Hello.
On Wed, Apr 11, 2001 at 11:09:35PM +0400, kuznet@ms2.inr.ac.ru wrote:
> BTW if that cursed socket is still alive, try to make the experiment
> with filling window on it. It must stuck, or my theory is completely wrong.
Filling the socket via writing to pty (controlled by sshd), I found
the state, which seems very similar to that I have reported:
# netstat -n -eot | grep 1018
tcp 0 37684 194.190.166.31:22 194.190.161.106:1018
ESTABLISHED 0 11964 off (0.00/0/0)
You see, timers are zero and send-q is not. Zero probes were NOT observed,
exactly as you predict, but keepalive is correct.
However, this is _not_ a staled state. When I resume ssh on 194.190.166.31,
buffer gets empty and connection behaves as normal. I made this experiment
waiting for keepalive packets from both sides, as well as resuming ssh
before keepalives. In both cases connection did not become stale.
So, my conclusion is that your statement about zero probe breakdown due
to keepalives is right, and, I hope, your patch also makes the right thing.
However, this is not an answer for the question how such a stale connection
could arise, and predicted machanism to get it "stuck" does not work.
[I hope we will continue this discussion later.]
-- Eugene Berdnikov - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sun Apr 15 2001 - 21:00:20 EST