Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.debian.devel > #107691
| From | Boian Bonev <bbonev@ipacct.com> |
|---|---|
| Newsgroups | linux.debian.devel |
| Subject | Re: Upgrade package from init script to Systemd, move config folder |
| Date | 2023-04-26 22:10 +0200 |
| Message-ID | <GoTCF-4J8E-7@gated-at.bofh.it> (permalink) |
| References | <GoTCF-4J8E-9@gated-at.bofh.it> <GmILn-3mGO-5@gated-at.bofh.it> <Go6BX-4dvH-1@gated-at.bofh.it> <GoOCZ-4FXw-7@gated-at.bofh.it> |
| Organization | linux.* mail to news gateway |
Hi, On Wed, 2023-04-26 at 16:25 +0200, Marc Haber wrote: > On Mon, 24 Apr 2023 15:48:02 +0100, Matthew Vernon > <matthew@debian.org> wrote: > > One thing I'd say is: please keep the init script in your package, > > so that people using inits other than systemd can continue to use > > it. > > I am not sure whether it is doing non-systemd users a favor to keep a > probably outdated, bitrotting and untested init script in the > canonical place. OTOH it is an un-favor to move or remove it, no matter if it is half- working or completely non-working. For a less experienced user it will be easier to fix a half-/non-working script in place than to look for it in obscure places, copy it from another distro or get it from a magic 'curl http... | sh' link. > My gut feeling is that it might be better to ship the old init script > in /usr/share/doc/package/examples unless the package maintainer is > reasonably sure that the init script will actually work. It is better to keep the init script in place. That will not affect systemd users in any way, besides <1k in disk space (same as with /usr/share/doc/package/examples). Also in case the maintainer is not interested in the init script, then the task for keeping the init script in good order should be left to those interested on a best effort basis. Who cares if a package ships a 100% bit-rotten init script? Only the people who use that un-popular init system. Then let them fix it. On the maintainer side - how much does it hurt to occasionally accept a patch for a small file? It would have been much easier if dpkg did allow to declare in package B that package A depends on B and in case A is removed then B should also go (to name it precisely, the relation is supplement)... Unlike sysvinit upstart is obsolete and removing upstart configuration is a good thing. But that is not the case with sysvinit. Removing the init scripts really hurts a group of people. Compared to Debian's user base that group is very very small. But tell me how small is a good number to be 'safely' ignored/neglected? 100, 1000 or 10000? With best regards, b.
Back to linux.debian.devel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
Upgrade package from init script to Systemd, move config folder Perry Naseck <mailinglists@perrynaseck.com> - 2023-04-20 22:10 +0200
Re: Upgrade package from init script to Systemd, move config folder Matthew Vernon <matthew@debian.org> - 2023-04-24 17:50 +0200
Re: Upgrade package from init script to Systemd, move config folder Marc Haber <mh+debian-devel@zugschlus.de> - 2023-04-26 16:50 +0200
Re: Upgrade package from init script to Systemd, move config folder Boian Bonev <bbonev@ipacct.com> - 2023-04-26 22:10 +0200
Re: Upgrade package from init script to Systemd, move config folder Paul Wise <pabs@debian.org> - 2023-04-27 03:30 +0200
Re: Upgrade package from init script to Systemd, move config folder Mark Hindley <mark@hindley.org.uk> - 2023-04-27 09:20 +0200
Re: Upgrade package from init script to Systemd, move config folder Simon Richter <sjr@debian.org> - 2023-04-27 08:40 +0200
Re: Upgrade package from init script to Systemd, move config folder Marc Haber <mh+debian-devel@zugschlus.de> - 2023-04-27 11:20 +0200
Re: Upgrade package from init script to Systemd, move config folder Peter Pentchev <roam@ringlet.net> - 2023-04-27 13:00 +0200
csiph-web