How to reboot after a S3 wake-up?

jbjbjb

One way might be, to define a key combination from remote control as initiator for a reboot, e.g. create a file".lircrc" in home of vdr-user, e.g. "/home/vdr/.lircrc":

begin
	prog   = irexec
	button = Stop
	button = Stop
	button = Stop
	button = Stop
	config = /sbin/reboot
end

Put "/sbin/reboot" into "/etc/sudoers"

#/> EDITOR=vi sudo -E visudo
...
vdr ALL=NOPASSWD:/sbin/reboot
<BLANK LINE>

Put this to the end of the file and don't forget the very last line has to be a blank one.

Then define anywhere an automatic start with the environment of vdr-user of, e.g. "/home/vdr/.config/openbox/autostart.sh" if you use Openbox:

/usr/bin/irexec -d /home/vdr/.lircrc &

With this example the VDR box will be reboot after pressing the Stop key 4x. Four times to prevent reboot by accident and the stop key, because it is used rarely and does not disturb the normal VDR operation.

Regards
fnu

Why would you want to do that ? Then you can configure normal shutdown or the fake-S4 in case you need that for remote wakeup.

You can see from "cat /proc/acpi/wakeup" what is supported.

Hi,

may be you'd like to ask e-tobi - afaik he has a working vdr with s3-acpi-wakeup.

I tried the same with an AT3N7A and had to give up. The acpi-events worked only with the testscript.
Using the vdrs shutdown-wrapper no acpi-event rised on wakeup - so there was no trigger to work with.
After all I changed vdr MB to a gigabyte wich works fine with acpi.

kind regards

Gero

Zitat

Original von jbjbjb
Can it work and reboot the system after wake-up from S3?

It is possible, but you risk that the next fsck will hit you soon and your recording starts too late. I don't understand why you don't follow steffen_b's advise and try fake-s4?

Gerald

Zitat

Original von gda
I don't understand why you don't follow steffen_b's advise and try fake-s4?
Gerald

Maybe because he has more processes running than you think and does not want to ruin data.
From my point of view doing a s2disk and afterwards simply booting is highly dangerous for your data not to mention consistency.
Just imaging a user runs a vm under yavdr - the way yavdr simply drops all processes is almost guaranteed to kill the filesystem on the vm-disk.
Or think about a local rsync process to backup your data to another harddrive. With working s2ram or s2disk this is no problem, your data will be back upped completely regardless how many suspends the pc does in between while your fake-S4 might simply stop the backup procedure without even telling you.

And please don't get me wrong, I think that yavdr is a great distribution, but you should make it
a) more obvious for users that you do not make a real s2disk in the web-menu and warn about the potential dataloss your fake-S4 might include
b) make a second entry for real S4 so the user has a choice to choose this way

I can't even see a real speed advantage of not resuming, it takes only 3 more seconds to come back to a working vdr from real S4 than from fake-S4 - and that is on a plain yavdr. As soon as more processes run (like a vm) this dramatically changes to real S4 beeing much faster than fake-S4. And btw, the 3 seconds come from VDR + Frontend starting up after resume, showing a live picture for about half a second and than restarting the frontend - showing the shutdown animation and coming back up again.
If you ask me I will always trade 3 seconds per vdr resume for data security.

Anyway S4 will not work for him. USB Wakeup is only working in S3 for him. Yes the script mentioned should work. I still dont get why you would wont that, but that doesnt matter, its not my computer its yours , so you can do whatever you want

geronimo: acpi-wakeup is working in yavdr for machines for that its working. We are talking about usb wakeup i believe.

Egalus: The fake-S4 has been done for crappy hardware which can not be get to work otherwise. So most likely for this hardware the S4 will not work fully. But I DO agree that the web interface should be more obvious on what its doing and that its only last resort.

Zitat

Original von jbjbjb
I only want that, after reboot from S3, the RAM and the processes of the computer, remain as if it was a clean startup

Okay, but why do you want this?

Gerald

Zitat

Original von jbjbjb

Well, I may be wrong, but I think that the system will work better from a clean startup than from a previous not clean state. Please, tell me if I am saying a nonsense!!!

How do you define better? It works, or it works not. If parts of the system do not work after S3 you have to take measures to let them work after resume. This is not Windows, where you have difficulties to see what is happening before suspend and after resume. To reboot without specific reasons is indeed nonsense.

Gerald