Suspend/hibernate on systems that don't suspend/resume or hibernate properly

Michael Knepher mknepher at bluethingy.com
Thu Apr 7 00:30:23 UTC 2011


2011/4/6 "Jóhann B. Guðmundsson" <johannbg at gmail.com>:
> On 04/06/2011 10:56 PM, Michael Knepher wrote:
>> 2011/4/6 "Jóhann B. Guðmundsson"<johannbg at gmail.com>:
>>> <snip>
>>>
>>> All suspend resume bugs should be fixed hence if it fails for you or
>>> anyother reporter for that matter you should report it so please file a
>>> bug and attach /var/log/messages
>>> ,/var/log/pm-suspend.log and the file from su -c
>>> 'pm-utils-bugreport-info.sh>  pm-utils-bugreport.txt'
>>>
>>> You can test suspend from the graphical.target and multi-user.target by
>>> running
>>>
>>> su -c 'pm-suspend'
>>>
>>> and
>>>
>>> su -c 'echo mem>  /sys/power/state'
>> Am I supposed to run both or either of these commands? I ran the
>> first, with the same results (no screen), and after restarting the
>> system, attached the requested files to
>> https://bugzilla.redhat.com/show_bug.cgi?id=690648
>
> What happens if you add pci=noacpi to the kernel command line and try to
> suspend/resume?

After a normal boot, and logging in, and waiting for 5 minutes, I just
had the background and a cursor and a network connection (I have
synergy autostarting in my session). No top panel, no overlay. I
killed the session with ctrl+alt+backspace and logged into a clean
test user, which brought up the overlay after about 20 seconds, but
nothing sensitive to user interaction other than cursor movement, and
no user name in the user menu, and no network connection. So I was
never able to get to a point within a user session at which I could
suspend. I restarted with pci=noacpi and tried running suspend from
gdm. Same apparent outcome.

>
> you should also add the output from dmesg to that bug report ( su -c
> 'dmesg > dmesg.txt' )  given that this mostlikely is a kernel bug

Done. Would dmesg output following a boot with pci=noacpi appended be
useful as well?

>
> JBG
> --
> test mailing list
> test at lists.fedoraproject.org
> To unsubscribe:
> https://admin.fedoraproject.org/mailman/listinfo/test
>


More information about the test mailing list