On Wed, Jul 14, 2010 at 10:19 AM, Michael Albinus <[email protected]>wrote:

> Terrence Brannon <[email protected]> writes:
>
> > 09:31:13.331959 tramp-send-command (6) # tramp_perl_file_attributes
> /root/ integer 2>/dev/null; echo tramp_exit_status $?
> > 09:31:13.471961 tramp-wait-for-regexp (6) #
> > tramp_exit_status 255
> > ///4e226762a89429956796e848e1d0647a#$
>
> Unfortunately, this doesn't show us what happens. Could you, please,
> change in `tramp-send-command-and-check' the line
>
>           (if command (if dont-suppress-err "; " " 2>/dev/null; ") "")
>
> to
>
>           (if command (if dont-suppress-err "; " " 2>/tmp/123; ") "")
>
> and try to investigate, what error has been written?
>

(defun tramp-send-command-and-check
  (vec command &optional subshell dont-suppress-err)
  "Run COMMAND and check its exit status.
Sends `echo $?' along with the COMMAND for checking the exit status.  If
COMMAND is nil, just sends `echo $?'.  Returns the exit status found.

If the optional argument SUBSHELL is non-nil, the command is
executed in a subshell, ie surrounded by parentheses.  If
DONT-SUPPRESS-ERR is non-nil, stderr won't be sent to /dev/null."
  (tramp-send-command
   vec
   (concat (if subshell "( " "")
   command
   (if command (if dont-suppress-err "; " " 2>/tmp/123; ") "")
   "echo tramp_exit_status $?"
   (if subshell " )" "")))
  (with-current-buffer (tramp-get-connection-buffer vec)
    (goto-char (point-max))
    (unless (re-search-backward "tramp_exit_status [0-9]+" nil t)
      (tramp-error
       vec 'file-error "Couldn't find exit status of `%s'" command))
    (skip-chars-forward "^ ")
    (prog1
     (read (current-buffer))
     (let (buffer-read-only) (delete-region (match-beginning 0)
(point-max))))))




>
> Thanks, and best regards, Michael.
>


It was a zero-length file ...

-- 
Search! Search! [DuckDuckGo](http://duckduckgo.com/?t=terrencebrannon)
_______________________________________________
Tramp-devel mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/tramp-devel

Reply via email to