debootstrap maintainers, how do you want to solve this? I see three kind of solutions: - the udev postinst checks for [ -d /debootstrap ] and does not try to start the daemon if it exists - debootstrap --foreign creates /etc/udev/disabled and deletes it when it's done - debootstrap --foreign signals its existence to the packages being installed by e.g. exporting an environment variable On Apr 21, Marco d'Itri <md@Linux.IT> wrote: > On Apr 21, Junichi Uekawa <dancer@netfort.gr.jp> wrote: > > > then run on the generated chroot (OS image): > > > > /debootstrap/debootstrap --second-stage > > > > udev postinst will of course know it's a chroot if it's a chroot, but > > if it's running outside of chroot (like, inside qemu or natively), it > > will fail. > Why should daemons not be started if debootstrap is not being run in a > chroot? > Maybe that this should be fixed in deboostrap instead, i.e. it should > not disable s-s-d if it's not running in a chroot. > > > There are two ways to tackle the problem; the current problem at hand > > seems like using a fake start-stop-daemon to stop udevd and assuming > > udevd is stopped. > > > Why not use policy-rc.d / invoke-rc.d ? It sounds more generic. > invoke-rc.d cannot be used because udevd must be started without using > the init script. I don't know if using policy-rc.d would be possible > and/or appropriate in this case. > > -- > ciao, > Marco -- ciao, Marco
Attachment:
signature.asc
Description: Digital signature