Re: pkg/52861 (Adding an rc file to syncthing.)

[Greg Troxel <gdt%lexort.com@localhost>]

The following reply was made to PR pkg/52861; it has been noted by GNATS.

From: Greg Troxel <gdt%lexort.com@localhost>
To: nia%NetBSD.org@localhost
Cc: gdt%netbsd.org@localhost,  pkgsrc-bugs%netbsd.org@localhost,  gnats-admin%netbsd.org@localhost,  dziltener%lyrion.ch@localhost,  gnats-bugs%netbsd.org@localhost
Subject: Re: pkg/52861 (Adding an rc file to syncthing.)
Date: Sun, 07 Jul 2019 11:42:56 -0400

 nia%NetBSD.org@localhost writes:
 
 > Synopsis: Adding an rc file to syncthing.
 >
 > State-Changed-From-To: open->feedback
 > State-Changed-By: nia%NetBSD.org@localhost
 > State-Changed-When: Sun, 07 Jul 2019 15:25:00 +0000
 > State-Changed-Why:
 > I've added an rc script to the syncthing package already, not being aware
 > of this PR. Let me know what you think of it?
 
 It looks basically fine to me.
 
 (Part of the reason I didn't address this is that I never got around to
 it and kept forgetting.)
 
 One of the reasons I hesitated to add an rc.d script is that syncthing
 is not clearly a system daemon vs. a user daemon.   And if it's a user
 daemon, then there should be able to be an instance per user.   So a
 variable scheme that admits possibly many users is in order, something
 like
 
 syncthing_users="alice bob charlie"
 syncthing_home.alice = /home/alice
 [and so on]
 
 And this is a specific case of users starting up programs at boot, for
 which there should be a more general mechanism.  On NetBSD, @reboot in
 crontab might be that, or might not.  For pkgsrc in general with rc.d,
 crontab features don't address the issue.
 
 Certainly, what you've done works for one user, and what was there before
 worked for zero users, so it's a big improvement.
 
 I too am curious to hear what the original submitter thinks; if
 satisfactory we can certainly close the PR.