I think it is time to birn this one with dignity. Upstream is fairly dead. People are either using elogind or systemd's logind.
Referencing more abandonware.
elogind seems to not have any capability to manage cgroup2 controllers for users logging in (setting resource weights and limits, etc). For OpenRC users only consolekit2 has this capability. It still works fine with small tweaks located in various Gentoo bug reports.
The bug has been referenced in the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=2a942c610c80f27824927abb163750dbe1c84647 commit 2a942c610c80f27824927abb163750dbe1c84647 Author: Mikle Kolyada <zlogene@gentoo.org> AuthorDate: 2020-08-02 07:53:32 +0000 Commit: Mikle Kolyada <zlogene@gentoo.org> CommitDate: 2020-08-02 07:56:28 +0000 profiles/base: mask consolekit use flag Bug: https://bugs.gentoo.org/727730 Signed-off-by: Mikle Kolyada <zlogene@gentoo.org> profiles/base/use.mask | 4 ++++ 1 file changed, 4 insertions(+)
The bug has been referenced in the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=f08fdab4d59f570a46b698599bad206c4c48cd06 commit f08fdab4d59f570a46b698599bad206c4c48cd06 Author: Mikle Kolyada <zlogene@gentoo.org> AuthorDate: 2020-08-02 08:05:52 +0000 Commit: Mikle Kolyada <zlogene@gentoo.org> CommitDate: 2020-08-02 08:09:19 +0000 profiles: mask sys-auth/consolekit for removal Bug: https://bugs.gentoo.org/727730 Signed-off-by: Mikle Kolyada <zlogene@gentoo.org> profiles/package.mask | 8 ++++++++ 1 file changed, 8 insertions(+)
This really should have been package.deprecated and NOT removed from all packages, even stable packages, without a word.
(In reply to Adam Feldman from comment #5) > This really should have been package.deprecated and NOT removed from all > packages, even stable packages, without a word. +1. As I said in comment 2, elogind seems to be missing important cgroup2 functionality, so it is far from being a drop-in replacement.
Additionally, Bug #682196 is also still open. So basically from my understanding a SELinux enabled system will not be possible in the future. I will need to migrate off SELinux in order to use elogind, or keep a package which is not in portage anymore. Is this assumption true?
The bug has been referenced in the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=39ece3ff2467f0c1a3dc946767f3896d31055198 commit 39ece3ff2467f0c1a3dc946767f3896d31055198 Author: Mikle Kolyada <zlogene@gentoo.org> AuthorDate: 2020-10-04 14:33:57 +0000 Commit: Mikle Kolyada <zlogene@gentoo.org> CommitDate: 2020-10-04 14:39:06 +0000 sys-auth/consolekit: remove last-rited pkg Bug: https://bugs.gentoo.org/727730 Closes: https://bugs.gentoo.org/644550 Closes: https://bugs.gentoo.org/667912 Closes: https://bugs.gentoo.org/644548 Closes: https://bugs.gentoo.org/601878 Closes: https://bugs.gentoo.org/699592 Closes: https://bugs.gentoo.org/651788 Closes: https://bugs.gentoo.org/614658 Closes: https://bugs.gentoo.org/593306 Signed-off-by: Mikle Kolyada <zlogene@gentoo.org> sys-auth/consolekit/Manifest | 6 - sys-auth/consolekit/consolekit-0.4.6.ebuild | 125 ------------------- sys-auth/consolekit/consolekit-1.1.0-r1.ebuild | 120 ------------------- sys-auth/consolekit/consolekit-1.1.2.ebuild | 131 -------------------- sys-auth/consolekit/consolekit-1.2.0.ebuild | 132 --------------------- sys-auth/consolekit/consolekit-1.2.1.ebuild | 132 --------------------- sys-auth/consolekit/consolekit-9999.ebuild | 128 -------------------- sys-auth/consolekit/files/90-consolekit-3 | 32 ----- sys-auth/consolekit/files/consolekit-0.2.rc | 25 ---- .../files/consolekit-0.4.5-polkit-automagic.patch | 30 ----- sys-auth/consolekit/files/consolekit-1.0.0.initd | 16 --- ...ove-the-root-restriction-for-runtime-dirs.patch | 57 --------- .../files/consolekit-cleanup_console_tags.patch | 67 ----------- ...nsolekit-shutdown-reboot-without-policies.patch | 26 ---- .../files/consolekit-udev-acl-install_to_usr.patch | 28 ----- sys-auth/consolekit/files/pam-foreground-compat.ck | 17 --- sys-auth/consolekit/metadata.xml | 22 ---- 17 files changed, 1094 deletions(-)
The bug has been referenced in the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=167553ba3e35c92dda61e85fe3043b6e3aa67717 commit 167553ba3e35c92dda61e85fe3043b6e3aa67717 Author: Mikle Kolyada <zlogene@gentoo.org> AuthorDate: 2020-10-04 14:41:22 +0000 Commit: Mikle Kolyada <zlogene@gentoo.org> CommitDate: 2020-10-04 14:41:22 +0000 sec-policy/selinux-consolekit: remove last-rited pkg Bug: https://bugs.gentoo.org/727730 Signed-off-by: Mikle Kolyada <zlogene@gentoo.org> sec-policy/selinux-consolekit/Manifest | 4 ---- sec-policy/selinux-consolekit/metadata.xml | 8 -------- .../selinux-consolekit-2.20190201-r1.ebuild | 15 --------------- .../selinux-consolekit-2.20190609-r1.ebuild | 15 --------------- .../selinux-consolekit/selinux-consolekit-9999.ebuild | 15 --------------- 5 files changed, 57 deletions(-)
The bug has been closed via the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=66f8fb307e3b4765903b4010b9801ea37bcf8bd1 commit 66f8fb307e3b4765903b4010b9801ea37bcf8bd1 Author: Mikle Kolyada <zlogene@gentoo.org> AuthorDate: 2020-10-08 20:32:26 +0000 Commit: Mikle Kolyada <zlogene@gentoo.org> CommitDate: 2020-10-08 20:34:55 +0000 profiles/base/use.mask: remove consolekit use flag Closes: https://bugs.gentoo.org/727730 Signed-off-by: Mikle Kolyada <zlogene@gentoo.org> profiles/base/use.mask | 4 ---- 1 file changed, 4 deletions(-)
This was very underhanded and horrible. No word of warning, no deprecation, just one day: Bam, no consolekit in portage, all USE flags and packages for it were masked and removed. And to top it all off, when it was masked Portage was fighting me very hard with unmasking it. I actually left Gentoo for a while because of this. Fortunately for everyone affected by this disgusting mistake of a removal, I made an overlay containing ConsoleKit and PolicyKit. While I was at it, I helped fix some other terrible mistakes made by Gentoo maintainers with the help of a friend. https://github.com/Phate6660/undeprecated I wanted to message this when all this first happened, but life got in the way, hence the delayed response.
Thankfully we've got the timeline in this very bug to prove you wrong.
