Server issues

Server issues

[Miguel Guedes]

Two seconds after emacs starts (server-start) is executed and I can then 
issue,

  emacsclient -n /path/to/file

 and emacs visits the specified file.

However, if I visit any /su::/path/to/file, using emacsclient to visit 
files from the command line no longer works.

Is this a bug or am I doing something wrong?

Re: Server issues

[W. Greenhouse]

Hi Miguel,

Miguel Guedes <miguel.a.guedes-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> writes:

> Two seconds after emacs starts (server-start) is executed and I can then 
> issue,
>
>   emacsclient -n /path/to/file
>
>  and emacs visits the specified file.
>
> However, if I visit any /su::/path/to/file, using emacsclient to visit 
> files from the command line no longer works.
>
> Is this a bug or am I doing something wrong?
>

This still works fine for me ( GNU Emacs 24.2.1 (i486-pc-linux-gnu, GTK+
Version 3.4.2) of 2013-01-22 on biber, modified by Debian ).

-Will
-- 
BOFH excuse #414:

tachyon emissions overloading the system

Re: Server issues

[Miguel Guedes]

Hi Will,

On Wed, 20 Mar 2013 20:52:33 +0000, W. Greenhouse wrote:

> This still works fine for me ( GNU Emacs 24.2.1 (i486-pc-linux-gnu, GTK+
> Version 3.4.2) of 2013-01-22 on biber, modified by Debian ).

I was secretly hoping someone would say that as it means it can be fixed 
without having to look (too deeply) under the hood! I would imagine 
something in my emacs conf files are loading is causing this issue 
perhaps by interfering with the code in server.el .  

Do you (or anyone else!) know how I could go about diagnosing this?

Many thanks for you reply.


* BTW, also using GNU Emacs 24.2.1

Re: Server issues

[W. Greenhouse]

Hi Miguel,

> I was secretly hoping someone would say that as it means it can be fixed 
> without having to look (too deeply) under the hood! I would imagine 
> something in my emacs conf files are loading is causing this issue 
> perhaps by interfering with the code in server.el .  
>
> Do you (or anyone else!) know how I could go about diagnosing this?
>
> Many thanks for you reply.
>
>
> * BTW, also using GNU Emacs 24.2.1

Where in the process of raising an emacsclient and using TRAMP to access
a file via su do things go wrong for you?  Is Emacs itself still able to
access /su::/path type filenames, or is that broken, too?

M-x toggle-debug-on-error does work for diagnosing many server errors,
so try that at first.  Next, the usual litany of the Emacs process of
elimination: try connecting to an Emacs server which was initialized
with emacs -q (skipping your own init file), and if that still has
trouble, escalate to emacs -Q (skipping the distro/site-wide init files
too).

FWIW, some of the most irritating and hard-to-diagnose emacsclient
invocation problems I've had have had to do with the behavior of the
*shell* rather than of Emacs (i.e. the shell interpreting things when we
want it to just pass the args to Emacs).  So experiment with different
shell quoting strategies too, especially if you are calling this inside
a shell script.

I noticed also in this list that you were writing some code to warn on
the modeline when su-editing a file; I don't see how that might've
contributed to your troubles, but maybe...

Sorry this is such general advice, but I can't think of much else that's
special about the su backend to Tramp.

Best,
Will

-- 
BOFH excuse #289:

Interference between the keyboard and the chair.