
Originally Posted by
Deamos
I am having the same issues with Jaunty. Suspend drops to a single cursor blink.
Maybe it has something to do with Video Card.
I am using the Restricted Nvidia 180.00 Drivers
I've had problems with waking up from suspend all the life of this machine, but last time I managed to find a piece of info to try and find out what was the cause. I've synthesized the procedure below:
remove /etc/init.d/[checkroot.sh | checkfs.sh]
echo 1 > /sys/power/pm_trace
echo mem > /sys/power/state -> suspend
[supposedly failed attempt -> hard reboot with power button]
-> date will be screwed, as well as filesystem checks
save dmesg somewhere
restore /etc/init.d/[checkroot.sh | checkfs.sh]
In the dmesg file saved before, look for:
[ 1.500776] Magic number: 4:256:725
[ 1.500851] pci 0000:01:00.0: hash matches
compare trailing numbers with lspci output to know which device was
culprit (last to be attempted to suspend or resume)
In this case, the fault was due to...
01:00.0 VGA compatible controller: nVidia Corporation Quadro NVS 140M (rev a1)
Yeah, last time it was the nvidia driver alright. In 8.10 I had frozen nvidia do 173 because of other problems, and a few days before 9.04 came out, the update manager offered me 180 again and I decided to give it a try. Previous problems seemed to be gone, but the machine wouldn't wake up anymore (blank screen with flashing cursor only, no disk activity).
Now I'm 2--3 days on 9.04 and already noticed the waking up problem (100% of the attempts). I'll either try newer versions (185) or get back to good old 173.
If you really want to be sure about the cause, do the procedure above. The procedure I found through googling didn't mention about removing the checkfs init script, which is essential. Ignore eventual msgs about filesystem problems; what is vital is to reboot and collect the dmesg info in a few minutes time (<5min) after the suspend event, or else the clock digits in the RTC will increase the counting and overwrite the needed info.
Bookmarks