[Bese-devel] How to play safe?

Marco Baringer mb at bese.it
Thu Feb 3 00:54:34 UTC 2005


etyurin at comcast.net writes:

> When I run Slime/Swank combo and make a mistake in my lisp code, I
> don't get to debugger even if I re-attach to the lisp socket (using
> standard attachtty/detachtty setup).

using slime's slime-list-threads (i think that's what its called) you
can attach to a running lisp and get a debugger in any thread. once
you have the debugger you should have the usual server-error and
retry-action restarts available.

> All I can do at this point is (ucw:restart-server *default-server*)
>
> What am I missing in my setup?

i generally develop through emacs, so errors always immediatly pop up,
are you running ucw standalone and only attaching slime when an error
occurs?

> Also, if I am running several applications at once, mistake in one
> takes them all out.  How can I make UCW continue serving other
> application while waiting for my response on the broken
> application's debugger?

the mod_lisp backend is, atm, single threaded. this means that if you
can only take one request/response at a time, no matter how many (or
how few) applictions you have running. on the aserve and araneida
backends (if the lisp supports threads) you shouldn't have this
problem.

-- 
-Marco
Ring the bells that still can ring.
Forget the perfect offering.
There is a crack in everything.
That's how the light gets in.
	-Leonard Cohen



More information about the bese-devel mailing list