[plt-scheme] Surprising behavior from "run-server"

I think I don't understand the test.

It looks like you have set up a server using `run-server' with a
timeout of 0.1 seconds on each connection. The you create two
connections. You ask the second one to get stuck, so the associated
handler thread on the server times out, and the client gets back an
EOF.

Since the timeout is the same for all connections, and since the client
waits for the one that gets stuck to time out, then the other
connection (the one created first) also times out by then --- even
though it's not doing anything.

So, maybe the issue is that the timeout is wall-clock time, not
processor time?

Matthew

At Sun, 04 Nov 2007 05:41:35 -0800, Eric Hanchrow wrote:
> This is r7563 of the PLT trunk.
> 
> Run the file "test.ss"; you'll see
> 
>     Sent hang; got #<eof>
>     Sent list-tables; got #<eof>
> 
> That first line makes sense: the server to which we sent the 'hang
> command has been killed because it ran too long.  But the other server
> should have kept running, as far as I can tell.
> 
> Is this a bug, or am I failing to understand something about how
> run-server works?
> 
> 
> ------------------------------------------------------------------------------
> [application/octet-stream "server.ss"] [save to ~/Desktop/ & open]
> 
> ------------------------------------------------------------------------------
> [application/octet-stream "tables.ss"] [save to ~/Desktop/ & open]
> 
> ------------------------------------------------------------------------------
> [application/octet-stream "test.ss"] [save to ~/Desktop/ & open]
> 
> -- 
> I read the opening sentence [of The Lord Of The Rings] and
> phoned around several friends to say goodbye, because suicide
> seemed so obviously preferable to 500 more pages of the same.
>         -- Helene Hanff
> _________________________________________________
>   For list-related administrative tasks:
>   http://list.cs.brown.edu/mailman/listinfo/plt-scheme