resume failed for the 2nd time since I upgraded to F18

[Julian Sikorski]

W dniu 15.12.2012 20:38, Adam Williamson pisze:
> On Sat, 2012-12-15 at 10:02 +0100, Julian Sikorski wrote:
>> Hi,
>>
>> since I upgraded to F18, resume from S3 has failed for me 2 times
>> already. It worked many more times, but still it is a bit concerning,
>> and also makes it way harder to pinpoint. All I can see in the logs is this:
>>
>> Dec 15 04:19:55 snowball2 systemd-logind[753]: Lid closed.
>> Dec 15 04:19:55 snowball2 systemd-logind[753]: Suspending...
>> Dec 15 04:19:57 snowball2 systemd[1]: SELinux Got Sender :1.0
>> Dec 15 04:19:57 snowball2 systemd[1]: Starting Sleep.
>> Dec 15 04:19:57 snowball2 systemd[1]: Reached target Sleep.
>> Dec 15 04:19:57 snowball2 systemd[1]: Starting Suspend...
>> Dec 15 04:19:57 snowball2 systemd-sleep[8033]:
>> /usr/lib/systemd/system-sleep/notify-upower.sh exited with exit status 1.
>> Dec 15 04:19:57 snowball2 systemd-sleep[8033]: Suspending system...
>>
>> When I tried to resume the machine in the morning, all I could see was a
>> blinking cursor. Where should I look in order to debug this problem and
>> investigate what is going on?
> 
> Suspend issues are rarely Fedora-specific, they're almost always
> graphics driver or kernel related, both of which we follow upstream
> closely on. You'll likely get more useful info on debugging suspend from
> some kind of kernel development/help place.
> 
OK, I finally managed to catch the problem with pm_trace enabled:
[julas at snowball2 ~]$ dmesg | grep hash\ matches
[    3.440566]   hash matches drivers/base/power/main.c:575
[    3.440585] pci 0000:03:00.0: hash matches
[julas at snowball2 ~]$ lspci | grep 03:00.0
03:00.0 Ethernet controller: JMicron Technology Corp. JMC250 PCI Express
Gigabit Ethernet Controller (rev 05)

How do I proceed?

Regards,
Julian