Thank you for being a valued part of the CNET community. As of December 1, 2020, the forums are in read-only format. In early 2021, CNET Forums will no longer be available. We are grateful for the participation and advice you have provided to one another over the years.

Thanks,

CNET Support

General discussion

Outlook 2003 causing IDE-related system crashes?

Jan 27, 2005 2:08AM PST

I have been dealing with a major PC problem for the last few weeks: Whenever cold-booting the computer (as opposed to restarting) after being off for more than a few hours, I was getting massive errors with the hard drive, which would either BSOD or freeze the system. The event logs would list ATAPI errors saying "Device\Ide\HD\0" wouldn't access, etc. When this happened, you could physically hear the hard drive whirring on and off repeatedly before the crash.

Whenever this happened, the restart after the crash would sometimes cause CheckDisk to start, but once XP came back up, nothing was wrong. No crashes, no problems, nothing.

Over the past few weeks I've been trying to isolate the problem, and nothing's worked: I've replaced both my hard drives and their IDE cables, thinking at first the drive was failing. I then removed both CD-drives, and tried doing clean XP installs without the sound cards installed. MemTest86 shows no errors at all with my RAM.

This morning, I turned on my computer after waking up, expecting the familiar screaming-hard-drive noises. That didn't happen... UNTIL I clicked the quicklaunch for Outlook 2003, which started the crash cycle as soon as it launched.

My event logs show no errors save for that pesky DCOM error with netman.dll and the Norton services that everyone seems to get and can't stop. I really don't think that has to do with my problem.

I know that this happens when I start Outlook 2003. I know it also only happens the first time I start it when powering on the computer from a power-off state for at least a few hours. So I feel like it's a COMBINATION of a hardware and a software problem. I just have no idea what is going on.

Is Outlook activating a flaw in a failing piece of hardware? Are my IDE ports just busted? Does the DCOM error have anything to do with it? Has anyone experienced anything like this before? Thanks in advance for any help; this is driving me crazy!

System stats:

Tyan Tiger MPX 2466 w/ ver 1.06 BIOS
2 sticks of 512MB DDR SDRAM
2 Athlop MP 2400 processors
Onboard LAN ethernet
SB Audigy 2
Nvidia GeForce MX 440
2 HDs, one 80 GB, one 40GB, both 7200 RPM/8mb
Cendyne CD-RW drive
Toshiba 16x DVD-ROM drive
Iomega ZIP 100MB internal
Floppy drive

Running Windows XP Professional, problems are identical with both SP1 and SP2 installed

Discussion is locked

- Collapse -
I bet you'll have to call this in. Here's why.
Jan 27, 2005 2:30AM PST

Most of MS's testing is on Single CPU machines. You may have found a new issue. Certainly one I haven't seen or heard of.

Good luck with that.

Bob

PS. A few seem to think these OSes and applications are bug free. Not many, but a few.

- Collapse -
So does that mean you're pretty sure it's Office?
Jan 27, 2005 3:41AM PST

It sounds like you think this is mainly, if not entirely, an Office 2003 problem, not a hardware problem. Is that the case?

I ask because my searching online for solutions about the various Event IDs related to my DCOM and IDE errors range anywhere from faulty devices to damaged RAM to software conflicts. Has anything like my hardware problem occured under different circumstances?

I have an old copy of Office 2000 lying around somewhere; I can swap out and see if that solves anything. I'd just like to know if I don't need to worry about a broken motherboard or something anymore Wink

- Collapse -
Why I noted such.
Jan 27, 2005 6:18AM PST

Your post contained nothing systemic to point to a hardware fault. Most DCOM errors I see are self inflicted with services being stopped, but your story isn't complete there.

As it stands you spent a lot of money on Office and should make the call.

Did you know there are patches you can't get unless you call in?

Bob

- Collapse -
Hmm...
Jan 27, 2005 11:29AM PST

...well I just came home from work and fired up the PC, and my theory might be partially blown now.

Before anything else, I went to the event log and checked what was there. The single DCOM error ("DCOM got error "The service database is locked. " attempting to start the service netman with arguments "" in order to run the server:" followed by a long registry key for netman.dll) and the usual service start messages.

But the event log also listed several repeated "Application popup: regs" and "Application popup: Machine check" events. Not sure what that means.

I opened up IE, and I got the sound of the hard drive clicking; the usual sign of a crash coming. But just one whir, and then everything was ok. I refreshed the event log and nothing came up.

So THEN I opened up Outlook, and got the same single whirring. I refreshed the event log, and lo and behold, an Atapi error came up of "The device, \Device\Ide\IdePort0, did not respond within the timeout period."

Again, I'm not really an expert, but I know the DCOM error is related to netman- a service that controls networking functions. I don't really have a "local area connection," but it's listed in my computer because an ethernet cable connects from my onboard port to my cable modem. Is there a chance that the netman error is affecting two programs that both access the internet?

- Collapse -
Sounds like bum hardware now.
Jan 27, 2005 11:57AM PST

"The device, \Device\Ide\IdePort0, did not respond within the timeout period." points to driver issues or a nearly failing hard disk. What usually happens is the owner goes into a denial phase about hardware issues and we get to wait for a total failure.

How does a DRIVE FITNESS TEST from the drive maker turn out?

Another fine mess are just big enough power supplies. Again, owners go into denial with "it worked fine last year", "it should be big enough" and other statements. Only the more seasoned will know that cold issues point to hardware issues such as aging capacitors...

Bob

Bob

- Collapse -
I doubt it
Jan 27, 2005 12:42PM PST

The IDE error would occur with different things: I actually have a CD-ROM drive unplugged right now thinking it meant the drive was faulty. This is a brand-new HD, because I originally thought the drive was bad and replaced it. New cables too. Sometimes the error is on the IDE-1 port, not IDE-2. I used a controller card for a fifth IDE device; right now that's removed too. Still get that error. If it's hardware, the only devices that are always in the computer are the graphics card, required to use the monitor, and the motherboard itself.

I'm curious about the board, because I've had problems with Tyan before- my previous board was the 2460, famously known for being the one that fried its PSU connector port.

(Speaking of which, following that incident last April, I repaced the PSU- it's an Antec Cool Blue 550, so I don't think it's a power level problem. I'm also using an APC backup so I don't think it's surges either.)

- Collapse -
Interesting
Jan 27, 2005 8:47PM PST

I did a search for the "Machine Check: Regs" event that appears on any startup that leads to a crash. One board mentioned this:

"This is the information I see in the Win Xp event viewer. That is, after I've rebooted, since I always crash when these messages are there. It repeats about 4 to 6 times. The first one has the same title minus the "REGS". I can find no data on this from MS knowlege base. All that I have been able to track down is that this has something to do with XP and AMD processors. The message has something to do with XP looking for a Pentium ID number (which of course my Athlon XP 2000+ does not have). I have disabled the bios entry that says to identify by processor ID, but I still crash from time to time (by crash I mean system locks up...all I can do is turn off power (it hangs when I try to reset). I've seen this question come up in other forums...but have yet to find an answer. Seems rather odd also that this is suppose to be a benign type message, not a warning or error message."

This is almost EXACTLY what happens to me. I am running dual AMD Athlon MP's, and have been since July 2003, so I'm not sure why this all would have started happening now. After work today I will try to make sure I have all my drivers for everything updated, but I think it's not a matter of updating anything, but uninstalling something... this seems like it might be a DLL file somewhere not knowing how to handle an AMD processor.

- Collapse -
Yes. It's going to be fun to find.
Jan 27, 2005 10:19PM PST

The issue of aging power supplies and parts will cause deep, hard to fathom issues. "It worked before" is very hard for people to get past. It does not matter.

The power or voltage can dip, brown out, vanish for a split second and with all I've seen I know to reduce the load, bring out the 650Watt PSU and try again.

If the machine is a few years old, then one has to take the stoic stance that such may be how this machine will be till it's replaced. It may cost too much in time or money to clear the error up.

In closing, I don't see where you are calling Microsoft to see if any patches address this issue. AGAIN, there are patches they have that are NOT PUBLISHED.

Bob

- Collapse -
ok
Jan 27, 2005 11:08PM PST

I'll see what I can get from Microsoft when I'm at home with necessary information.

As for the PC's age: the oldest parts of the system are from July 2003. The motherboard was put in in April of 2004, as was the PSU. The original PSU was about 350, I'm running I believe a 450 right now. If I have to buy ANOTHER 550 or 600, so be it, I'm just amazed that an 8-month old Antec wouldn't be able to handle my system, bulky as it is.

My only other thought for a hardware problem is that the reason this only happens from a cold start has something to do with the temperature of the CPUs. But why a system would have errors when the processors are at the one point they wouldn't be overheating (right at startup) is beyond me.

- Collapse -
About aging.
Jan 27, 2005 11:34PM PST

Look at Figure 7 of http://www.edpsciences.org/articles/epjap/pdf/1999/01/ap8043.pdf

That chart is all of 6 months of aging. Today's "lowest cost wins" designs push such designs into your PC and why you may find me repeatedly looking at power supplies that were just big enough when new. Your 350 Watt supply is not what it was when new.

There are more studies about this, but I hope this answers why this is and why I never skimp on PSUs. If there were available I'd just fit 1KW supplies so they would last 7 to 10 years...

Bob

- Collapse -
Yes...
Jan 28, 2005 12:24AM PST

...but I've said this I think twice in this thread already, the 350 Watt PSU is long gone. I'm running a name-brand 450-watt supply that's less than nine months old. If you're issue is "aging" the PSU is the newest thing on my computer right now.

Furthermore, I really don't understand how a PSU problem would only affect a cold-started PC... if the PSU was faulty it would be faulty all the time, wouldn't it?

I'm not "in denial" as you've suggested- if I need a new PSU I'd be happy to get one. I'm just asking about the physical rational for the PSU being to blame.

- Collapse -
You won't know till you change it.
Jan 28, 2005 12:30AM PST

That's where the shop end of the office has it all over home repair. We keep some monster supplies ready when we suspect a power issue.

If you were looking for me to write "it will cure it"... Such will never happen.

As to 350/450/etc. Yup, I saw that and decided I would not edit it.

Bob

- Collapse -
I can't afford to keep doing it that way.
Jan 28, 2005 1:10AM PST

I'm not asking you to say "this will cure it." I'm just asking if this type of problem sounds synonimous with a bad PSU, and if you had any theories as to why a PSU problem would only occur under these specific conditions. I've been told it's everything from Office to my motherboard to my graphics card to my sound card to the processors and now the PSU- if I'm supposed to buy new ones and play "won't know till you try it" I might as well just buy an entire new computer.

I'm willing to buy new hardware if there's at least a _precedent_ for that being the problem, but after doing that with a harddrive and two IDE cables I'm just asking for at least a little bit of theory beyond "buy this and see if it works." There's a limit to how many open-box devices my Best Buy will take back when I suddenly realize I didn't need them after all.

- Collapse -
Why it's not possible.
Jan 28, 2005 4:52AM PST

Today's machines are very complex (you know this) and the vendors are not providing sufficient test features into the boxes for us to look at some output and say "replace the trickypart."

I'm offering my view from many thousands of repairs and what we do is a quick swap with the monster PSU and retest. We do that since it's the cheapest part. Then we have the next cheapest part... Motherboard. then next and so on.

Usually the software issues have been tried before we see the machine.

Again, if you haven't called it in then you may not get some patch or other software that MS keeps in their support line only arsenal.

-> I still wonder if its worth fixing. It's only failing once, when cold. Maybe you should wait for a hard failure. Something like the AE-35 unit that HAL had an issue with.

- Collapse -
Worth fixing
Jan 28, 2005 5:10AM PST

Oh, it's worth fixing. I've had to reinstall Windows four times in the last three months because of this problem. During previous runs, the system wouldn't recover from a crash except in safe mode. Before my last re-format a week ago, the problem was that the PC wouldn't turn off after I selected "shut down." Instead windows would just shut down and the PC would keep running. That's why I'm leaning toward a driver conflict or a bad device port, because the only thing that always remains the same is that freaking DCOM netman error.

If it's hardware, it's either the PSU, the graphics card, or the AGP slot the graphics card is plugged into, because they're the only things in every instance of installation that weren't either replaced, temporarily removed, or thoroughly tested.

I'll go to Best Buy or CompUSA tonight, but the biggest PSU they'd sell would probably be 550 watts.

- Collapse -
Just an example.
Jan 28, 2005 5:30AM PST
- Collapse -
ok
Jan 28, 2005 5:42AM PST

Well I guess I'll order this then, unless you think it's available at one of the common PC stores.

- Collapse -
Most likely as common as...
Jan 28, 2005 5:57AM PST

Dual MP systems... Wink