Am 30.07.2016 um 20:47 schrieb Michael Biebl: > Am 30.07.2016 um 19:52 schrieb Felipe Sateler: >> On 29 July 2016 at 17:05, Michael Biebl <bi...@debian.org> wrote: >>> Am 29.07.2016 um 22:44 schrieb Michael Biebl: >>>> Package: init-system-helpers >>>> Version: 1.41 >>>> Severity: serious >>>> >>>> Running >>>> update-rc.d foo disable >>>> update-rc.d foo defaults >>>> will create start symlinks although this should not happen for a >>>> disabled service. >>> >>> This happens when insserv is not installed, so is another regression in >>> our fallback code. >> >> Insserv does not create new links if there are already-existing links >> for that runlevel. So I think the fix (untested) would be:
... > I think what insserv does, is to compare the LSB header to what symlinks > exist on the disk. If there is a difference, it does nothing at all. > Your patch behaves slightly different from what I can see. > It does the check per runlevel and not as a whole. Just noticed this after upgrading keyboard-setup: $ ls /etc/rc?.d/???keyboard* /etc/rcS.d/S01keyboard-setup.sh /etc/rcS.d/S02keyboard-setup.sh S02keyboard-setup.sh was the existing symlink which had been created by insserv. S01keyboard-setup.sh was created by todays update by the fallback code in update-rc.d. It's a related but slightly different bug. Your patch would handle also handle this case. So we should add it in any case. The question is, if we also add the additional check, which does nothing at all, if the symlinks on disk differ from what's in the LSB header. Michael -- Why is it that all of the instruments seeking intelligent life in the universe are pointed away from Earth?
signature.asc
Description: OpenPGP digital signature
