> My apps are all working fine with this new behavior.
Probably, our apps set large backlog. Normal apps used to set it to 5.
F.e. even emacs client starts to died after this.
Probably, minimum at 128 will be enough to move the problem
to area of "never occurs in real life". Only ECONNREFUSED
should be replaced with something more appropriate.
F.e. we could sleep for blocking sockets and EAGAIN for non-blocking.
Your patch is bad because it prefers to break innocent programs
to fix hole, which can be used by maliciuous program.
Honest citizen should not suffer 8)
In any case, the same problem still exists with SOCK_DGRAM.
Alexey
-
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/