nina
looks fine, i'm not very familiar with typescript though...
I guess a proper implementation should also do credentials checking and have only *one* socket that is created by dinit on start? Doing this is simpler, but also all sockets...
my main issue with using plain old sockets is the same, i.e. permissions protect them but in the end *any* process running as the same user is allowed to poke...
the way systemd gets around it is that it checks the process credentials that come with the datagram, but we can't do that, because the allowed processes should include any...
thinking about this more, if i end up doing this at all (which i'm not 100% sure about yet), i will do what you said - i.e. use regular unix...
I updated the PR to implement things like above. Documentation still pending, but you can review what is there...
a wrapper is unreliable regardless, the process may never signal readiness and if an approach like dbus-wait-for is taken you'll end up with a forked off child process that will...
I also wonder if i should add functionality for setting scheduling and i/o priorities while at it, and perhaps oom-killer score adjustment (they are useful to have especially for services...
this is what the setting of the 3 values would look like without the parsing (the input values for `nice` would go integer -20 to 19, for `ionice` it would...
I implemented everything. What's left now is basically 1) documentation 2) testing 3) deciding how to guard the features; whether we should have a separate feature for each mainly 4)...