2nd try: reboot hangs
 

On Mi, 28 mar 12, 09:26:55, Miles Fidelman wrote:
> Haven't heard a peep on this one. Anybody at least have a
> suggestion re. other places to ask this question (perhaps a
> kernel-related list?)

You missed http://lists.debian.org/4F72711D.8080206@gmail.com but I
would wait a few days anyway.

Kind regards,
Andrei
--
Offtopic discussions among Debian users and developers:
http://lists.alioth.debian.org/mailman/listinfo/d-community-offtopic

2nd try: reboot hangs
 

Andrei POPESCU wrote:

On Mi, 28 mar 12, 09:26:55, Miles Fidelman wrote:

Haven't heard a peep on this one. Anybody at least have a
suggestion re. other places to ask this question (perhaps a
kernel-related list?)

You missed http://lists.debian.org/4F72711D.8080206@gmail.com but I
would wait a few days anyway.



ooops... did, sorry...

--
In theory, there is no difference between theory and practice.
In practice, there is. .... Yogi Berra



--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

Archive: 4F7318ED.6070209@meetinghouse.net">http://lists.debian.org/4F7318ED.6070209@meetinghouse.net

2nd try: reboot hangs
 

Scott Ferguson<scott.ferguson.debian.user@gmail.com <mailto:scott.ferguson.debian.user%40gmail.com>> wrote

On 28/03/12 12:24, Miles Fidelman wrote:


> I have an old tower machine - a homebrew that I use as a sandbox. Just
> loaded it up with a copy of squeeze, and a xen installation - works
> fine, but...
> - shutdown -h works just fine
> - reboot goes down, gives the console message "restarting system", then
> hangs
>
> On an earlier installation (Lenny, x86 install) adding "reboot=bios"
> solved the problem. This time around - 64bit image (amd64), squeeze,
> none of the reboot=[....] parameters seem to make a difference. (I note
> that the Debian installer documentation indicates that reboot=bios only
> works with x86).
>
> Any suggestions? Any configuration things I should try to uncover that
> might help find a solution (I don't have paperwork on the box, anymore -
> so it's going to be a matter of running various probes to figure out
> what's on the motherboard and such - suggested commands welcome.)


I suspect you've already thought of this but... could it be an ACPI
problem? I use a number of older machines for development - some require
acpi=force to shutdown properly or they just hang at the last stage of halt.


didn't help, actually didn't expect it to, since "shutdown" works fine, reboot is the issue

but... just discovered a new symptom... SOMETIMES reboot works
(with reboot=force on the command line) - if I try reboot right after the system has come up,
then it hangs at the final step before it should restart, but if I wait a while, then the
system will sometimes reboot rather than hanging -- seems like something is interfering at a
very low level

come to think of it... maybe something funny is happening in the hypervisor (xen), ... hmmm





--
In theory, there is no difference between theory and practice.
In practice, there is. .... Yogi Berra



--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

Archive: 4F731E4C.8020007@meetinghouse.net">http://lists.debian.org/4F731E4C.8020007@meetinghouse.net

2nd try: reboot hangs
 

On 29/03/12 00:58, Miles Fidelman wrote:
> Andrei POPESCU wrote:
>> On Mi, 28 mar 12, 09:26:55, Miles Fidelman wrote:
>>> Haven't heard a peep on this one.


<snipped, there's an irony there somewhere>

Have you also tried both soft *and* hard reboots? (or should I make you
wait a few days? ;-p)

There is a list of things I'd check in BIOS and dmesg, but I'd start
with using another kernel, then looking for ACPI issues when calling
halt, *then* look at reboot. But that's just my approach to isolation
testing those sorts of issues, not necessarily the "right" way.


Kind regards

--
Iceweasel/Firefox/Chrome/Chromium/Iceape/IE extensions for finding
answers to Debian questions:-
https://addons.mozilla.org/en-US/firefox/collections/Scott_Ferguson/debian/


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: 4F73972F.3030501@gmail.com">http://lists.debian.org/4F73972F.3030501@gmail.com

2nd try: reboot hangs
 

Hi Scott,

Scott Ferguson wrote:

On 29/03/12 00:58, Miles Fidelman wrote:

Andrei POPESCU wrote:

On Mi, 28 mar 12, 09:26:55, Miles Fidelman wrote:

Haven't heard a peep on this one.


<snipped, there's an irony there somewhere>

Have you also tried both soft *and* hard reboots? (or should I make you
wait a few days? ;-p)


I've tried every option I could find for reboot=[warm, cold, bios, efi,
pci, force, .... (can't remember the list offhand)].


Also beginning to wonder if the problem is at the hypervisor level (I'm
running on Xen). Tried a bunch of variations there, too.

There is a list of things I'd check in BIOS and dmesg, but I'd start
with using another kernel, then looking for ACPI issues when calling
halt, *then* look at reboot. But that's just my approach to isolation
testing those sorts of issues, not necessarily the "right" way.
Tried setting logging to as verbose as I can get (both Dom0 and Xen),
and it keeps coming down to the console message "restarting system" and
then a hang. Not quite sure what to look at next.


Not quite sure I want to start digging through finding or building an
alternate kernel, what with having to maintain compatibility with Xen -
besides this is really starting to be more a matter of an itch to
scratch than something that really gets in the way of using the system.
I'm at the point of really wanting to isolate exactly the point at which
something isn't working right, rather than avoiding having to push the
reset button (it's not like this is one of the production servers
sitting in a remote data center).


Any suggestions re. how to isolate this would be very much appreciated.
In particular, I'm trying to figure out if there are some kernel or
hypervisor messages that I'm just not seeing.


Thanks Very Much,

Miles



--
In theory, there is no difference between theory and practice.
In practice, there is. .... Yogi Berra



--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

Archive: 4F739D39.6030504@meetinghouse.net">http://lists.debian.org/4F739D39.6030504@meetinghouse.net

2nd try: reboot hangs
 

On 29/03/12 09:56, Scott Ferguson wrote:
> On 29/03/12 00:58, Miles Fidelman wrote:
>> Andrei POPESCU wrote:
>>> On Mi, 28 mar 12, 09:26:55, Miles Fidelman wrote:
>>>> Haven't heard a peep on this one.
>
>
> <snipped, there's an irony there somewhere>
>
> Have you also tried both soft *and* hard reboots? (or should I make you
> wait a few days? ;-p)

Ooops - sorry, just remembered you're using xen so a hard reboot isn't
possible. Compare shutdown and destroy instead (hard and soft shutdowns).

>
> There is a list of things I'd check in BIOS and dmesg, but I'd start
> with using another kernel, then looking for ACPI issues when calling
> halt, *then* look at reboot. But that's just my approach to isolation
> testing those sorts of issues, not necessarily the "right" way.
>
>
> Kind regards
>


--
Iceweasel/Firefox/Chrome/Chromium/Iceape/IE extensions for finding
answers to Debian questions:-
https://addons.mozilla.org/en-US/firefox/collections/Scott_Ferguson/debian/


--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: 4F739D4D.7050606@gmail.com">http://lists.debian.org/4F739D4D.7050606@gmail.com

2nd try: reboot hangs
 

Scott Ferguson wrote:

On 29/03/12 09:56, Scott Ferguson wrote:

On 29/03/12 00:58, Miles Fidelman wrote:

Andrei POPESCU wrote:

On Mi, 28 mar 12, 09:26:55, Miles Fidelman wrote:

Haven't heard a peep on this one.


<snipped, there's an irony there somewhere>

Have you also tried both soft *and* hard reboots? (or should I make you
wait a few days? ;-p)

Ooops - sorry, just remembered you're using xen so a hard reboot isn't
possible. Compare shutdown and destroy instead (hard and soft shutdowns).


Not sure what you mean here. It's not the DomUs that I'm having
problems with, it's the Dom0 kernel and the hypervisor, which should be
able to go down to a hard reboot. Are there additional boot options for
either the Dom0 or hypervisor lines /etc/default/grub?


Miles


--
In theory, there is no difference between theory and practice.
In practice, there is. .... Yogi Berra



--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

Archive: 4F739E90.8030605@meetinghouse.net">http://lists.debian.org/4F739E90.8030605@meetinghouse.net

2nd try: reboot hangs
 

Miles Fidelman wrote:
> ... since "shutdown" works fine, reboot is the issue

I recently had a very similar problem on a new Intel Atom
motherboard. An update of the BIOS fixed it. Reboot issues were
listed in the changes for that particular model. Perhaps that is also
the problem that you are having?

Bob

2nd try: reboot hangs
 

Bob Proulx wrote:

Miles Fidelman wrote:

... since "shutdown" works fine, reboot is the issue

I recently had a very similar problem on a new Intel Atom
motherboard. An update of the BIOS fixed it. Reboot issues were
listed in the changes for that particular model. Perhaps that is also
the problem that you are having?



Not sure I want to go to the hassle (and risk) of flashing a new
motherboard ROM - it's not like I can't reboot this machine by doing a
shutdown (which works) and then hitting the power button. If it were
one of my production servers, sitting in a remote data center, that
would be one thing, but this is just a sandbox for playing with new
tools, and it sits under my desk.


It's more that this has become an itch that I really want to scratch.
It all worked just fine with previous releases of Debian (Lenny, 32
bit), and Xen. With Sarge, a 64bit kernel, and the latest Xen package,
it doesn't -- and the "reboot=bios" command line option (which is what
was working previously, suggesting that the bios reboot code works just
fine) is documented as applying only to the x86 kernel.


So... it kind of bugs me... somewhere, at the very last step of the
reboot process, after everything has been shut down, and either the
kernel, or the xen hypervisor sends "restarting now" to the console, either:
- something hangs just before issuing the jump to machine instruction
that starts the reboot process, or,
- or the instruction executed to kick off rebooting isn't one that works
on my machine, or,

- the startup code isn't really there, or,
- the startup code doesn't work on my particular machine (i.e., the
wrong code is set up).


I kind of want to isolate this down to the specific cause. Again, more
our of sheer stubbornness at this point.


Any suggestions on:
- very low level configuration items to fiddle with,
- adding even more detailed logging to the very low level code,
- or kicking into single step or debug mode down at the very low level,
- or something else?

Sigh....

Miles



--
In theory, there is no difference between theory and practice.
In practice, there is. .... Yogi Berra



--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org

Archive: 4F75A7A3.4030803@meetinghouse.net">http://lists.debian.org/4F75A7A3.4030803@meetinghouse.net

2nd try: reboot hangs
 

On Vi, 30 mar 12, 08:31:31, Miles Fidelman wrote:
>
> Any suggestions on:
> - very low level configuration items to fiddle with,
> - adding even more detailed logging to the very low level code,
> - or kicking into single step or debug mode down at the very low level,
> - or something else?

Just to eliminate any Xen issues:

- stop all virtual machines before reboot
- try with the normal kernel (i.e. non-Xen)

Kind regards,
Andrei
--
Offtopic discussions among Debian users and developers:
http://lists.alioth.debian.org/mailman/listinfo/d-community-offtopic