Thorsten Glaser <firstname.lastname@example.org> (2014-11-23): > Cyril Brulebois dixit: > > >bad, because we're not going to change debootstrap, especially not at > >this late stage of the release cycle, to behave differently depending > >on the target distribution. The current script covers all suites from > > This is not necessary. Well, not at first. > > The changes to debootstrap I’d “love” to have are: > > ① add a new --variant called sysvinit, which excludes all the > systemd bits and explicitly adds sysvinit-core; this variant > will only be usable with jessie and up, but as it must be > invoked explicitly will not do anything to older releases. > > This is the change we’d need now. No we don't need that, for the reasons I listed, and the ones below. > ② change the scripts for --variant minbase, buildd, fakechroot, > scratchbox to exclude systemd, sysvinit (and, for good measure > on certain downstreams) upstart. This will not touch either > --variant sysvinit or base (the default). > > This is a change that could be put into experimental, as it > is not necessary for either the release or d-i. Then upload > that to sid after the unfreeze. Having init not pulled by debootstrap was mentioned recently but I'm not sure this led to a bug report tracking this. Might be worth investigating for jessie+1. > ③ To be fair, add --variant=upstart based on the sysvinit variant. > And openrc. But this all will end up in experimental, too. Once/if there's a way to have an initless system, people can install whatever init they want on top of it; having a flavour per init looks like a bad idea. > I’d actually be happy to try and hack these myself and submit > them for inclusion. Is the --exclude not working bug fixed by > now btw? Read the bug log? Mraw, KiBi.
Description: Digital signature