virtual/init-0.ebuild depends on >=sys-apps/sysvinit-2.86-r6 or sys-process/runit. But sys-apps/systemd-sysv-utils is also another init program. Reproducible: Always
Created attachment 320994 [details] init-0.ebuild
Comment on attachment 320994 [details] init-0.ebuild --- init-0.ebuild 2012-05-16 18:55:44.000000000 +0200 +++ - 2012-08-12 16:42:16.448176182 +0200 @@ -12,7 +12,7 @@ IUSE="" RDEPEND="!prefix? ( - kernel_linux? ( || ( >=sys-apps/sysvinit-2.86-r6 sys-process/runit ) ) + kernel_linux? ( || ( >=sys-apps/sysvinit-2.86-r6 sys-process/runit sys-apps/systemd-sysv-utils ) ) kernel_FreeBSD? ( sys-freebsd/freebsd-sbin ) )" DEPEND=""
And now with the whitespace done correctly: --- init-0.ebuild 16 May 2012 07:40:58 -0000 1.16 +++ init-0.ebuild 12 Aug 2012 14:43:50 -0000 @@ -12,7 +12,13 @@ IUSE="" RDEPEND="!prefix? ( - kernel_linux? ( || ( >=sys-apps/sysvinit-2.86-r6 sys-process/runit ) ) + kernel_linux? ( + || ( + >=sys-apps/sysvinit-2.86-r6 + sys-process/runit + sys-apps/systemd-sysv-utils + ) + ) kernel_FreeBSD? ( sys-freebsd/freebsd-sbin ) )" DEPEND="" Also: profiles/package.mask: # Michał Górny <mgorny@gentoo.org> (21 Jan 2012) # Blocks sysvinit yet doesn't provide all tools provided by it. # Masking until we get the necessary tools out of sysvinit. sys-apps/systemd-sysv-utils
(In reply to comment #3) can't tell if your indentation is off or bugzilla mangled the comment. either way, i'd leave it up to the systemd guys to pick out the value for this.
Is that right that we are waiting for (closed for some reason) bug 399615? Any temporary workaround for those who want to get `reboot` etc. symlinks to `systemctl`?
(In reply to comment #5) > Is that right that we are waiting for (closed for some reason) bug 399615? Yes, mostly that. > Any temporary workaround for those who want to get `reboot` etc. symlinks to > `systemctl`? Well, you can just unmask the package and use it. Just note that you'd have a few tools missing, like wall, mesg, last, fstab-decode, killall5, pidof.
Missing them makes me sad, especially `pidof`. I think I'll just remove intefering binaries from sysvinit evuild in my local overlay for now. Any chances to speed up the proccess? Are they refusing to split those tools out into, I don't know, sysvinit-tools (like Fedora does)?
(In reply to comment #7) > Missing them makes me sad, especially `pidof`. I think I'll just remove > intefering binaries from sysvinit evuild in my local overlay for now. > > Any chances to speed up the proccess? Are they refusing to split those tools > out into, I don't know, sysvinit-tools (like Fedora does)? You can try to reopen the bug, and give a big 'ping' there. But the maintainer mostly doesn't care.
Yeah, I was going to do that…
Seems that I can't reopen it, I hope just pinging will be enough…
Is killall5/pidof really needed for unmasking sys-apps/systemd-sysv-utils?
BTW, as a temporary solution, systemd-sysv-utils can provide killall5/pidof symlinks to busybox. :D
(In reply to Michał Górny from comment #6) > Well, you can just unmask the package and use it. Just note that you'd have > a few tools missing, like wall, mesg, last, fstab-decode, killall5, pidof. wall [0] mesg [0] last [0] fstab-decode [0] killall5 [0] pidof [sys-process/procps (/bin/pidof)] [0] not available So, first one fixed, 5 to go, I'll try search Arch forums :)
wall [util-linux --enable-wall] mesg [util-linux USE="+tty-helpers"] last [util-linux --enable-last ? ] fstab-decode [0] killall5 [0]
(In reply to David Heidelberger (okias) from comment #14) > wall [util-linux --enable-wall] > mesg [util-linux USE="+tty-helpers"] Both wall and mesg are provided by util-linux[tty-helpers] (>=2.23.2-r2) > last [util-linux --enable-last ? ] bug 491742
There is no such thing as virtual/init in Portage, nor there should be. Adjusting $summary accordingly.
So, are we all set to go?
(In reply to Kirill Elagin from comment #17) > So, are we all set to go? looks like it, unless killall5 and/or fstab-decode is required by something and I wouldn't worry about them too much before someone files bugs about them missing, and that won't be happening before this beast is actually released to ~arch i'll let systemd maintainers unmask this.
Bumped and unmasked. Since there's no virtual/init anymore, I guess that's all.
AMAZING! THANK YOU GUYS.
