21

My company issued laptop is Dell Latitude E7450 running Windows 7 Professional. If I hibernate the system, on boot it displays a blank screen and does not respond to input. Subsequent reboots do not yield different results. My IT department tells me that Windows hibernation is problematic on Dells, and that hibernating destroys the boot partition. They rebuilt the boot-loader, and told me to be very careful not to let the system hibernate again.

I am a software engineer, not a system administrator, but something about this doesn't sound right to me.

Is my IT department right? Is it a common problem that users brick their systems through hibernation? Or is it something about the image they have created? Interaction with UEFI? Is there a solution?

Hugh White
  • 313
  • 1
  • 4
  • 9
    It doesn't sound right to me either. Ask them for reference from Dell Support. Moreover, your IT department already has the ability [but perchance not the knowhow] to disable hibernation. – jscott Jan 24 '17 at 19:54
  • 7
    You are correct, this is simply not true. I've had systems not return from hibernation, but it doesn't change the bootloader. – Jim B Jan 24 '17 at 20:29
  • 4
    I exclusively used dell laptops for the past 5 years and have never had a problem you've described with Hibernation. I think it's more likely the image they built and deployed to your laptop is the culprit, not Dell. – corsiKa Jan 25 '17 at 02:18
  • 7
    I wonder if they are using some crappy snake-oil "security" solution that loads before the WIndows bootloader and somehow doesn't play nice with hibernation. – André Borie Jan 25 '17 at 10:39
  • 2
    Here's another IT department that's just not fit for its job... – dr_ Jan 25 '17 at 12:42
  • Perhaps a meta question, but why do you think people seem to like this question so much? I'm not complaining, but I didn't think this would be interesting to so many people. Is it the IT vs users factor? – Hugh White Feb 01 '17 at 22:24
  • I'm pretty certain the last part is it. The title is great, and is attractive without being clickbait. Plus, most people have had this issue with an IT outfit in some way or another. I'm interested in how the situation will be handled, but that's going out of the realm of SF and more into workplace / soap opera. – Spooler Feb 02 '17 at 05:28

2 Answers2

25

Nope, it's definitely not trashing the bootloader since hibernation never touches the bootloader with a write. It doesn't have to, since all that's required to boot from a hibernation image is for that image to be present at all. It's always checked for during boot.

It's failing to restore from hibernation, and keeps trying to restore every time it boots. This is usually due to a driver issue, and likely can be fixed by making sure all of your drivers are installed and at their latest versions (or at least not ancient versions).

One easy way to get around this is to delete c:\hiberfile.sys from the drive (obviously with some live OS, or with the OS drive connected to another working machine). With that file gone, it will no longer try to restore that image to RAM and will continue with its normal boot process. The thing to do here would be to disable hibernation on this laptop entirely, or fix the underlying device reset / driver issue.

Spooler
  • 7,016
  • 16
  • 29
  • 6
    Note that deleting hiberfile.sys is equivalent to power-cycling the computer, with the corresponding possibility of data loss. Don't do this unless you really need to. – Mark Jan 25 '17 at 04:49
  • 6
    That's true. I've never tried to pick apart a hiberfile in an effort to repair it. Hopefully it's about as easy as analyzing a core dump, but even that is out of the realm of most IT outfits. That said, I wouldn't recommend anyone use hibernation to save data, but rather commit your working data to disk and use hibernation to preserve only your workflow state. – Spooler Jan 25 '17 at 04:53
  • 1
    Well, well. Here's a *very* good answer that details how one could accomplish a data recovery from a hiberfile. http://superuser.com/questions/660649/how-to-read-windows-hibernation-file-hiberfile-sys-to-extract-data – Spooler Jan 25 '17 at 05:58
  • "Note that deleting hiberfile.sys is equivalent to power-cycling the computer..." Well, not quite. Power cycling the computer can interrupt writes to the disk itself, whereas hibernation still allows the computer to perform a proper ACPI shutdown, allowing disk writes to complete and reducing the likelihood of filesystem corruption. – Nathan Osman Feb 14 '17 at 00:33
2

I agree the claim is preposterous; however I would check next time before disbelieving them. It's not that hard to check if the MBR and NTFS boot sector are intact. They might have actually done this on trying to fix it before.

joshudson
  • 403
  • 4
  • 10
  • Checking is not that hard, bit you may want to add how to do that for BIOS (MBR&NTbootsector) as well for UEFI. – Hennes Jan 25 '17 at 08:47
  • 2
    _"however I would check next time before disbelieving them"_ That's exactly what the OP is doing by asking this question. Your answer does not appear to be useful. – Lightness Races in Orbit Jan 25 '17 at 10:25
  • 2
    @LightnessRacesinOrbit: I mean check the MBR and boot sector not ask on the internet. – joshudson Jan 25 '17 at 16:11
  • OP ain't a systems guy. Checking to see if any boot related functionality has been written to would require some pretty crafty systems skill, since you'd have to check the MBR, hidden areas after the MBR, boot structures, boot loader, and the BCD in a consistent and reliable way both before and after the failure. It's even more hairy when considering UEFI. – Spooler Jan 26 '17 at 11:39