Explorer hangs accessing drive after reboot 2

[colinsw]

Hi

My machine is a 1GHz PIII, 512Mb and more disk than it knows what to do with. It runs Win2k SP4 and has all critical patches applied. The m/board, RAM and disks have had diagnostics run upon them, both individually and together, outside the machine and found to be OK.

I have the problem that immediately after startup or a reboot and after waiting until all disk activity has stopped, if I then open Explorer and access the C, D or E drive, Explorer hangs and a Restart (through Task Manager) is required to get the machine going again.

The machine is virus, worm and trojan free. This problem has been happening for some many months and did not coincide with the installation of any particular software package or patch.

Has anyone seen this happen before and/or have any ideas where to start looking for the cause, please?

 

[bcastner]

colinsw,

O4 - HKLM\..\Run: [NvCplDaemon] RUNDLL32.EXE C:\WINNT\System32\NvCpl.dll,NvStartup
O4 - HKLM\..\Run: [nwiz] nwiz.exe /install


This says that Symantec (Norton) AV is running.
That may not be what you intended.

 

[colinsw]

I the course of checking the last few instructions, I got a bit adventurous and quite by accident accessed the CDrom before the C drive. Doing it this way does not appear to cause the problem. I then rebooted and tried accessing the parallel port attached ZIP drive, then the C drive, same thing no problem. Rebooted again, tried acessing the C drive first, problem occurs. Go figure!

 

[1stITMAN]

I think it could be controller issue try changes ur ide cables

I.T Systems Support Engineer
Bsc. (Hons).

 

[colinsw]

> O4 - HKLM\..\Run: [NvCplDaemon] RUNDLL32.EXE C:\WINNT\System32\NvCpl.dll,NvStartup
> O4 - HKLM\..\Run: [nwiz] nwiz.exe /install

These two entries are something to do with the NVidia graphic card. I do not have nor have ever had NAV installed on this machine.

 

[bcastner]

colinsw,

You are absolutely right.

I apologize.

 

[colinsw]

bcastner: no problem, with the multitude of software that gets installed on machines these days, and with very similar names, it can be very difficult to positively identify a single file.

1stITMAN: I have change the HDD cable. Couldn't find a spare ATA100 cable, but have used a standard IDE one instead. While the drive is running much slower, DMA66, after three reboots accessing C drive has not faulted. I don't yet consider this conclusive as without using the right cable, things are not exactly as they were. That is that running at a slower speed may hide a controller fault of some sort. I'll go out soon and get an ATA100 cable from DSE and try that. Will report back with the results.

Thanks guys for all your help so far.

 

[colinsw]

I spoke too soon!

After 4 reboot cycles the problem recurred, that is accessing C drive after a clean boot and waiting for drive activity to stop, causes Explorer to show as Not Responding and CPU usage goes to 100%.

This with the STD IDE cable still fitted. Maybe we have a faulty controller here?

I will swap the HDDs onto the second drive port, leaving the CD and DVD disconnected for the next round. Report back shortly.

 

[bcastner]

If appropriate, head to the Intel web site and download and apply their chipset update utility.

I think 1stITMAN's notion of this being a cable, or controller issue is likely on the right track.

If you are using ATA100 or better drives, be sure to use certified 80 conductor cables, not the standard IDE cables.

 

[colinsw]

> If appropriate, head to the Intel web site and download and apply their chipset update utility.

I wasn't aware that Intel did anything for non-Intel motherboards. Remember I'm using a Gigabyte m/board.

> I think 1stITMAN's notion of this being a cable, or controller issue is likely on the right track.

Current tests seem to have disproved both of these items. Although I was kind of hopeful from the initial results. Swapping the cable out has not solved the problem and putting only the drives onto the second controller likewise. The board appears to have two distinct controllers, so even if the Primary was faulty, the secondary should have worked. Also consider that apart from this issue, the drive system work perfectly. Reading and writing huge amounts of data 24/7. I would have thought that a controller would have had some effect on data going through it, like slow access or data corruption. Diagnostics have shown this not to be the case.

Do you think I am mistaken in my assumptions on that point?

> If you are using ATA100 or better drives, be sure to use certified 80 conductor cables, not the standard IDE cables.

Correct, but as mention I didn't have a spare ATA100 cable, so used STD IDE instead. Seemed to initially work ok, but then failed and did not resolve the issue. Have now tried a replacement ATA100 cable, problem remains. Also tried plugging the drives into another machine with the same motherboard, no change.

Given all this, I'm reasonably happy that we have proven off the drives, the drive cable and the drive controllers, but I am still somewhat suspicious of windows.

I have no problem with changing the motherboard for one that uses a different chipset (currently VIA), but I not sure I want to go there until we have proven it to be the cause. What do you think?

 

[bcastner]

Quote: "I wasn't aware that Intel did anything for non-Intel motherboards. Remember I'm using a Gigabyte m/board."

Intel, in conjunction with Microsoft, HP, Compaq and others created new device drivers for their support chips. Originally issued by Intel as the "Application Accelerator" it is no more modestly described.

If you have an Intel processor, and if you have the support ICs often used, go to the Intel Web site and download the upgrade utility software. If you have AMD and/or Sis and/or Via those sites are worth investigating as well. There have been a lot of driver updates no matter what ICs you are using on your motherboard.

Intel:

http://www.intel.com/design/software/drivers/platform/inf.htm

SiS:

http://download.sis.com/index1.htm

(they make this site a small p.i.a.)

Via:

http://www.via.com.tw/en/index/index.jsp

 

[colinsw]

I have heard a lot of bad things about the Intel Application Accelerator and the number of problems that have been attributed to it. These problem have not been across the board, but have been reported frequently enough to make me very suspicious. Has it now got to a level of reliability that it can be safely installed without fear of it causing more problems than it is worth? I'll go download it in anticipation.

Being a Gigiabyte motherboard with a VIA chipset, I have been to both these sites. I already had the current BIOS version, but VIA had put out a update v4.51. Downloaded and applied it, problem remains.

 

[colinsw]

I've now checked the Intel website which says that the Chipset update Utility requires an Intel 8xx chipset. There is nothing stated for the Application Accellerator, so i have downloaded and run it. It just brings up a message saying incompatible hardware.

Further testing: Above I mentioned that I had some success with accessing devices attached to the secondary controller. Well, I've just rebooted three times in a row, after the first two of those reboots I clicked on the ZIP drive, Explorer locked up, so while that looked good initially, it has now been proven to be not so.

 

[bcastner]

I created confusion for you I did not intend. My underlying point was the nobody pushes BIOS upgrades, chipset upgrades out to end users, they need to actively take the time to make their system current.

The Intel Application Acclerator was superceded by the software I linked way above. The whole notion of the original package (and similar steps by SiS and Via) was that drive devices should be independent. The only issue I ever faced with the original Intel Application Accelerator was an odd issue with early Pentium IVs and the MS java JIT compiler. I have recommended, used, and have been happy keeping my critical workstations upgraded with BIOS and chipset level device drivers for Intel, SiS and Via, and encourage others to spend the time to do the same.

 

[colinsw]

Thanks for making that clear.
I was getting kind of worried, trying to understand why I was being referred to an Intel site for a non-Intel motherboard and chipset.

I have spend some time this afternoon, fiddling. Removed all except video card and made sure the drivers were removed. Also reduced the installed software to absolute minimum and made sure that everything associated with the removed programs was cleaned from the registry, Program Files folder/menu and System Tray are looking quite 'naked'. Nothing improved, except the machine sped up quite a bit when O2k was removed.

Ghosted the partition at this stage, for safekeeping and updated the ERD disk. Did the Repair install suggested earlier. No change. Just in case its relevant, I did notice that when the machine is running in Safe mode the problem does not happen, ever. Given the time Safe mode takes to start running several reboot cycles took ages.

I'm leaning towards the theory that one of the components associated with the Windows drive system might be mismatched or faulty. What do you think?

 

[1stITMAN]

You were saying that u have been getting reboot loops!
Have u viewed the boot log ntbtlog.txt.

I.T Systems Support Engineer
Bsc. (Hons).

 

[colinsw]

Sorry, but I don't know where you got that idea from. I didn't mention a 'reboot loop' anywhere in my previous messages.

What exactly are you referring to?

 

[1stITMAN]

sorry this is what i read in u previous post

"After 4 reboot cycles the problem recurred, that is accessing C drive after a clean boot and waiting for drive activity to stop, causes Explorer to show as Not Responding and CPU usage goes to 100%".

I miss interpret it to myself. Have u checked the event log since.



I.T Systems Support Engineer
Bsc. (Hons).

 

[colinsw]

Hmmm, I guess I'll try to be a bit more clear, sorry.

'Reboot cycles' in context means that I purposely rebooted the machine, as the problem only occurs immediately after a clean boot.

Having checked the log files, there is nothing other than normal system messages.

 

[junkiemeel]

i know from experience that
C:\WINNT\System32\nvsvc32.exe
causes the same problems you mentioned. Evil Nvidia help program.

i read somewhere you can disable this in services without experiencing trouble. try reading up on this somewhere else just to make sure.

lev

 

[colinsw]

Thank you so much for that interesting idea.

I haven't heard of it before and will certainly do some research this afternoon. I have set the service in SCM to manual and stopped the coppy currently running. Seemingl;y without adverse effects. I run througha few reboots in a minute and see what happens.

Thanks again.