tramp-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: tramp (2.2.3-24.1); Using tramp to run ess-remote session. Submittin


From: Tobias Muhlhofer
Subject: Re: tramp (2.2.3-24.1); Using tramp to run ess-remote session. Submitting more than one line of code at a time, causes emacs to wait indefinitely.
Date: Fri, 15 Feb 2013 13:02:35 -0500

Michael:

One more thing. I poked around a little bit more and found that a *R*
buffer (which is what ESS uses to run R sessions) is actually
established. Its contents are:

-----------------------------------------
env: R: No such file or directory

Process R exited abnormally with code 127 at Fri Feb 15 12:48:11 2013
-----------------------------------------

This seems interesting. The contents of the *ESS* buffer are:

---------------------------------------
[ess-site.el]: ess-customize-alist=nil
[ess-site.el _2_]: ess-customize-alist=nil
(S): ess-s-versions-create making M-x defuns for

(R): ess-r-versions-create making M-x defuns for


(R): ess-dialect=nil, buf=*GNU Emacs*, start-arg=nil
 current-prefix-arg=nil
(inferior-ess 0): ess-start-args=--no-readline
(inf-ess 1): lang=nil, dialect=nil, tmp-dialect=R, buf=*GNU Emacs*
(inf-ess 1.1): procname=R temp-dialect=R, buf-name=*R*
(inferior-ess) Method #3
start=/scpc:address@hidden:/N/dc/scratch/tmuhlhof/ buf=*R*
(ess-setq-vars-LOCAL): language=S, dialect=R, buf=nil,
comint..echoes=t, comint..sender=comint-simple-send
(inf-ess 2.1): ess-language=S, ess-dialect=R buf=*R*
(inf-ess 2.2): start args = --no-readline  , inf-ess-start-args=--no-readline
(inf-ess finish [S(R), R(nil)]
(ess-multi 0):  inf-ess-start-args=--no-readline  , comint-..echoes=t
(ess-setq-vars-LOCAL): language=S, dialect=R, buf=nil,
comint..echoes=t, comint..sender=inferior-ess-input-sender
(i-ess 1): buf=*R*, lang=S, comint..echo=nil,
comint..sender=inferior-ess-input-sender,
(i-ess end): buf=*R*, lang=S, comint..echo=nil,
comint..sender=inferior-ess-input-sender,
(ess-multi post inf-ess: start-args=--no-readline  , comint-echoes=nil
(ess-multi 1):  start-args=--no-readline
Making Process...Buf *R*, :Proc R, :Prog R
 :Args= --no-readline
Start File=nil
(ess-multi 2): waiting for process to start (before hook)

--------------------------------------------------

That one, I can't make any sense of.

But I think I understand what's going on. That particular remote
server requires a script to be executed upon login that puts the R
executable into the user's PATH. [They use the Modules package to
manage the user's environment.] That script doesn't seem to be
executed. Tried a different remote server that doesn't require this,
and I don't have that problem there at all. I can start an R remote
process there without a problem. And, lo and behold, like you, I can
feed multiple lines to that process through C-c M-r, like you.  I bet
that's the solution to that particular puzzle.

So, let me know about how you want to proceed with this one, as well
as the original issue about using M-x shell to start the R process.

Thanks!
   Toby

On Fri, Feb 15, 2013 at 12:45 PM, Tobias Muhlhofer
<address@hidden> wrote:
> Hi, Michael:
>
> Sorry about that. Here is the debug tramp buffer. Same steps as before
>
>> 1) Started emacs.
>> 2) In the *scratch* buffer that comes up, did M-x R, and selected a
>> tramp path as the starting directory (/<username>@<server>:/<path>).
>> 3) I was asked for my password in the minibuffer.
>> 4) Emacs hung until I did C-g.
>
> By the way, between steps 3 and 4, tramp's auto-complete actually
> worked to lead me to the starting directory that I wanted for R on the
> remote machine.
>
> Anyway, here is the output:
>
> -----------------------------------------
>
> ///bb9194a24d6cf50f33e1507790424290#$
> 12:35:20.145068 tramp-send-command (6) # echo \"`tty`\" 2>/dev/null;
> echo tramp_exit_status $?
> 12:35:20.157275 tramp-wait-for-regexp (6) #
> "/dev/pts/19"
> tramp_exit_status 0
> ///bb9194a24d6cf50f33e1507790424290#$
> 12:35:20.157807 tramp-open-connection-setup-interactive-shell (5) #
> Setting default environment
> 12:35:20.158113 tramp-send-command (6) #
> HISTFILE=$HOME/.tramp_history; export HISTFILE
> 12:35:20.168098 tramp-wait-for-regexp (6) #
> ///bb9194a24d6cf50f33e1507790424290#$
> 12:35:20.168644 tramp-send-command (6) # HISTSIZE=1; export HISTSIZE
> 12:35:20.178041 tramp-wait-for-regexp (6) #
> ///bb9194a24d6cf50f33e1507790424290#$
> 12:35:20.178589 tramp-send-command (6) # LC_ALL=C; export LC_ALL
> 12:35:20.188198 tramp-wait-for-regexp (6) #
> ///bb9194a24d6cf50f33e1507790424290#$
> 12:35:20.188692 tramp-send-command (6) # TERM=dumb; export TERM
> 12:35:20.199076 tramp-wait-for-regexp (6) #
> ///bb9194a24d6cf50f33e1507790424290#$
> 12:35:20.199629 tramp-send-command (6) # EMACS=t; export EMACS
> 12:35:20.209075 tramp-wait-for-regexp (6) #
> ///bb9194a24d6cf50f33e1507790424290#$
> 12:35:20.209627 tramp-send-command (6) #
> INSIDE_EMACS='24.1.1,tramp:2.2.3-24.1'; export INSIDE_EMACS
> 12:35:20.219084 tramp-wait-for-regexp (6) #
> ///bb9194a24d6cf50f33e1507790424290#$
> 12:35:20.219658 tramp-send-command (6) # PAGER=""; export PAGER
> 12:35:20.229111 tramp-wait-for-regexp (6) #
> ///bb9194a24d6cf50f33e1507790424290#$
> 12:35:20.229472 tramp-send-command (6) # unset correct autocorrect
> MAILPATH MAILCHECK MAIL HISTORY CDPATH
> 12:35:20.238889 tramp-wait-for-regexp (6) #
> ///bb9194a24d6cf50f33e1507790424290#$
> 12:35:20.239632 tramp-maybe-open-connection (3) # Opening connection
> for address@hidden using scpc...done
> 12:35:20.239807 tramp-send-command (6) # cd
> /N/dc/scratch/tmuhlhof/<redacted>/; exec env
> PS1=/scpc\:address@hidden:\ \#\$\  R --no-readline
>
> ---------------------------------------------------
>
> I have redacted the actual path to help a little with security.
>
> One guess (by someone who doesn't quite understand everything that's
> going on here): would the expression 'exec env
> PS1=/scpc\:address@hidden:\ \#\$\', imply that the only
> two available prompt characters are "#" and "$"? That would present a
> problem, since the prompt character in R's interpreter is ">". But
> maybe that means something else.
>
> Thanks!
>    Toby
>
>
>
> --
>            Dr. Tobias Mühlhofer
>            Assistant Professor
>            Department of Finance
>            Kelley School of Business
>            Indiana University
>            +1-812-855-9270
>            http://tobias.muhlhofer.com



--
           Dr. Tobias Mühlhofer
           Assistant Professor
           Department of Finance
           Kelley School of Business
           Indiana University
           +1-812-855-9270
           http://tobias.muhlhofer.com



reply via email to

[Prev in Thread] Current Thread [Next in Thread]