> i'm genuinely curious to know, btw, what pathological conditions
> do you think might cause a catastrophic memory shortage?
The last time I tried a 2.2.x kernel connect() kept returning ENOBUFS if
called many times quickly in non-blocking mode to a large number of different
IP addresses; I quickly withdrew to 2.0.x where things are nice and stable. I
don't know if this was a catastrophic memory shortage, or simply a lack of one
type of buffer able to be allocated in a non-blocking fashion to kernel use,
but it was definitely horrible behaviour caused by a user-level network
scanner, which was then rendered useless in itself and caused problems to
other applications on the system.
(this was 2.2.0-pre4 I think - I decided to wait till the traffic about general
problems died down before testing the 2.2.x waters again; the list still seems
to be filled with enough problems to not try any real testing yet)
David.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/