usbhub.sys can often cause a BSOD when an XP or W2K3 PC resumes from power-saving mode (standby or hibernate) if a usb device is removed whilst the power was off.
I apologise if I have missed some key update regarding this issue but I can find lots of references to the problem over the *years* but no real resolutions.
So far, I have reproduced this problem with:
XPSP2 usbhub.sys 5.1.2600.2180
XPSP2 usbhub.sys 5.1.2600.3114 (hotfix KB935192)
W2K3 SP1 usbhub.sys 5.2.3790.1830
W2K3 SP2 usbhub.sys 5.2.3790.3959
I have already contacted Microsoft support and they have provided the above hotfix which does not fix the problem.
Can this problem really still be there after so many years?
Has anyone made any progress re: the technical analysis of this problem and/or how to fix/avoid it (other than telling the user to not do something that they would often do quite inadvertently?
In the past there has been mention of a potential QFE but I have found no evidence of it from our Microsoft support channels so far.
This one still potentially affects a lot of people/users/devices.
> ----------
From: xxxxx@lists.osr.com[SMTP:xxxxx@lists.osr.com] on behalf of xxxxx@farsite.co.uk[SMTP:xxxxx@farsite.co.uk]
Reply To: Windows System Software Devs Interest List
Sent: Tuesday, October 09, 2007 12:10 PM
To: Windows System Software Devs Interest List
Subject: [ntdev] usbhub BSOD on XPSP2 and Server 2003 SP2 if usb device removed during power-saving
Can this problem really still be there after so many years?
Yes. Problem will stay there until you or somebody else convinces MS to really fix it. Please note there can be several different problems with the same symptoms and some of them already fixed by mentioned hotfixes.
Has anyone made any progress re: the technical analysis of this problem and/or how to fix/avoid it (other than telling the user to not do something that they would often do quite inadvertently?
In the past there has been mention of a potential QFE but I have found no evidence of it from our Microsoft support channels so far.
If you can reproduce the problem and it is really OS problem, open support case standard way and don’t allow to close it until it is really fixed. It can be a long and painful task; count with several months at least. If the problem can be reproduced at Vista, it can be probably fixed faster. It may be necessary that MS support people are able to reproduce problem themselves; they usually resist on it even if they have crashdumps (sigh).
Good luck.
Best regards,
Michal Vodicka
UPEK, Inc.
[xxxxx@upek.com, http://www.upek.com]