GNU bug report logs - #47439
27.2; In daemon mode, if after-init-hook errors out, the server does not start

Previous Next

Package: emacs;

Reported by: Gilles <gilles.usenet <at> gmail.com>

Date: Sun, 28 Mar 2021 01:02:02 UTC

Severity: normal

Found in version 27.2

Full log


View this message in rfc822 format

From: Eli Zaretskii <eliz <at> gnu.org>
To: Lars Ingebrigtsen <larsi <at> gnus.org>
Cc: gilles.usenet <at> gmail.com, 47439 <at> debbugs.gnu.org
Subject: bug#47439: 27.2; In daemon mode, if after-init-hook errors out, the server does not start
Date: Sun, 28 Mar 2021 17:55:31 +0300
> From: Lars Ingebrigtsen <larsi <at> gnus.org>
> Cc: gilles.usenet <at> gmail.com,  47439 <at> debbugs.gnu.org
> Date: Sun, 28 Mar 2021 16:43:48 +0200
> 
> I don't know why the doc string here specifies that there's no error
> handling of this hook in particular -- that's the default for hooks,
> right?
> 
> It seems that it first appeared in 1992:
> 
> +  "Functions to call after loading the init file (~/.emacs).
> +The call is not protected by a condition-case, so you can set `debug-on-error'
> +in .emacs, and put all the actual code on `after-init-hook'.")
> 
> Which is even more confusing -- it seems like the point here is that
> this is a way to run init code, but get debugging?  We now have a
> separate facility for that, so that bit doesn't seem very important any
> more.

It definitely predates the daemon. and I think it relates to what
"--debug-init" does.

> Anyway -- I think perhaps continuing on here and starting the daemon
> might be the most useful solution here, perhaps?

You mean, catching the error if we are starting the daemon? that's one
possibility, assuming that the error message will then be visible when
the first client connects.

In non-daemon mode I think we should keep not catching errors.




This bug report was last modified 4 years and 47 days ago.

Previous Next


GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.