The exact error message is:
(stream/warning) failure to open network stream to host "localhost" for service "4005"
Took a look at the changelog--Luke, something to confess?
The relevant init.el part:
;; start up slime (global-set-key [(f5)] '(lambda () (interactive) (shell-command "c:/programme/acl62-demo/allegro-ansi.exe +B +cm -L ~/slime.lisp&") (sleep-for 3) (slime-connect "localhost" 4005) (cd "~/Eigene Dateien/")))
and ~/slime.lisp is:
(load "~/Eigene Dateien/devel/lisp/slime/swank-loader.lisp") (swank::create-swank-server 4005 :spawn #'swank::simple-announce-function t)
Thomas Schilling tjs_ng@yahoo.de writes:
The exact error message is:
(stream/warning) failure to open network stream to host "localhost" for service "4005"
Took a look at the changelog--Luke, something to confess?
The relevant init.el part:
;; start up slime (global-set-key [(f5)] '(lambda () (interactive) (shell-command "c:/programme/acl62-demo/allegro-ansi.exe +B +cm -L ~/slime.lisp&") (sleep-for 3) (slime-connect "localhost" 4005) (cd "~/Eigene Dateien/")))
Is it just the obvious race condition? That function assumes that Lisp will have the server up and running within 3 seconds. If it's not up in time then you'd get the error you respoted. Note that if the Lisp sources have changed (e.g. after cvs update) then swank-loader will recompile them, and that could easily take more than 3 seconds.
The smallest fix would be to change the connecting part to something like:
(while (not (ignore-errors (slime-connect "localhost" 4005) t)) (sit-for 1))
.. to loop until the connection succeeds.
Am Thu, 22 Jul 2004 08:26:00 +0200 schrieb Luke Gorrie luke@bluetail.com:
Thomas Schilling tjs_ng@yahoo.de writes:
The exact error message is:
(stream/warning) failure to open network stream to host "localhost" for service "4005"
Took a look at the changelog--Luke, something to confess?
The relevant init.el part:
;; start up slime (global-set-key [(f5)] '(lambda () (interactive) (shell-command "c:/programme/acl62-demo/allegro-ansi.exe +B +cm -L ~/slime.lisp&") (sleep-for 3) (slime-connect "localhost" 4005) (cd "~/Eigene Dateien/")))
Is it just the obvious race condition? That function assumes that Lisp will have the server up and running within 3 seconds. If it's not up in time then you'd get the error you respoted. Note that if the Lisp sources have changed (e.g. after cvs update) then swank-loader will recompile them, and that could easily take more than 3 seconds.
The smallest fix would be to change the connecting part to something like:
(while (not (ignore-errors (slime-connect "localhost" 4005) t)) (sit-for 1))
.. to loop until the connection succeeds.
That would run forever since this is not my problem.
Even a manual M-x slime-connect doesn't work. It says:
connection failed: No such file or directory, 127.0.0.1, SLIME Lisp
And the problem first occured after I checked out yesterday, though I often have long-running sessions where I cannot really say when I started it.
Thomas Schilling tjs_ng@yahoo.de writes:
Even a manual M-x slime-connect doesn't work. It says:
connection failed: No such file or directory, 127.0.0.1, SLIME Lisp
And the problem first occured after I checked out yesterday, though I often have long-running sessions where I cannot really say when I started it.
Does 'telnet localhost 4005' work?
And how old was the version of SLIME you were using previously? If more than a month or two you could try connecting to "localhost" instead of "127.0.0.1".
Am Thu, 22 Jul 2004 16:24:22 +0200 schrieb Luke Gorrie luke@bluetail.com:
Thomas Schilling tjs_ng@yahoo.de writes:
Even a manual M-x slime-connect doesn't work. It says:
connection failed: No such file or directory, 127.0.0.1, SLIME Lisp
And the problem first occured after I checked out yesterday, though I often have long-running sessions where I cannot really say when I started it.
Does 'telnet localhost 4005' work?
No. I already tried to shut off my firewall. But it didn't help.
And how old was the version of SLIME you were using previously? If more than a month or two you could try connecting to "localhost" instead of "127.0.0.1".
Yes, tried that, too. Seems there's no socket open ...
I'm now trying to check out an earlier version. I think my last checkout was 7 July. So I'm trying this now ...
Thomas Schilling tjs_ng@yahoo.de writes:
Yes, tried that, too. Seems there's no socket open ...
Ok, sounds like a server problem. Can you paste the ~/slime.lisp file?
Am Thu, 22 Jul 2004 16:32:19 +0200 schrieb Luke Gorrie luke@bluetail.com:
Thomas Schilling tjs_ng@yahoo.de writes:
Yes, tried that, too. Seems there's no socket open ...
Ok, sounds like a server problem. Can you paste the ~/slime.lisp file?
Already did :) It's just a two-liner. But here it is again:
(load "~/Eigene Dateien/devel/lisp/slime/swank-loader.lisp") (swank::create-swank-server 4005 :spawn #'swank::simple-announce-function t)
It's definitely a slime problem: I checked out from July 7 and it works. So I'm now checking out stepwise - then I can tell you where exactly the problem is.
Thomas Schilling tjs_ng@yahoo.de writes:
Already did :) It's just a two-liner. But here it is again:
(load "~/Eigene Dateien/devel/lisp/slime/swank-loader.lisp") (swank::create-swank-server 4005 :spawn #'swank::simple-announce-function t)
Okay. That works under ACL6.2 on Linux for me.
Can you try debugging on the Lisp side? I don't know much about Windows, but maybe e.g. it would not let you bind port 4005 on the loopback interface if it is already bound to another interface - probably something else though.
BTW, you could abbreviate the second line to just:
(swank:create-server)
.. which was added ages ago but not publicised very well.
Hm, *that* is strange:
I Updated slime to be of version before 2004-7-7. worked. I Updated slime to be of version before 2004-7-8. worked. I Updated slime to be of version before 2004-7-14. worked. I Updated slime to be of version before 2004-7-21. worked. I Updated slime to be of version before 2004-7-22. worked. I Updated slime to be of version HEAD. works ...
Restarted Emacs. F5. works.
What the ... ?
Can you try debugging on the Lisp side? I don't know much about Windows, but maybe e.g. it would not let you bind port 4005 on the loopback interface if it is already bound to another interface - probably something else though.
BTW, you could abbreviate the second line to just:
(swank:create-server)
.. which was added ages ago but not publicised very well.
Okay, doing that and deleting fasl files now.
Very strange things happening here ...
Thomas Schilling tjs_ng@yahoo.de writes:
Hm, *that* is strange:
Probably had some corrupted FASL files. That can happen in some Lisps if you e.g. kill the Lisp process half way through compiling a file.
Am Thu, 22 Jul 2004 16:58:17 +0200 schrieb Thomas Schilling tjs_ng@yahoo.de:
Okay, doing that and deleting fasl files now.
Still works ...
-
Luke Gorrie -
Thomas Schilling