Sam Hartman <hartm...@debian.org> writes: > my position is that socket activation is a bad choice for network > services where the primary user of the socket is non-local. The issue > is that inherently socket-activation requires the socket configuration > to live in systemd.
> That involves splitting configuration between systemd and the service. > I have the information on where to listen in systemd and the information > on how to configure the sockets in a service specific manner still in > the service's configuration. That's actually one of the things that I like about it. From a systems administration perspective, I would far rather configure all of the sockets in one place with one syntax rather than fighting with obscure flags and configuration formats to configure this separately for every service in some unique, special snowflake way. That's particularly true given that most services do not support properly configuring sockets. For example, I usually can't set IPv6 freebind, I often can't even configure three bind addresses but not all addresses, I often don't have control over whether IPv6 gets IPv4-mapped addresses, and so forth. However, that said, the MIT Kerberos KDC may be a special case here, since, now that I think about it, it has to do dynamic UDP rebinding as interfaces are added and removed from the system, which is a poor fit for socket activation (for the same reason that ntpd probably doesn't benefit that much from socket activation). But, regardless, I think slapd ideally should support socket activation, which would resolve this issue since it would mean that there's no longer an ordering dependency between the services. -- Russ Allbery (r...@debian.org) <http://www.eyrie.org/~eagle/> -- To UNSUBSCRIBE, email to debian-bugs-dist-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org
