New major release soon?

[joakim-tjernlund]

241.4 is getting old, is there a 243.x on the horizon?

[Yamakuzure]

Yes it is. I am currently testing and cleaning up the sources from unused functions.Guess I'll push tomorrow, you can clone from v243-stable branch and test, too, if you like.

[joakim-tjernlund]

Yes it is. I am currently testing and cleaning up the sources from unused functions.Guess I'll push tomorrow, you can clone from v243-stable branch and test, too, if you like.

Would be much easier if there was en 9999 ebuild in Gentoo, could you push one?

[Yamakuzure]

There is one in my Overlay "seden", available via Layman.I am not a Gentoo dev, just a proxy maintainer for a few packages. (not elogind) 

[Yamakuzure]

I just pushed. Travis is no longer working, but that I have to sort out later. (They changed some default package in the void linux image we use.)

[joakim-tjernlund]

I tried your elogind-243.9999 ebuild but didn't work out:

1. the elogind-241.1-docs.patch does not apply
2. building fails with:

o' -MF 'elogind-uaccess-command@exe/src_uaccess-command_uaccess-command.c.o.d' -o 'elogind-uaccess-command@exe/src_uaccess-command_uaccess-command.c.o' -c ../elogind-243.9999/src/uaccess-command/uaccess-command.c
../elogind-243.9999/src/uaccess-command/uaccess-command.c: In function ‘dev_uaccess’:
../elogind-243.9999/src/uaccess-command/uaccess-command.c:44:9: error: implicit declaration of function ‘umask’; did you mean ‘sigmask’? [-Werror=implicit-function-declaration]
   44 |         umask(0022);
      |         ^~~~~
      |         sigmask
../elogind-243.9999/src/uaccess-command/uaccess-command.c:44:9: warning: nested extern declaration of ‘umask’ [-Wnested-externs]
cc1: some warnings being treated as errors

[joakim-tjernlund]

This little patch makes it build:

--- src/uaccess-command/uaccess-command.c.org	2019-12-29 22:18:46.849371284 +0100
+++ src/uaccess-command/uaccess-command.c	2019-12-29 22:19:40.636321047 +0100
@@ -22,6 +22,8 @@
 #include <errno.h>
 #include <stdio.h>
 #include <stdlib.h>
+#include <sys/types.h>
+#include <sys/stat.h>
 
 #include "sd-login.h"

w.r.t elogind-241.1-docs.patch, could you not integrate that into the elogind src?

[Yamakuzure]

Sorry for the late reply.

I have just published: https://github.com/elogind/elogind/releases/tag/v243.4

The elogind-243.9999 should work just fine until the update hits the portage tree.

The reason for the long time is my stupidity leading to a typo that lead to an rm -rf /. Luckily as a user, but thanks to the wonders of nvme, the short two seconds until I slammed CTRL-C was enough to delete half of the containers I have my stuff in and all of my backups.

So after restoring enough to test and get this release out, my current main project is:
https://github.com/Yamakuzure/xfs_undelete

I hope I get my stuff back...

[joakim-tjernlund]

Sorry for the late reply.

I have just published: https://github.com/elogind/elogind/releases/tag/v243.4

The elogind-243.9999 should work just fine until the update hits the portage tree.

Great !

The reason for the long time is my stupidity leading to a typo that lead to an rm -rf /. Luckily as a user, but thanks to the wonders of nvme, the short two seconds until I slammed CTRL-C was enough to delete half of the containers I have my stuff in and all of my backups.

So after restoring enough to test and get this release out, my current main project is:
https://github.com/Yamakuzure/xfs_undelete

I hope I get my stuff back...

Ouch, been there myself too.
I haven't looked into you xfs code, just wanted to mention http://inai.de/projects/hxtools/
There is some xfs recover code in there. There is also https://github.com/salviati/xfsr
Good luck

[joakim-tjernlund]

BTW, the tag, v243.4 seem to be missing.
My little patch above is also missing I think?

[Yamakuzure]

I had to retract the release due to some heavy issues on travis.

And good it was, because some more issues popped up all of a sudden. I'll re-release when everything is ready. Hopefully tomorrow or on Wednesday.

[Yamakuzure]

My little patch above is also missing I think?

It should be fixed now.

[joakim-tjernlund]

My little patch above is also missing I think?

It should be fixed now.

Still don't see it.

[Yamakuzure]

My little patch above is also missing I think?

It should be fixed now.

Still don't see it.

I need a little bit more info. I am using elogind-243.9999 for a few weeks now, and everything is working well, built with gcc-9.2 against glibc-2.30. No build failures so far, so I don't know why I should add includes when they don't solve a standing problem.

Do you build still fails with elogind-243.9999 ?

[joakim-tjernlund]

umask(2) has these includes documeted:

       #include <sys/types.h>
       #include <sys/stat.h>

       mode_t umask(mode_t mask);

Omitting them may work depending on glibc version which by accident pull in these

Have you fixed the two Gentoo patches elogind-241.4-broken-test.patch and elogind-241.4-nodocs.patch ?
They don't apply any more.

[joakim-tjernlund]

umask(2) has these includes documeted:

       #include <sys/types.h>
       #include <sys/stat.h>

       mode_t umask(mode_t mask);

Omitting them may work depending on glibc version which by accident pull in these

Especially since elogind defaults to -Werror
-Werror is nice for development work but a pain for distributions

[joakim-tjernlund]

Still breaks form me at home:
gcc-9.2.0-r2 and glibc-2.29-r7

[Yamakuzure]

Now I get it....

[perfect7gentleman]

 ~ $ loginctl --version
elogind 243.4 (243.4)
+PAM -AUDIT -SELINUX +SMACK +UTMP +ACL default-hierarchy=hybrid

[Yamakuzure]

@perfect7gentleman That's also my experience.
And yet we have another shred of evidence that installations can be different without seeming to be so... 🤷‍♂️

That's why I need your feedback, guys, so thank you very much!

[Yamakuzure]

@joakim-tjernlund Done!
https://github.com/elogind/elogind/releases/tag/v243.4
Finally! 😀