Hi Am 25.01.2018 um 07:53 schrieb Niels Thykier: > I think we should do it in a separate compat bump. With where things > are headed, I think I got enough changes to do a compat 12 as well in > buster.
ok > Can we just have " [ -d /run/systemd/system ] " be the gate both ways? > Is there any case, where that would do the wrong thing? > > I.e. > > # dh_installinit inserts: > if [ ! -d /run/systemd/system ] && ... ; then > ... > fi > > # dh_installsystem inserts: > if [ -d /run/systemd/system ] && ... ; then > ... > fi > > This means that dh_installinit's snippet is run when we are not running > systemd and dh_installsystemd's snippet is run when we are running > systemd. It is simple to write and is simple to explain, so if that > does the right thing, then it feels like the right approach. We still do have the case of packages, which only ship a SysV init script but no native .service file (yet). With a simplistic check like above, we'd have to change dh_installsystemd to start those as well. Somehow I would prefer if dh_installsystemd would only take care of real, native .service files and leave SysV only services to dh_installinit/invoke-rc.d. That's why I added the additional ! [ -e /lib/systemd/system/foo.service ] check. It would be up to the package maintainer to ensure that for a package shipping both, to either use the same name or add a symlink. Atm we try to detect during build time, whether such a init script and matching service file exists and generate code conditionally. This is tricky, as you need to make sure the service files or such a symlink is installed at the time when dh_installinit resp. dh_installsystemd is run. By always adding this check, and letting this be detected during runtime, we avoid this issue. 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
