Reboot instead of shutdown on DH77EB, OS 12.1 and tumbleweed

Hello community,

Background info:
I recently bought a new box. Motherboard Intel DH77EB, CPU Intel Core i7-3770, No additional GPU, 16GB of Ram (4*4 1333MHz), LVM setup with 3 HDD. BIOS is 075, the most recent. The on-board graphics is Ivy bridge HD4000, from what I googled. I have a dual monitor setup. I am an average linux user, technically rather not very familiar with shutdown and power-management process. This is a linux-only box, I have no spare windows license to check if this would work with more popular operating systems (and I really don’t care except if this would prove that I have a hardware problem).
The story:
I originally installed OpenSuse 12.1. I had a terrible first impression (mouse pointer was not visible on the second monitor, the whole feeling was awfully slow for such hardware). I applied all available updates and changed to KDE SC4.8 stable repositories. Dual monitor setup, worked, the overall feeling was still kind of slow and unresponsive (slower than on my 6 year old older box), and moreover, I could not properly shutdown. This means, the whole shutdown procedure works, but it ends with a reboot.
The problem:
I decided to go for tumbleweed (So now I am on kernel 3.4.4). I have no additional repositories except for the ones suggested for tumbleweed and packman (the tumbleweed one). Finally, the feeling of having a new computer arrived. Responsiveness is snappy. All is perfect, except for the inability to properly power off.

Any good suggestions?

I would rather not go for acpi=off and loose all power management.

PS: I would have submitted this thread in the tumbleweed forum, but since it occurred before my upgrade, I think it is not tumbleweed specific. I could have gone, for the hardware forum, but that would be drawing conclusions…

On 2012-07-10 08:56, Nikos78 wrote:

> The problem:
> I decided to go for tumbleweed (So now I am on kernel 3.4.4). I have
> no additional repositories except for the ones suggested for tumbleweed
> and packman (the tumbleweed one). Finally, the feeling of having a new
> computer arrived. Responsiveness is snappy. All is perfect, except for
> the inability to properly power off.

How do you request the power off? Tried command line?

> PS: I would have submitted this thread in the tumbleweed forum, but
> since it occurred before my upgrade, I think it is not tumbleweed
> specific. I could have gone, for the hardware forum, but that would be
> drawing conclusions…

Here is fine, for the moment. Later, if we find something wrong with the
software, you will have to report in factory mail list or bugzilla to call
the attentions of the devs.


Cheers / Saludos,

Carlos E. R.
(from 11.4 x86_64 “Celadon” at Telcontar)

How do you request the power off? Tried command line?

sudo /sbin/shutdown -hP now

from a terminal window.

and of course also via the respective KDE menu entry.

The effect is the same. A perfect shutdown, but a reboot instead of a poweroff.

A perfect shutdown,

As I realized yesterday that is not true. The last line I can see before the screen is blacked is:

Could not finalize…

Now I know the dots are the important part. Where can I read the system log regarding this (if this is not deactivated in a previous step of the shutdown process)?

and …giving up

,but I still haven’t managed to read the whole log entry. Does anybody know, if this is still logged somewhere?

On 2012-07-12 07:56, Nikos78 wrote:
> ,but I still haven’t managed to read the whole log entry. Does anybody
> know, if this is still logged somewhere?

I don’t think so.

These things are obtained via serial port loggin, or sometimes making a
video instead of taking a photo.


Cheers / Saludos,

Carlos E. R.
(from 11.4 x86_64 “Celadon” at Telcontar)

I don’t know, but can imagine what that is. Therefore, I guess it is out of the question…(especially if serial port logging really requires a serial port)

Now, that is a good idea. I just hope my video camera provides enough fps to catch the message.

On 2012-07-12 15:16, Nikos78 wrote:
>
> robin_listas;2473939 Wrote:
>>
>> These things are obtained via serial port loggin
>
> I don’t know, but can imagine what that is. Therefore, I guess it is
> out of the question…(especially if serial port logging really
> requires a serial port)

It does.

Although I read recently, on the ubuntu wiki, that a special usb cable can
now be used, too. I gave the links recently, talking about hibernation
somewhere

incantations.

I think I posted more links somewhere… No, that’s the one, but it doesn’t
explain a lot :frowning:

> robin_listas;2473939 Wrote:
>>
>> or sometimes making a
>> video instead of taking a photo.
>
> Now, that is a good idea. I just hope my video camera provides enough
> fps to catch the message.

fps are usually good enough, not that fast. The problem is resolution to
read the characters.


Cheers / Saludos,

Carlos E. R.
(from 11.4 x86_64 “Celadon” at Telcontar)

I am going to try out on the weekend and let you know…I think it has something to do with my LVM setup (I have two LVM Groups, one encrypted, one not) or with Ivy bridge…but now I am guessing again.

Ok here it is:


Disabling swaps
Detaching loop devivces
Detaching DM devices
Could not delete dm /dev/dm-0: Device or resource busy
Not all DM devices detached, 2 left.
Detaching DM devices
Could not delete dm /dev/dm-0: Device or resource busy
Not all DM devices detached, 2 left.
Could not finalize remaining file systems and sevices, trying to kill remaining processes
Detaching DM devices
Could not delete dm /dev/dm-0: Device or resource busy
Not all DM devices detached, 2 left.
Cannot finalize remaining file systems and devices, giving up

And now my questions:

  1. How to find out what /dev/dm-0 is, and which process keeps it busy?
  2. I have an LVM setup with two Volume groups, one encrypted one not. I somehow think the problem might be linked to the way I have mounted them?

Hopefully, somebody reads till this point…

On 2012-07-15 10:16, Nikos78 wrote:

> And now my questions:
> 1. How to find out what /dev/dm-0 is, and which process keeps it busy?

dm is a raid device.

“lsof /dev/dm-0” might tell you what holds it.

> 2. I have an LVM setup with two Volume groups, one encrypted one not. I
> somehow think the problem might be linked to the way I have mounted
> them?

>
> Hopefully, somebody reads till this point…

:slight_smile:


Cheers / Saludos,

Carlos E. R.
(from 11.4 x86_64 “Celadon” at Telcontar)

dm is a raid device

I have not configured raid, so I think it is the LVM…

 lsof /dev/dm-0

gives me nothing if run as a user, and this:

lsof: WARNING: can't stat() fuse.gvfs-fuse-daemon file system /home/myname/.gvfs
      Output information may be incomplete.

if run as root.

On 2012-07-15 16:06, Nikos78 wrote:
>
>> dm is a raid device
> I have not configured raid, so I think it is the LVM…

Might be, dunno.

> Code:
> --------------------
> lsof: WARNING: can’t stat() fuse.gvfs-fuse-daemon file system /home/myname/.gvfs
> Output information may be incomplete.
>
> --------------------
>
>
> if run as root.

Ignore the warning, the important thing is that it doesn’t find any match on dm-0. In my
machine it doesn’t find a match either, and dm-0 is in use, so the command is anyway useless.


Cheers / Saludos,

Carlos E. R.
(from 11.4 x86_64 “Celadon” at Telcontar)

I found a probably similar problem described here:
Weblog for gpall - can’t deactivate volume group (LVM2 shutdown…, but no solution provided.

I guess the thing is, the shutdown script tries to unmount the LVM on which it is running. This would explain, why the encrypted LVM is not unmounted, but it does not explain the other one.

I cannot imagine being the only one using LVM with OpenSuse and having this problem.

On 2012-07-16 11:06, Nikos78 wrote:

> I cannot imagine being the only one using LVM with OpenSuse and having
> this problem.

Don’t know. I don’t use LVM, never liked/trusted it much. But I doubt that it is the cause of
the reboot instead of the shutdown. I would try posting in the openSUSE mail list, it is a
different group of people. As you are using tumbleweed, try the opnsuse-factory mail list.


Cheers / Saludos,

Carlos E. R.
(from 11.4 x86_64 “Celadon” at Telcontar)

After update to kernel 3.5.3 my problem was resolved (might also be another update though, as yesterday there were a lot of those in Tumbleweed). So I don’t really know what it was, but probably it had nothing to do with my LVM setup.

On 2012-08-31 08:36, Nikos78 wrote:
>
> After update to kernel 3.5.3 my problem was resolved (might also be
> another update though, as yesterday there were a lot of those in
> Tumbleweed). So I don’t really know what it was, but probably it had
> nothing to do with my LVM setup.

So after all it was a tumbleweed issue :slight_smile:


Cheers / Saludos,

Carlos E. R.
(from 11.4 x86_64 “Celadon” (Minas Tirith))

I would say it was a kernel issue…since it occurred first time on OS 12.1 (without any updates). I still think it had to do with the Ivy bridge Graphics which is still very fresh…but it works now, and that is good enough for me.

Interesting though, the last lines of my shutdown process remain the same. So…
…giving up
is the last thing I see from my OpenSuse every day

I am having the EXACT same issue of the system not shutting down completely in 12.2. It reboots instead. Same motherboard. No raid, no lvm, just a plain jane install.

I am having the EXACT same issue of the system not shutting down completely in 12.2. It reboots instead. Same motherboard. No raid, no lvm, just a plain jane install.

You need a kernel >3.5.2