[opensuse] Resume From Windows Hibernation

older
[opensuse] openSuSE Bugzilla Dead...

Sudhir Anand

8 Jul 2017 8 Jul '17

14:54

All. After windows goes into hibernation mode, the computer will not boot into Linux. It stops and asks for root password and drops to command prompt. The only way to reboot into Linux is to start windows to recover from its hibernation mode. After shutting windows down normally, Linux will boot normally. Is this normal behavior? Can any changes be made to recover from Windows hibernation mode? Regards. -- Sudhir Anand -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Show replies by date

Carlos E. R.

8 Jul 8 Jul

16:25

On 2017-07-08 16:54, Sudhir Anand wrote:

...

All.

After windows goes into hibernation mode, the computer will not boot into Linux. It stops and asks for root password and drops to command prompt.

Surely there will be a message about the reason. The system is expecting you to correct the situation.

...

The only way to reboot into Linux is to start windows to recover from its hibernation mode. After shutting windows down normally, Linux will boot normally.

Is this normal behavior? Can any changes be made to recover from Windows hibernation mode? Yes. No.

Let me guess. You probably got an error message in Linux about the impossibility to mount the Windows partitions because it is in an inconsistent state or some similar words. When you hibernate a machine you can not mount the partitions it had opened in any other operating system, nor in another instance of the same operating system. You need to awake the hibernated system, then fully close it. If you force the mount, the destruction caused to those filesystem can be horrible. In Linux, you can comment out the Windows mount line, then continue booting. Or write the option "nofail". -- Cheers / Saludos, Carlos E. R. (from 42.2 x86_64 "Malachite" at Telcontar)

Sudhir Anand

16:48

On Saturday, 8 July 2017 17:25:10 BST Carlos E. R. wrote:

...

On 2017-07-08 16:54, Sudhir Anand wrote:

...
All.

After windows goes into hibernation mode, the computer will not boot into Linux. It stops and asks for root password and drops to command prompt.

Surely there will be a message about the reason. The system is expecting you to correct the situation.

...
The only way to reboot into Linux is to start windows to recover from its hibernation mode. After shutting windows down normally, Linux will boot normally.

Is this normal behavior? Can any changes be made to recover from Windows hibernation mode?

Yes. No.

Let me guess. You probably got an error message in Linux about the impossibility to mount the Windows partitions because it is in an inconsistent state or some similar words.

When you hibernate a machine you can not mount the partitions it had opened in any other operating system, nor in another instance of the same operating system. You need to awake the hibernated system, then fully close it.

If you force the mount, the destruction caused to those filesystem can be horrible.

In Linux, you can comment out the Windows mount line, then continue booting. Or write the option "nofail".

Carlos, thank you for your prompt response and for clarifying the cause of the error message and the solution suggested. Kind Regards. -- Sudhir Anand -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Doug

9 Jul 9 Jul

02:47

On 07/08/2017 11:25 AM, Carlos E. R. wrote:

...

On 2017-07-08 16:54, Sudhir Anand wrote:

...
All.

After windows goes into hibernation mode, the computer will not boot into Linux. It stops and asks for root password and drops to command prompt. Surely there will be a message about the reason. The system is expecting you to correct the situation.

...
The only way to reboot into Linux is to start windows to recover from its hibernation mode. After shutting windows down normally, Linux will boot normally.

Is this normal behavior? Can any changes be made to recover from Windows hibernation mode? Yes. No.

Let me guess. You probably got an error message in Linux about the impossibility to mount the Windows partitions because it is in an inconsistent state or some similar words.

When you hibernate a machine you can not mount the partitions it had opened in any other operating system, nor in another instance of the same operating system. You need to awake the hibernated system, then fully close it.

If you force the mount, the destruction caused to those filesystem can be horrible.

In Linux, you can comment out the Windows mount line, then continue booting. Or write the option "nofail".

I think the answer to the problem is to turn FAST BOOT off, in Windows.Then it won't hibernate when you shut it down. It will take longer to boot up, but so what? Wait another minute! --doug -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

David C. Rankin

21:57

On 07/08/2017 11:25 AM, Carlos E. R. wrote:

...

When you hibernate a machine you can not mount the partitions it had opened in any other operating system, nor in another instance of the same operating system. You need to awake the hibernated system, then

I'm glad Carlos snapped to what was going on... I was still scratching my head wondering what the heck Sudhir was talking about. I dual-boot (just because I happen to have 2 separate drives in the laptop and left the Win10 SSD alone when I installed Leap). I would NEVER, not in a MILLION YEARS, have considered trying to dual-boot when one OS was hibernated. That is just such a bizarre thought in my world it never occurred to me what Sudhir was actually talking about. I don't know why, but it just seems like common-sense that you fully shutdown one OS before trying to boot another -- maybe that's just me... (and a one reason I will not allow grub to touch the win10 drive, relying on 'Select Boot Device' from the BIOS to chose which drive to boot) Virtualbox of KVM is the way to go to virtualize windows within Leap. You want to run windows? just start it like another application. You can even create a .desktop file that launches the vboxmanage commandline. I've even got to where I virtualize windows on my server and give it 2-cores and 4GB of ram. I access it via rdesktop -- even over wifi it is fine (not snappy, but fine). -- David C. Rankin, J.D.,P.E.

Carlos E. R.

10 Jul 10 Jul

12:50

On 2017-07-09 23:57, David C. Rankin wrote:

...

On 07/08/2017 11:25 AM, Carlos E. R. wrote:

...
When you hibernate a machine you can not mount the partitions it had opened in any other operating system, nor in another instance of the same operating system. You need to awake the hibernated system, then

I'm glad Carlos snapped to what was going on... I was still scratching my head wondering what the heck Sudhir was talking about. I dual-boot (just because I happen to have 2 separate drives in the laptop and left the Win10 SSD alone when I installed Leap).

I would NEVER, not in a MILLION YEARS, have considered trying to dual-boot when one OS was hibernated. That is just such a bizarre thought in my world it never occurred to me what Sudhir was actually talking about.

Why bizarre? It seems a very normal thing to want to do and I have done it. One is working on one system, now wants to do something else on the other system, so hibernate the current system because it preserves the status, and boot to the other. I have done it. It is nice. Grub/lilo will not allow it, but there are tricks to let it allow the thing. The problem is trying to use hibernated filesystem on the other system. The corruption that results is horrible (specially if both are Linux). The system that boots sees the filesystem as not closed properly, so runs fsck on it. When you restore the hibernated filesystem, it expects the filesystem as it left it, and has a memory image of open files that are now closed, and writes things as if it was still holding the previous state. Horrible. The result is horrible. Better format and restore from backup. The good news is that it is possible to know that NTFS is in hibernated state, and Linux knows it and refuses to mount that disk, so it is not corrupted. But not such a thing with Linux filesystems.

...

I don't know why, but it just seems like common-sense that you fully shutdown one OS before trying to boot another -- maybe that's just me... (and a one reason I will not allow grub to touch the win10 drive, relying on 'Select Boot Device' from the BIOS to chose which drive to boot)

To me it seems reasonable to expect that technology would know how to do it :-) (no, I know that this is not possible and why)

...

Virtualbox of KVM is the way to go to virtualize windows within Leap. You want to run windows? just start it like another application. You can even create a .desktop file that launches the vboxmanage commandline. I've even got to where I virtualize windows on my server and give it 2-cores and 4GB of ram. I access it via rdesktop -- even over wifi it is fine (not snappy, but fine).

Good for you. My Windows laptop has only 4 GiB, so I can at most give 2 GiB to the other system, making the current one run slow. And running the other system in a window, which is what I prefer to do, gives it too little screen space in a laptop. -- Cheers / Saludos, Carlos E. R. (from 42.2 x86_64 "Malachite" at Telcontar)

John Andersen

8 Jul 8 Jul

18:33

On 07/08/2017 07:54 AM, Sudhir Anand wrote:

...

All.

After windows goes into hibernation mode, the computer will not boot into Linux. It stops and asks for root password and drops to command prompt.

The only way to reboot into Linux is to start windows to recover from its hibernation mode. After shutting windows down normally, Linux will boot

This is a known problem with windows. Hybernation is not the same thing as off. Especially with UEFI boot, where windows code is doing some of the boot process. It knows it is hybernated, and its not going to let go. This is why dualboot sucks. It was never great, and its getting worse every month. Are you SURE you wouldn't rather run Windows or Linux in a VirtualBox VM? You will NEVER be happy with dual boot. I promise. -- After all is said and done, more is said than done. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Andrei Borzenkov

18:40

08.07.2017 21:33, John Andersen пишет:

...

This is a known problem with windows.

How exactly is it specific to Windows? At least, it won't let you mount drives; while with other Linux instance you'd have already destroyed your filesystem. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

John Andersen

18:49

On 07/08/2017 11:40 AM, Andrei Borzenkov wrote:

...

08.07.2017 21:33, John Andersen пишет:

...
This is a known problem with windows.

How exactly is it specific to Windows? At least, it won't let you mount drives; while with other Linux instance you'd have already destroyed your filesystem.

When your lowest level boot loader is WINDOWS how is it possibly specific to anything but windows? As long as you are not using the same Swap space (or hibernation file) there should never be a problem hibernating one system to boot another. Hybernation is not the same thing as suspend. People get around this using Refind in many cases, although not so much in the opensuse world. http://www.rodsbooks.com/refind/ -- After all is said and done, more is said than done. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Carlos E. R.

20:25

On 2017-07-08 20:49, John Andersen wrote:

...

On 07/08/2017 11:40 AM, Andrei Borzenkov wrote:

...
08.07.2017 21:33, John Andersen пишет:

...
This is a known problem with windows.

How exactly is it specific to Windows? At least, it won't let you mount drives; while with other Linux instance you'd have already destroyed your filesystem.

When your lowest level boot loader is WINDOWS how is it possibly specific to anything but windows?

The lowest boot loader is UEFI or BIOS, not Windows.

...

As long as you are not using the same Swap space (or hibernation file) there should never be a problem hibernating one system to boot another. Hybernation is not the same thing as suspend.

ROTFL! Try it. I did. You will destroy the hibernated Linux filesystems beyond repair. -- Cheers / Saludos, Carlos E. R. (from 42.2 x86_64 "Malachite" at Telcontar)

John Andersen

21:31

On 07/08/2017 01:25 PM, Carlos E. R. wrote:

...

On 2017-07-08 20:49, John Andersen wrote:

...
On 07/08/2017 11:40 AM, Andrei Borzenkov wrote:

...
08.07.2017 21:33, John Andersen пишет:

...
This is a known problem with windows.

How exactly is it specific to Windows? At least, it won't let you mount drives; while with other Linux instance you'd have already destroyed your filesystem.

When your lowest level boot loader is WINDOWS how is it possibly specific to anything but windows?

The lowest boot loader is UEFI or BIOS, not Windows.

...
As long as you are not using the same Swap space (or hibernation file) there should never be a problem hibernating one system to boot another. Hybernation is not the same thing as suspend.

ROTFL! Try it. I did. You will destroy the hibernated Linux filesystems beyond repair.

He didn't have hibernated linux file systems. He had hibernated windows file systems. -- After all is said and done, more is said than done.

Carlos E. R.

21:42

On 2017-07-08 23:31, John Andersen wrote:

...

On 07/08/2017 01:25 PM, Carlos E. R. wrote:

...
On 2017-07-08 20:49, John Andersen wrote:

...
On 07/08/2017 11:40 AM, Andrei Borzenkov wrote:

...
08.07.2017 21:33, John Andersen пишет:

...
This is a known problem with windows.

How exactly is it specific to Windows? At least, it won't let you mount drives; while with other Linux instance you'd have already destroyed your filesystem.

When your lowest level boot loader is WINDOWS how is it possibly specific to anything but windows?

The lowest boot loader is UEFI or BIOS, not Windows.

...
As long as you are not using the same Swap space (or hibernation file) there should never be a problem hibernating one system to boot another. Hybernation is not the same thing as suspend.

ROTFL! Try it. I did. You will destroy the hibernated Linux filesystems beyond repair.

He didn't have hibernated linux file systems. He had hibernated windows file systems.

Come on. You said that this issue only affects machines that have Windows installed, that it is a Windows issue. We are saying it is not, it affects any machine that boots more than one operating systems and one is hibernated, even if it is two Linux systems. -- Cheers / Saludos, Carlos E. R. (from 42.2 x86_64 "Malachite" at Telcontar)

David C. Rankin

9 Jul 9 Jul

21:46

On 07/08/2017 01:33 PM, John Andersen wrote:

...

Are you SURE you wouldn't rather run Windows or Linux in a VirtualBox VM? You will NEVER be happy with dual boot. I promise.

Not to mention changing the BIOS time from UTC to local and back as you boot between OS's.... -- David C. Rankin, J.D.,P.E. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

James Knott

21:52

On 07/09/2017 05:46 PM, David C. Rankin wrote:

...

On 07/08/2017 01:33 PM, John Andersen wrote:

...
Are you SURE you wouldn't rather run Windows or Linux in a VirtualBox VM? You will NEVER be happy with dual boot. I promise.

Not to mention changing the BIOS time from UTC to local and back as you boot between OS's....

I configured my ThinkPad, so that Linux expects the hardware clock to be local time and configured Windows to not change anything. Works well enough. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

David C. Rankin

22:15

On 07/09/2017 04:52 PM, James Knott wrote:

...

...
Not to mention changing the BIOS time from UTC to local and back as you boot between OS's....

I configured my ThinkPad, so that Linux expects the hardware clock to be local time and configured Windows to not change anything. Works well enough.

I used to do just that, but moved completely away from localtime on Linux as Arch dropped support for it a couple of years back. There were technical reasons for doing so (which I've long since forgotten, but were sound) You have a registry hack on windows that allows it to use UTC, but it has side-effects on the windows side as well. I guess I'm just old-school and work with the UTC-5 and UTC-6 for daylight and standard time. That's generally why I have 14-days of uptime on Linux separated by 2 hrs of uptime on windows just to let updates apply. It's a bit of a pain swapping back and forth :) -- David C. Rankin, J.D.,P.E. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Carlos E. R.

10 Jul 10 Jul

12:28

On 2017-07-10 00:15, David C. Rankin wrote:

...

On 07/09/2017 04:52 PM, James Knott wrote:

...
...
Not to mention changing the BIOS time from UTC to local and back as you boot between OS's....

I configured my ThinkPad, so that Linux expects the hardware clock to be local time and configured Windows to not change anything. Works well enough.

I used to do just that, but moved completely away from localtime on Linux as Arch dropped support for it a couple of years back. There were technical reasons for doing so (which I've long since forgotten, but were sound) You have a registry hack on windows that allows it to use UTC, but it has side-effects on the windows side as well. I guess I'm just old-school and work with the UTC-5 and UTC-6 for daylight and standard time. That's generally why I have 14-days of uptime on Linux separated by 2 hrs of uptime on windows just to let updates apply. It's a bit of a pain swapping back and forth :)

Several years ago I changed Windows to expect UTC time on the CMOS clock and it works fine, and I haven't noticed any side effect. -- Cheers / Saludos, Carlos E. R. (from 42.2 x86_64 "Malachite" at Telcontar)

Carl Hartung

12:46

On Mon, 10 Jul 2017 14:28:02 +0200 Carlos E. R. wrote:

...

Several years ago I changed Windows to expect UTC time on the CMOS clock and it works fine, and I haven't noticed any side effect.

+1 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Carlos E. R.

13:01

On 2017-07-10 14:46, Carl Hartung wrote:

...

On Mon, 10 Jul 2017 14:28:02 +0200 Carlos E. R. wrote:

...
Several years ago I changed Windows to expect UTC time on the CMOS clock and it works fine, and I haven't noticed any side effect.

+1

For the record (again) the instructions are here: https://en.opensuse.org/SDB%3AConfiguring_the_clock However, I don't understand this paragraph: "Next problems can be Enhanced Write Filter and File-Based Write Filter in Windows[tm]." -- Cheers / Saludos, Carlos E. R. (from 42.2 x86_64 "Malachite" at Telcontar)

James Knott

13:28

On 07/10/2017 08:28 AM, Carlos E. R. wrote:

...

Several years ago I changed Windows to expect UTC time on the CMOS clock and it works fine

How did you do that? -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Carlos E. R.

13:48

On 2017-07-10 15:28, James Knott wrote:

...

On 07/10/2017 08:28 AM, Carlos E. R. wrote:

...
Several years ago I changed Windows to expect UTC time on the CMOS clock and it works fine

How did you do that?

Posted on another post in this thread :-) copied from here: Date: Mon, 26 Mar 2012 15:40:05 +0200 From: Dr. Werner Fink <...@suse.de> To: opensuse@opensuse.org ... Indeed ... please compare with http://en.opensuse.org/SDB%3AConfiguring_the_clock I've written with my HOWTO in the lunch hour. -- Cheers / Saludos, Carlos E. R. (from 42.2 x86_64 "Malachite" at Telcontar)