Re: [tlug] Um, so... systemd?

[Curt Sampson <cjs@example.com>]

Actually, some of the complaints about systemd show such ignorance as
to go beyond pathetic to amusing. Take for example [this page] by Nils
Dagsson Moskopp, where he complaints about code making assumptions
such as the "We don't know, so assume yes" bit in the
unix_socket_alive() function in src/tmpfiles/tmpfiles.c.

Well, as it turns out, that's checking /proc/net/unix to see if the
unix domain socket is listed there (meaning that a process is
definitely listening on it). If it can't read /proc/net/unix for
whatever reason, it cannot tell if the given socket is in use or not,
so the correct thing here to do is, yes, assume that it is alive
rather than removing it and hoping that nothing was listening on it.

Sheesh.

I knew that people had some complaints about systemd, and I think that
some of them are even valid for certain situations, but this is
absurd. And of course there's no way to comment on that page to stop
the spread of FUD.

[this page]: http://news.dieweltistgarnichtso.net/posts/systemd-assumptions-bullying-consent.html

cjs
-- 
Curt Sampson         <cjs@example.com>         +81 90 7737 2974

To iterate is human, to recurse divine.
    - L Peter Deutsch