Hello everybody,

maybe somebody can help me with this one: I wrote a prog which fetches
pages from the web with multiple threads. However, the URLs given
sometimes cause a network timeout:

#<Errno::ETIMEDOUT: Die Wartezeit f die Verbindung ist abgelaufen - "connect(2)">
(German for: connection timeout)

But this error only appears after a long pause, where all other threads
are blocked as well. So I thought I use the timeout library and only wait
for some shorter time and continue with the next URL if I don't get
response fast enough. This may work in a single threaded app, but not
really with multiple threads: The threads don't really get the time
they're supposed to have.

Anyway, how would I avoid blocking threads because of network timeouts
while guaranteeing each thread enough time to fulfill its task?

So long
Arno Erpenbeck

BTW:
Praise for Ruby: I've never met a language where you get productive so fast!