SIGPWR sent to init, and /etc/inittab

randyding · 03-26-2005, 11:59 PM

The /etc/inittab has this line to define init's behavior on receiving SIGPWR
pf:

owerfail:/sbin/shutdown -f -h +2 "Power Failure; System Shutting Down"

Then it has this line
pr:12345

owerokwait:/sbin/shutdown -c "Power Restored; Shutdown Cancelled"

But what signal (or other) actually invokes this ?

itsme86 · 03-27-2005, 12:05 AM

Check out man 7 signal:

Code:

SIGPWR      29,30,19    Term    Power failure (System V)

The signal is sent by the kernel. If you have the correct cable hooked up to your UPS, it will tell the kernel that there's been a power failure.

randyding · 03-27-2005, 12:29 AM

I may be mistaken about this, and doing it totally wrong, but I thought SIGPWR was actually sent to PID #1 by powerd, a daemon that monitors the serial port CTS line (* goes active on power failure *).
Right now I wrote and installed my own powerd that does exactly this. When CTS goes active, I do in C code...

Code:

kill(1,SIGPWR);

Then if CTS goes inactive (before shutdown kills my daemon) I do this in C code...

Code:

system("/sbin/shutdown -c \"Power restored.\"");

My system() function call here bypasses the action defined in the /etc/inittab for power-restore.

Is my powerd doing this incorrectly then ?

itsme86 · 03-27-2005, 12:34 AM

Oh, I see. You're trying to find out how to tell init that the power has been restored...I'm not quite sure how that's done...hmm...

itsme86 · 03-27-2005, 12:40 AM

I think this is the information you're looking for:

Code:

# powerwait – The process will be executed when init receives the SIGPWR signal. Init will wait for the process to finish before continuing.
# powerfail – Same as powerwait but init does not wait for the process to complete.
# powerokwait – The process will be executed when init receives the SIGPWR signal provided there is a file called "/etc/powerstatus" containing the word "OK".

So it looks like when your program sees that power is okay, it should write OK to /etc/powerstatus and then send SIGPWR to init again.

randyding · 03-27-2005, 12:53 AM

Yes, thanks. All other code I found via google did it the way I showed,
which didn't feel right.
I'll have to experiment and dig into it some more then. Its not clear
at first glance when the /etc/powerstatus file has to be unlinked.
I'm assuming prior to the next power failure at a minimum.
Thanks again...