On 11/01/2014 at 05:24 PM, Jonathan de Boyne Pollard wrote: > Dimitrios Chr. Ioannidis: > >> A quick search reveals the following. >> >> I've a software that use libuuid. Until now, the uuidd had the >> ability to start on-demand the uuidd if the later, quotting "... >> setuid to an unprivileged user (e.g. uuidd:uuidd)". >> >> After that commit, i'm forced to use systemd, if i don't want to >> start uuidd from the beginning. That's new for me... > > Aside from the security implications of doing things the old way, > which Christian Seiler touched upon, you are not forced to use > systemd. You are forced to use a service management system that is > capable of listening on AF_LOCAL sockets and starting up daemon > processes using the systemd socket-file-descriptor transfer protocol, > a subtly different thing.
And a perfectly fine one, unless you either think there are problems with that protocol (in which case you should probably report them for analysis and possible fixing) or think that socket activation is undesirable as itself. I'm not sure that some people don't think the latter, and if so that could help explain some of the objection to systemd even though systemd itself is at this point mostly optional - but at this point I think an ecosystem switch to socket activation s pretty much a done deal, given the current level (and, apparently, pace) of upstream adoption. -- The Wanderer The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man. -- George Bernard Shaw
signature.asc
Description: OpenPGP digital signature
