headers
Chris Baker | 6 Jul 2002 20:11

latest ssh incompatible with CLX?


I'm running debian unstable, and I recently upgraded to the latest
ssh.  Unfortunately, it looks like CLX no longer works over ssh via X
forwarding.  Has anyone else encountered this problem?  (Debugging
output below).

TIA,

cbb

cbb <at> clever-hans:~$ ssh -v -X localhost
OpenSSH_3.4p1 Debian 1:3.4p1-2, SSH protocols 1.5/2.0, OpenSSL 0x0090604f
debug1: Reading configuration data /etc/ssh/ssh_config
debug1: Rhosts Authentication disabled, originating port will not be trusted.
debug1: ssh_connect: needpriv 0
debug1: Connecting to localhost [127.0.0.1] port 22.
debug1: Connection established.
debug1: identity file /home/cbb/.ssh/identity type 0
debug1: identity file /home/cbb/.ssh/id_rsa type -1
debug1: identity file /home/cbb/.ssh/id_dsa type -1
debug1: Remote protocol version 1.99, remote software version OpenSSH_3.4p1 Debian 1:3.4p1-2
debug1: match: OpenSSH_3.4p1 Debian 1:3.4p1-2 pat OpenSSH*
Enabling compatibility mode for protocol 2.0
debug1: Local version string SSH-2.0-OpenSSH_3.4p1 Debian 1:3.4p1-2
debug1: SSH2_MSG_KEXINIT sent
debug1: SSH2_MSG_KEXINIT received
debug1: kex: server->client aes128-cbc hmac-md5 none
debug1: kex: client->server aes128-cbc hmac-md5 none
debug1: SSH2_MSG_KEX_DH_GEX_REQUEST sent
debug1: expecting SSH2_MSG_KEX_DH_GEX_GROUP
(Continue reading)

Permalink | Reply | 3 comments |
headers
Eric Marsden | 11 Jul 2002 17:18
Picon
Picon
Favicon

Re: latest ssh incompatible with CLX?


>>>>> "cb" == Chris Baker <electro <at> 1939worldsfair.com> writes:

  cb> I'm running debian unstable, and I recently upgraded to the latest
  cb> ssh.  Unfortunately, it looks like CLX no longer works over ssh via X
  cb> forwarding.  Has anyone else encountered this problem?  (Debugging
  cb> output below).

I can reproduce the problem with a recent ssh. I suspect that it is
because the format of the key in the $HOME/.Xauthority file has
changed: its format now looks like

   hostname/unix:13 MIT-MAGIC-COOKIE-1 abcdefetc

whereas I believe that it used to use localhost rather than hostname
(use "xauth list" to see this). I shall see whether I can fix the
problem.

--

-- 
Eric Marsden                          <URL:http://www.laas.fr/~emarsden/>
Permalink | Reply | 2 comments |
headers
Eric Marsden | 11 Jul 2002 19:24
Picon
Picon
Favicon

Re: latest ssh incompatible with CLX?


>>>>> "ecm" == Eric Marsden <emarsden <at> laas.fr> writes:
>>>>> "cb" == Chris Baker <electro <at> 1939worldsfair.com> writes:

  cb> I'm running debian unstable, and I recently upgraded to the latest
  cb> ssh.  Unfortunately, it looks like CLX no longer works over ssh via X
  cb> forwarding.  Has anyone else encountered this problem?  (Debugging
  cb> output below).

actually, this is the same problem that Scott Fahlman reported here a
few weeks ago. Try adding the following to your .cmucl-init file,
after having loaded CLX.

(in-package "XLIB")

(defun get-best-authorization (host display protocol)
  (labels ((read-short (stream &optional (eof-errorp t))
	     (let ((high-byte (read-byte stream eof-errorp)))
	       (and high-byte
		    (dpb high-byte (byte 8 8) (read-byte stream)))))
	   (read-short-length-string (stream)
	     (let ((length (read-short stream)))
	       (let ((string (make-string length)))
		 (dotimes (k length)
		   (setf (schar string k) (card8->char (read-byte stream))))
		 string)))
	   (read-short-length-vector (stream)
	     (let ((length (read-short stream)))
	       (let ((vector (make-array length :element-type '(unsigned-byte 8))))
		 (dotimes (k length)
(Continue reading)

Permalink | Reply | 1 comment |
headers
Chris Baker | 12 Jul 2002 01:42

Re: latest ssh incompatible with CLX?


Eric Marsden <emarsden <at> laas.fr> writes:

> >>>>> "ecm" == Eric Marsden <emarsden <at> laas.fr> writes:
> >>>>> "cb" == Chris Baker <electro <at> 1939worldsfair.com> writes:
> 
>   cb> I'm running debian unstable, and I recently upgraded to the latest
>   cb> ssh.  Unfortunately, it looks like CLX no longer works over ssh via X
>   cb> forwarding.  Has anyone else encountered this problem?  (Debugging
>   cb> output below).
> 
> actually, this is the same problem that Scott Fahlman reported here a
> few weeks ago. Try adding the following to your .cmucl-init file,
> after having loaded CLX.

[snip code to fix problem]

Thanks!  That worked like a charm.  I saw that earlier thread but
tuned it out since I thought it was a Solaris problem (and it was
somewhat over my head :)

Thanks again,

cbb

--

-- 
  4:30pm  up 119 days, 19:19,  2 users,  load average: 0.28, 0.11, 0.03
Permalink | Reply | 0 comments |
headers
Lilo | 12 Jul 2002 05:15

Help with "logical-host"


What exactly is a logical-host?  I'm trying to run Closure[1] under
CMUCL and I get this error:

CMU Common Lisp 18d, running on afx
Send questions to cmucl-help <at> cons.org. and bug reports to
cmucl-imp <at> cons.org.
Loaded subsystems:
    Python 1.0, target Intel x86
    CLOS based on PCL version:  September 16 92 PCL (f)
    CLX X Library MIT R5.02
    Motif toolkit and graphical debugger 1.0
    Gray Streams Protocol Support
* (require :clue)
[...]
(load "load.lisp")
[...]

File-error in function COMMON-LISP::FIND-LOGICAL-HOST:
   Logical host not yet defined: ""

Thanks,
Lilo

1. http://www.uni-karlsruhe.de/~unk6/closure/
--

-- 
http://fastmail.fm - Sent .0000002 seconds ago
Permalink | Reply | 0 comments |
headers
Lilo | 12 Jul 2002 05:26

Help with logical host.


What exactly is a logical-host?  I'm trying to run Closure[1] under
CMUCL and I get this error:

CMU Common Lisp 18d, running on afx
Send questions to cmucl-help <at> cons.org. and bug reports to
cmucl-imp <at> cons.org.
Loaded subsystems:
    Python 1.0, target Intel x86
    CLOS based on PCL version:  September 16 92 PCL (f)
    CLX X Library MIT R5.02
    Motif toolkit and graphical debugger 1.0
    Gray Streams Protocol Support
* (require :clue)
[...]
(load "load.lisp")
[...]
(gui::init-closure)
File-error in function COMMON-LISP::FIND-LOGICAL-HOST:
   Logical host not yet defined: ""

Thanks,
Lilo
--

-- 
Access all of your messages and folders wherever you are! 
http://fastmail.fm - Get your mail using the web or your email software
Permalink | Reply | 0 comments |
headers
Fred Gilham | 12 Jul 2002 21:27
Favicon

undumpable constant


What does the `referencing an undumpable constant' error mean?

I'm trying to recompile the Ergo Lisp system and I'm stuck on this
problem.  The error comes when compiling the following file:

;;; -*- Mode: Lisp; Package: SB-RUNTIME -*-
(in-package "SB-RUNTIME")  (use-package :ergolisp)

(use-package '("OPER" "OCC" "TERM" "SORT" "LANG"))

(defparameter *sbrt-sort-table*
  (make-sort-table
   '((#t(:sort id) . #t(:union))
     (#t(:sort cid) . #t(:union))
     (#t(:sort number) . #t(:union))
     (#t(:sort string) . #t(:union))
     (#t(:sort literal) . #t(:union))
     (#t(:sort keyword) . #t(:union)))))

I don't know if the problem is due to make-sort-table creating a hash
table (which it does) or to it using the #t read macros.  Or if
there's some other problem.

--

-- 
Fred Gilham                                      gilham <at> csl.sri.com
[Some of the] Top ten reasons why the God of Jesus Christ makes a
better God than `Caesar':
7. God doesn't circulate worthless chunks of metal or paper with `In
Caesar we trust' written on them.
(Continue reading)

Permalink | Reply | 1 comment |
headers
Martin Cracauer | 12 Jul 2002 22:03

Re: undumpable constant


Fred Gilham wrote on Fri, Jul 12, 2002 at 12:27:59PM -0700: 
> 
> What does the `referencing an undumpable constant' error mean?
> 
> I'm trying to recompile the Ergo Lisp system and I'm stuck on this
> problem.  The error comes when compiling the following file:
> 
> ;;; -*- Mode: Lisp; Package: SB-RUNTIME -*-
> (in-package "SB-RUNTIME")  (use-package :ergolisp)
> 
> (use-package '("OPER" "OCC" "TERM" "SORT" "LANG"))
> 
> (defparameter *sbrt-sort-table*
>   (make-sort-table
>    '((#t(:sort id) . #t(:union))
>      (#t(:sort cid) . #t(:union))
>      (#t(:sort number) . #t(:union))
>      (#t(:sort string) . #t(:union))
>      (#t(:sort literal) . #t(:union))
>      (#t(:sort keyword) . #t(:union)))))

Can you post the source for make-sort-table?

> I don't know if the problem is due to make-sort-table creating a hash
> table (which it does) or to it using the #t read macros.  Or if
> there's some other problem.

I usually get these in third-party code when a function takes a (then
higher-order) function as an argument and a default funtion is
(Continue reading)

Permalink | Reply | 0 comments |
headers
Martin Cracauer | 12 Jul 2002 22:32

Re: undumpable constant


Fred Gilham wrote on Fri, Jul 12, 2002 at 01:22:20PM -0700: 
> 
> > Can you post the source for make-sort-table?
> 
> 
> (defvar *global-sort-table* nil
>   "The global sort-table")
> 
> (defun make-sort-table (&optional (contents ()))
>   "Build a sort-table.  Contents is a list of oper,opsig pairs to be inserted
>    in the table."
>   (let ((result (make-hash-table :test #'eq)))
>     (mapc #'(lambda (entry)
> 	      (let ((key (car entry))
> 		    (value (cdr entry)))
> 		(declare (type ttype value))
> 		(assert (is-sort-ttype key) ()
> 			"Key value must be a sort: ~S" key)
> 		(sort-table-insert key value result))) ; here
> 	  contents)
>     result))

I don't really see why, but I am sure the operation part of the
contents (let-variable "key") ends up trying to put the literal
function into the hastable like I indicated in my previous message.

Try this:
 		(sort-table-insert (function key) value result))) ; here
(Continue reading)

Permalink | Reply | 0 comments |
headers
Rolf Wester | 15 Jul 2002 11:35
Picon
Picon

Graphical debugger


Hi,

I would like to use CMUCL's graphical debugger, but I couldn't figure
out how to set breakpoints, step through functions etc. Is there any 
documentation on this (the CMUCL manual only covers the tty debugger
interface) or could anybody give me some advice?

I would be very appreciative for help.

Regards

Rolf Wester 
-------------------------------------
Rolf Wester
rolf.wester <at> ilt.fraunhofer.de
Permalink | Reply | 0 comments |

Gmane