I wish to suggest to developers to add this new knob : tcp_retries2_time - INTEGER This value influences the timeout (seconds) of an alive TCP connection, when RTO retransmissions remain unacknowledged. RFC 1122 recommends at least 100 seconds for the timeout, which corresponds to a value of at least 8.
Rationale : * I see a lot of misuse of tcp_retries2, because even if you check the RTO, it will change with different connections, routing, etc... and so tcp_retries2 timeout will change. * An absolute value will be friendly and better ton read and document. See other parameters with two knobs ( example : dirty_ratio, dirty_ratio_bytes ). Best regards, Sala