that not true. so the drive space it uses should be refreshed - true not true

I don`t think it does get refreshed. That`s why I recommend that you replace the page file with a brand new one.

I wish I could tell you exactly what the problem is, but I can`t.

It may well be that your memory is fine. If that is the case, then I`m scratching me head too.

Maybe you should send cpc2004 a pm, and see what he thinks.

He`s the real expert on these matters.

A second opinion could be helpful I feel.

Regards Howard :)
 
howard - i really do appreciate your assistance here and i will pm this other person you mention.

there are a lot of memory sellers out there - karlon, crucial, memoryx and all with slightly different prices. their analizers seems to be able to read my board and the memory i need do you have a certain organization you buy you memory from ???

also a dumb question on xp how do i replace the paging file ???
 
Personally I use Kingston ram, and have never had any problems.

However, since your ram is compatible with the mobo, and as far as you`re aware isn`t faulty. I don`t think it`s a matter of ram manufacturers.

To replace your page file.

Right click my computer, and select properties, and then the advanced tab.

Under performance, click settings. Click on the advanced tab, and then on the change button.

Click on the custom size button, and click in the initial size box and make the size of the page file 0. Click in the maximum size box, and make the size of the page file 0.

Click set.

Now reboot your machine. After your system has restarted run a defrag.

Once the defrag has finished, go back to the pagefile settings above, and click on the change button. For the initial, and maximum pagefile size make it 1.5 times the ammount of ram your have. I.E if you have 512mb of ram set the page file to 768mb.

Click on the set button again, and reboot your system. You now have a brand new page file.

Regards Howard :)
 
howard - unfortunately kingston not compatible with my boards manual. i also like kingston.

removed the page file, defrag's and reset page file. one more thing off the list to try. also, pm'd cpc2004 but no word back from him.

thanks again.

so running my memory testers for literally hours still isnt long enough. why would they not indicate a problem ???
 
Hi gghartman,

I believe that it is faulty memory (ie faulty ram or CPU).
Suggestion
1. Clean the dust within the memory slot. Reseat the RAM to another memory slot
2. Downclock the ram
3. Downclock the CPU
4. Make sure your PC is not overheat

Summary of dump analysis
Two minidumps are crashed with memory corruption.

Mini013006-01.dmp BugCheck 1000008E, {c0000005, 805783a5, f43bac04, 0}
DEFAULT_BUCKET_ID: CODE_CORRUPTION
CHKIMG_EXTENSION: !chkimg -lo 50 -db !nt
!chkimg -lo 50 -db !nt
4 errors : !nt (805783a6-805783be)
805783a0 ff 74 44 53 56 ff *85 04 85 c0 74 42 83 65 *83 00 .tDSV.....tB.e..
805783b0 ff 75 1c ff 75 18 *85 75 14 ff 75 10 50 56 *83 f7 .u..u..u..u.PV..

Owning Process 82db83f8 Image: TeaTimer.exe
Probably caused by : memory_corruption

--------------------------------------------------------------------------------------------
Mini013006-02.dmp BugCheck 1000008E, {c000001d, f7c6100e, f434c99c, 0}
DEFAULT_BUCKET_ID: CODE_CORRUPTION
CHKIMG_EXTENSION: !chkimg -lo 50 -db !USBPORT
!chkimg -lo 50 -db !USBPORT
8 errors : !USBPORT (f7c61007-f7c6103f)
f7c61000 f7 04 74 37 8d 48 0c *69 4d f4 c6 45 fc 71 c6 *20 ..t7.H.iM..E.q.
f7c61010 fd 75 c6 45 fe 54 c6 *00 ff 52 ff d3 8b 55 f4 *00 .u.E.T...R...U..
f7c61020 4a 04 23 c8 8b 45 fc *00 e1 04 03 4a 08 89 01 *00 J.#..E.....J....
f7c61030 45 08 89 79 04 89 71 *00 89 41 0c 83 3d dc 1d *00 E..y..q..A..=...

Owning Process 82c68960 Image: mainserv.exe
Probably caused by : memory_corruption

--------------------------------------------------------------------------------------------
Mini013106-01.dmp BugCheck 1000008E, {c0000005, f79ea04d, f4ff699c, 0}
Owning Process 8270a520 Image: apcsystray.exe
Probably caused by : hidusb.sys ( hidusb!HumGetSetReport+158 )
 
thats strange I didnt get any notification that howard or cpc2004 replied to me. just thought id check and there the last 2 messages were. strange.

dang i will order another memory module and if problem still persist then i'll put both in and i'll have a gig of ram. i will switch to another slot and blow the dust out real good. machines not overheating i dont believe. i have 2 extra case fans inside the box and here in winter country i have the side of the case open and cold air blowing in and its not even warm to the touch.

if i downclock will i not be loosing performance to the machine ??? would i not be better off to just buy new ram and cpu and see what results are ???

so running my memory testers for literally hours still isnt long enough. why would they not indicate a problem ???

thank you cpc2004 much appreciated.
 
I had the same issue with an ECS mobo I had. Rebuilt the OS more then once and still got random crash's. I had nothing but problems with that ECS motherboard so it is not in the closet and a MSI motherboard resides in it's place. I have not had a single crash since. I don't know what it is about the ECS boards but they suck IMHO.
 
Status
Not open for further replies.
Back