Quoting Russ Allbery (2014-11-19 19:28:09) > Bob Proulx <bob@proulx.com> writes: > > > No. That is too late. By the time you are disabling something it has > > already been installed and started in postinst scripts. Using > > policy-rc.d is the only way to prevent unknown anythings from being > > enabled before installing. > > Ah, yes, that's true. > > > P.S. Related to this is that I really think that if you want a daemon > > running and install it and the package can configure it and start it > > then it should do so. If you don't want something running then don't > > install it. Or remove/purge it. I am not advocating any change that > > would get in the way of making an installation not start a daemon that > > it can do so. The chroot case really is a special case. > > Yes, the general Debian practice is to assume that, if you install a > daemon, you want the daemon running. If that's not the case and you don't > want it to start even temporarily, you're correct that policy.d is the > only mechanism to achieve that. Which implies, I believe, that any other way of starting daemons should also respect policy-rc.d if it can lead to automated triggering. Example: if a logrotate snippet uses "update-rc.d force-restart ..." then suppressing that deamon with policy-rc.d will be circumvented when cron triggers log rotation. This can easily go unnoticed, because our most common use of policy-rc.d (done also within debootstrap and cdebootstrap) suppresses _all_ daemons, including cron daemon. - Jonas -- * Jonas Smedegaard - idealist & Internet-arkitekt * Tlf.: +45 40843136 Website: http://dr.jones.dk/ [x] quote me freely [ ] ask before reusing [ ] keep private
Attachment:
signature.asc
Description: signature