service stop & rcorder

[fluca1978]

I noticed that service(1) performs a raw exec(1) of the script passed as its first argument using the verb as argument to such script, therefore something like

service blah stop

results in

exec env -i HOME=/ PATH=/sbin:/bin:/usr/sbin:/usr/bin /somewhere/blah stop

This is asymmetric with the boot/startup process, that is driven by rcorder(1) and is based on the REQUIRE/PROVIDE annotations. Therefore if I, for instance, do the stop of a daemon in my stack that provides services to other daemons, other daemons are not notified, correct? Is there a reason I cannot see for this behaviour or another tool to use instead of service(1) to handle intra-daemon state changes?

 

[SirDice]

Therefore if I, for instance, do the stop of a daemon in my stack that provides services to other daemons, other daemons are not notified, correct?

Correct.

Is there a reason I cannot see for this behaviour or another tool to use instead of service(1) to handle intra-daemon state changes?

I'm not sure if that functionality is there but it does have other nice features. I'm hoping this will someday make it to the base.

sysutils/fsc