Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.debian.user > #272704
| From | Greg Wooledge <greg@wooledge.org> |
|---|---|
| Newsgroups | linux.debian.user |
| Subject | Re: emacs service and session start |
| Date | 2024-08-30 14:20 +0200 |
| Message-ID | <Jh8LE-8I5G-5@gated-at.bofh.it> (permalink) |
| References | (5 earlier) <JfGeJ-7KDF-11@gated-at.bofh.it> <JfKBH-7NME-5@gated-at.bofh.it> <JgGml-8oTT-5@gated-at.bofh.it> <JgP6h-8vdR-1@gated-at.bofh.it> <Jh8iC-8HFO-7@gated-at.bofh.it> |
| Organization | linux.* mail to news gateway |
On Fri, Aug 30, 2024 at 13:45:43 +0200, Erwan David wrote:
> The after/wants does not work (starnge since ssh-agent.service seems
> to see the SSH_AUTH_SOCK variable.
>
> But /etc/X11/Xsession.d/90x11-common_ssh-agent is a configuration file
> and we can add SSHARGS. I added a
> SSHARGS="-a $XDG_RUNTIME_DIR/ssh-agent"
>
> Then I added an override to emacs.service
> Environment=SSH_AUTH_SOCK=%t/ssh-agent
>
> And it works
It's amazing how badly the systemd folks managed to break *everything*.
I'm pretty old-fashioned. I use startx from a console login, and I
configure it with a ~/.xsession file which overrides the Debian stuff.
In my .xsession file I have:
hash ssh-agent 2>/dev/null && eval "$(ssh-agent -s)"
hash ssh-add 2>/dev/null && ssh-add </dev/null
This launches an agent, and evaluates the output (variables defining
where the socket lives), exactly the way it's supposed to be done since
decades before systemd existed.
I invoke my window manager later in the .xsession file, so it inherits
the ssh-agent environment variables. All of my X11 programs are children
of the window manager (or of this .xsession file), so *they* all inherit
the variables as well.
None of that "child of dbus" or "child of systemctl --user" bullshit
for me, thank you very much.
Back to linux.debian.user | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
emacs service and session start Erwan David <erwan@rail.eu.org> - 2024-08-23 10:10 +0200
Re: emacs service and session start didier gaumet <didier.gaumet@gmail.com> - 2024-08-23 11:20 +0200
Re: emacs service and session start Erwan David <erwan@rail.eu.org> - 2024-08-23 11:40 +0200
Re: emacs service and session start Erwan David <erwan@rail.eu.org> - 2024-08-23 18:20 +0200
Re: emacs service and session start Max Nikulin <manikulin@gmail.com> - 2024-08-23 18:40 +0200
Re: emacs service and session start Max Nikulin <manikulin@gmail.com> - 2024-08-25 05:20 +0200
Re: emacs service and session start Erwan David <erwan@rail.eu.org> - 2024-08-26 10:10 +0200
Re: emacs service and session start Erwan David <erwan@rail.eu.org> - 2024-08-26 13:40 +0200
Re: emacs service and session start Max Nikulin <manikulin@gmail.com> - 2024-08-26 18:20 +0200
Re: emacs service and session start Erwan David <erwan@rail.eu.org> - 2024-08-29 08:00 +0200
Re: emacs service and session start Max Nikulin <manikulin@gmail.com> - 2024-08-29 17:20 +0200
Re: emacs service and session start Erwan David <erwan@rail.eu.org> - 2024-08-30 13:50 +0200
Re: emacs service and session start Greg Wooledge <greg@wooledge.org> - 2024-08-30 14:20 +0200
Re: emacs service and session start Max Nikulin <manikulin@gmail.com> - 2024-08-30 16:50 +0200
Re: emacs service and session start Max Nikulin <manikulin@gmail.com> - 2024-08-30 14:40 +0200
Re: emacs service and session start Erwan David <erwan@rail.eu.org> - 2024-08-30 15:30 +0200
csiph-web