Hello, Question: What is the maximum amount of time a packet can be buffered before calling nfq_set_verdict? Question: Is there a configurable timeout that controls whether or not a queued packet can be transmitted? Question: If there is a configurable timeout, how can the timeout be changed? I have a bridge with two interfaces, eth1 and usb0. I use iptables to queue all packets from the FORWARD chain: # iptables -A FORWARD -j QUEUE I use libnetfilter_queue to get copies of the packets and buffer them. If my application is in the forwarding state, it immediately copies the packet from the receive queue to the transmit queue. When the transmit thread sees that the transmit queue is not empty, it gets the next packet and uses nfq_set_verdict with NF_ACCEPT to transmit the packet. This works. If my application is in the buffering state, it buffers packets until it receives the command to flush all buffered packets. It copies packets from the receive queue to the transmit queue. When the transmit thread sees that the transmit queue is not empty, it gets the next packet and uses nfq_set_verdict with NF_ACCEPT to transmit the packet. This does not work Apparently, if there is a delay of more than a few seconds between when the application buffers a packet and when it calls nfq_set_verdict, not all packets get transmitted. In fact, only the first buffered packet gets transmitted. I suspect there is some timeout mechanism in the netfilter code. However, I was unable to find the exact location of the timeout code. Thank you, David -- To unsubscribe from this list: send the line "unsubscribe netfilter" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html