On 28-Nov-2001 Miquel van Smoorenburg wrote: > According to Sean 'Shaleh' Perry: >> > If I'm not mistaken that is not nessecary unless we plan to move >> > all .deb archives over to .lsb too, which is not going to happen. >> > Debian will stay Debian we just need to make it possible to install >> > .lsb files *as well* >> >> we can support the installation of lsb binaries HOWEVER the lsb spec adds a >> 'status' option to init scripts which lsb packages may expect to exist. So >> at >> the bare minimum we need to support that. > > Well no, packages in .lsb that have an /etc/init.d/initscript must > support the 'status' option but Debian packages don't have to do > that as they are Debian packages and not .LSB packages. >
right, but if you want to INSTALL and USE a lsb package that package is fully within its rights to expect /etc/init.d/apache status to work (not to pick on apache). >> If we actually want to call Debian an LSB system that involves the work I >> mentioned above. > > There's probably lots to do for LSB compliance but adding a status > option to every /etc/init.d/* script is probably not part of it. > > Disclaimer: > If I'm wrong then the LSB spec is even more stupid than I thought... > actually the spec calls for RH style shell functions in the inits instead of our method of start-stop-daemon. Yes as with all specs it is the worst of all possible worlds but the things all of them were willing to do.
