Can a packet crash a wi-fi driver?

I delivered a computer the other day -- it was fully lab tested.  

When I went to connect to the wireless network at the customer's house - BSOD (didn't catch the exact msg).

Had trouble rebooting the machine on site -- something about the Compaq's 'flea power' on the motherboard didn't allow it to boot and at the time I didn't know
about how to correctly do a 'power drain'.

Back at the lab, did a power drain with the help of a Compaq specialist.

Machine booted..  On my wireless network there has been no problem.

My concern is when I take the unit back to the customer site, will it crash again?  Why wouldn't it?

She has a Linksys range expander.  She never had the BSOD  until I reimaged the computer.  I took the wi-fi drivers on the COMPAQ driver disk.

So I'm feeling like there's some packet coming from her environment which crashes the wireless NIC (the driver).

Is that possible?

Who is Participating?
drawlinConnect With a Mentor Commented:
I was an Electronics Tech for 12 years before I touched my fist PC, 13 years ago.  just about every electronic circuit in every hardware conponent has some sort of over voltage or over current protection circuit engineered into it.  If you desgin, make and sell Harddrives, you don't want the reliability of a some other manufacturer's power supply to be the determining factor as to how often you will have to replace harddrives becasue they are under warranty.  Same if true for the motherboard, video cards, sound and even Chipsets that are attached to a motherboard.

Wicking is a soldering term used used to describe the phenomina of melted solder traveling up a wire toward the source of heat.  The same also applies to water moving through seams in plastic.

I would crack open the case, as much as possible, short of voiding any warrenties.  And look for any stains or residue.  Then Get some Isopropol alcohol (Technical grade if you can fide it)  and a some cotton swabs and just dab anything that looks like a stain.  Then let it dry.
> Is it possible to crash a computer, router, etc with a packet.... Yes... looking up "Ping of Death" for a good explanation of how it's possible to crash many systems with a packet.

Is that your problem, probably not.  Did it blue screen when it was booting, or when it was up and running?  Since you are moving the machine, it is more likely hardware jumbled and knocked loose to get a BSOD.  If not, then their might be an issue with the driver that was installed, or there could be an infected machine on her network.  More importantly, get the BSOD message if it happens again, otherwise, it's a guessing game.
supportorangesAuthor Commented:
thanks for your response.

it was up and running fine and seemed only to bluescreen on CONNECT to available network (??). all i was trying to do was reach the internet.   i discovered i accidentally placed the laptop on a rather wet carpet in my truck (didn't know somebody had spilled water back there).  when i picked up the laptop there was visible moisture on the bottom of the laptop.  in a bit  of a panic i quickly  brought it up fine just before delivering it (just to double check after drying it off).  so when it blue screened, i thought this was moisture related -- possible a short or something.  back in the lap unable to reproduce -- has been running for days.

i'll look up ping of death.  if you have any other ideas you wish to share -- they would be welcome.
Will You Be GDPR Compliant by 5/28/2018?

GDPR? That's a regulation for the European Union. But, if you collect data from customers or employees within the EU, then you need to know about GDPR and make sure your organization is compliant by May 2018. Check out our preparation checklist to make sure you're on track today!

If you have XP with latest service pascks and patches on OS and applications, Ping of death isn't the problem.  THat bug has been fixed for a long time.  The moisture issue seems much more likely.  Moisture in cracks, power up PC and heat is produced, heat causes moisture to wick through gaps in plastic and come in comtack with conductive surface.  A short or decrease in resistance in the circuit could cause a hardware based overload protection circuit to engage causing BSOD.  Now the warm computer evaporates the remainder of moisture on the trip back to the lab and problem goes away.
drawlin is correct about the Ping of Death not being an issue on an up to date XP machine.  I was giving it as an example of how you can cause a crash over a network; however if the right virus is floating on their network, you can see similar issues.

However, at this point, I would expect more of a hardware or driver issue.  Also, did you find out what they were doing at the time of bluescreen?  It could have been some app they are running, or a simple 1 time lockup.

Your best bet is to check the obvious, download all the latest XP updates from windows update (EXCLUDING drivers, I have seen these crash machines, vendors work better).  Go to the compac site and download and install all the latest drivers (including bios).  This will pretty much rule out driver issues.

For hardware, the moisture may have been an issue causing it to crash.  However, it may have dried out since then and the issue is gone or is very random since you havn't seen it for days.    You mentioned it happened when they connected the wireless card...perhaps they didn't have it fully seated and crash!

Where are the fans on the laptop (bottom or side)?  If they have a warm environment, it could be an overheating issue that crashes it, which could also explain why it doesn't crash in your lab (assuming it's cool).  You can have them elevate the corners of the laptop to allow more air flow, or put an external fan on it.

Good luck
Keith AlabasterEnterprise ArchitectCommented:
I am confused. If the box worked at your office, why would you need wireless drivers at all? The drivers must have been installed when you tested it (the PC and wireless card) in your lab.

If that logic is true, disable the wireless device when you take it back and power it up normally. Does it BSOD anyway? How long did it run previously at her location when you took it back before the blue screen appeared? Once its there and running, assuming it does, enable the card after the box is active.  
supportorangesAuthor Commented:
Thanks to everybody for their contribution.

To answer Keith's question, the wireless drivers were installed in the lab.  Also for everyone's info, all the service packs and all the hotfixes were applied as part of the original re-image.

A bios upgrade and latest wi-fi drivers would seem smart.

The BSOD didn't happen when they plugged in the wi-fi card, the wi-fi NIC is integrated.  It happened when I tried to connect to her network.  Perhaps this was coincidental since it never happened at the lab, before, or after delivery to the client.

The fan to the laptop is on the bottom but i only had it up for less than 10 minutes before the BSOD.

drawlin - i'm very interested in this overload protection circuit.  much of the bottom of the laptop appears to be sealed pretty well, such as the area memory is seated in.  i also didn't see any moisture where the hard drive lives. although i wiped down the bottom of the laptop as soon as i saw the moisture, i imagined that the laptop could have absorbed moisture inside the unit.  i ran the full quicktech pro hardware diags and everything passed.  i was frustrated and close to cracking the case and using a hairdryer on the unit and then I thought -- probably a bad idea as I was already more then a little stressed out.

drawlin - two short questions - 1) is there anything preventive I should do such as opening the case and looking for physical signs of moisture and 2) can I be confident that the hardware-based overload protection really protected the insides?

to everyone - this is definitely the weirdest most embarrassing problem i've ever encountered.  trust me - i couldn't make this up.  thank you for your participation.
supportorangesAuthor Commented:
the overload protection circuit is in the power supply.  definitely the thing would not boot at the customer site after the BSOD.  at the lab, a chat with Compaq attributed it to flea power on the motherboard needing to be 'discharged'.  it came up on its own during the chat (maybe I discharged the flea power without knowing it).
supportorangesAuthor Commented:
i have researched the bios and the wlan driver.  devmgmt is listing my wlan driver as 10/2006 so that's pretty new.  i'm running more burn-in h/w diags.  just had a thought.  better to leave this machine on to give a good chance at more evaporation.

i rebooted 5 times, ran ping /t for a while.  

i'm starting to think it must have been the moisture because the problem does not seem to be returning.

either that or one of those one-time-only ephemeral anomalies.
supportorangesAuthor Commented:
thanks for all who contributed.  i certainly have much more info now!
Keith AlabasterEnterprise ArchitectCommented:
Hmmm. O well, sorry our contributions did not help you as well. :(
All Courses

From novice to tech pro — start learning today.