次の方法で共有


An important hotfix if you can't hibernate your computer

My post about hotfixes certainly made it's way around various internal aliases over the last week. I've also been working (before and since the post) on some fixes with Product Support to the issues I found through the experiment. 

The other constructive bit of information I received because of this lengthy internal thread is that there is a hotfix I didn't know about that fixes a very ironic bug... 

https://support.microsoft.com/default.aspx?scid=kb;en-us;909095

It's been bothering me for a few months now that my laptop, with a gig of ram, didn't have enough resources to hibernate... now it can sleep easy... and you don't have to call to get this hotfix.

The other thing I've learned... not one person can agree on what "airing dirty laundry" means.  I think it's like saying "I know porn when I see it" for some people.  IMO, it's hard to say someone is "airing" dirty laundry if your customers can smell it and talk about it openly already.  "Airing dirty laundry", IMO, would be if I disclosed information about how poorly built/managed some internal process/system is that leads to user headaches.  To each his own.  This, however, is what I replied to some folks who think I'm simply airing dirty laundry.

One thing that driving transparency at this company has taught me is that there is a real positive gain to this approach for us and customers. The truth is that because of this (my post) we've discovered new customer validated bugs and are now driving the conversation rather than being on the sidelines.

I've also been thanked by a few customers in email for simply admitting to some issues. they've gone as far as volunteering to help fix the issues in whatever way possible.

We also got attention on how friction free calling is. I had one customer thank me for sending him down that path because it worked for him.

Comments

  • Anonymous
    September 06, 2006
    The comment has been removed
  • Anonymous
    September 10, 2006
    It's too early to guess whether the hotfix is really working and what its side effects might be, but it's not to early to have some first impressions and a suspicion.

    First impression is that the hotfix appears to allow hibernation as intended.  So far there were two occasions when, in the early stage of starting hibernation, accesses to the hard drive followed a different pattern than usual, and an intuitive guess says that those would have been failures without the patch.

    The suspicion of a side effect arises in a situation which calls for more testing to see if it reproduces, but I haven't had time yet.  One time I connected a USB hard drive, copied some files off of the hard drive to the notebook's internal drive, and then clicked on the icon in the notification area to ask Windows for permission to disconnect the USB hard drive.  Permission was denied, about 6 times in a row.  I had to close down all running applications and shut down Windows.

    Prior to the patch, here were my experiences:  As mentioned, the frequency where Windows XP failed to enter hibernation used to be around 3% but grew to around 20%.  The frequency of the apparently unrelated bug but which might be a side effect is as follows.  When asking for permission to disconnect a USB hard drive, around 25% of the time Windows 2000 or XP or 2003 rejected the first request (drive still being accessed by some unknown process), but usually Windows accepted the second request, so around 3% of the time it was necessary to close all running applications, shut down Windows, disconnect the USB hard drive, and then boot Windows again.

    In Vista, there's a misleading icon in the notification area which can still be clicked to ask for permission to disconnect a USB hard drive, but it's meaningless.  By design Vista does not allow a USB hard drive to be disconnected.  If you connect a USB hard drive, copy a few files off of it into the notebook's hard drive, and want to disconnect the USB hard drive, then 100% of the time it will be necessary to close down all running applications, shut down Windows, disconnect the USB hard drive, and then boot Windows again.

    Therefore, even though I only used a USB hard drive one time after this patch, and the failure to allow disconnecting occured on 100% of only one test case, the first impression is that it looks suspicious.  It looks like there's a possibility that a side effect of this Windows XP hotfix might be bringing some Vista compatibility to XP.

    Since I hibernate more frequently than I backup files to USB hard drives or restore backed up files from them, I am not uninstalling the patch.
  • Anonymous
    September 11, 2006
    In case my second feedback to this blog report gets posted, here is my third feedback with good news.

    Two more times I have connected USB hard drives to the machine with the new XP kernel, then asked XP for safe removal and received permission immediately.

    So the timing of the problem reported in my second feedback was probably just a coincidence.  Probably just an unrelated intermittent bug remaining unchanged, not a side effect of this fix, and not a deliberately induced compatibility with Vista.  Anyone who has a habit of temporarily connecting USB hard drives and then disconnecting them will be able to continue this habit under XP with this fix, probably.
  • Anonymous
    September 13, 2006
    > not one person can agree on what "airing dirty laundry"
    > means.

    Maybe you should stop worrying about it.  BSODs that are 100% reproducible outside of Microsoft are not reproducible inside Microsoft.  Microsoft can assure itself of non-reproducibility inside Microsoft by deleting uploads of full memory dumps that exceed 50MB, even in cases where Microsoft asked users to send full memory dumps.  Microsoft's dirty laundry will remain invisible and unsmellable inside Microsoft, and whatever it is that customers see and smell will not be any of Microsoft's concern.

    I wasted hours sending a full memory dump that Microsoft asked for.  I wasted many hours writing, getting screenshots, and uploading non-reproduceable garbage that only reproduces outside of Microsoft.  I wasted too many hours as a volunteer beta tester for dirty laundry that cannot be seen or smelled or read about inside of Microsoft.  Vista will be the same quality as it would have been if I hadn't bothered.
  • Anonymous
    June 09, 2009
    PingBack from http://insomniacuresite.info/story.php?id=9791