our problem is a network printer the has paper empty.
The socket connection to the printer broke down after ~45 minutes with
errno 110 (Connection timed out).
Linux base version is 2.4.18 on a strongarm machine.
Tracking this down brought us to the tcp_send_probe0 function in
The tp->backoff value becomes allways increased.
on this machine from 31 on (tp->rto << tp->backoff) is 0.
The xmit timer is set to this timeout value - resulting in an ACK burst.
If the TCP sender gets the (default) 16 ACKS out, before the receiver can
answer them, the connection dies.
This happened every night, when the printer received a huge job from a foreign
If this isn't a bug, it should be made configurable. Or do we miss something?
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to email@example.com
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/