mirror of
https://github.com/torvalds/linux.git
synced 2024-11-11 06:31:49 +00:00
1f37bf87aa
Ueki Kohei reported that when we are using NewReno with connections that
have a very low traffic, we may timeout the connection too early if a
second loss occurs after the first one was successfully acked but no
data was transfered later. Below is his description of it:
When SACK is disabled, and a socket suffers multiple separate TCP
retransmissions, that socket's ETIMEDOUT value is calculated from the
time of the *first* retransmission instead of the *latest*
retransmission.
This happens because the tcp_sock's retrans_stamp is set once then never
cleared.
Take the following connection:
Linux remote-machine
| |
send#1---->(*1)|--------> data#1 --------->|
| | |
RTO : :
| | |
---(*2)|----> data#1(retrans) ---->|
| (*3)|<---------- ACK <----------|
| | |
| : :
| : :
| : :
16 minutes (or more) :
| : :
| : :
| : :
| | |
send#2---->(*4)|--------> data#2 --------->|
| | |
RTO : :
| | |
---(*5)|----> data#2(retrans) ---->|
| | |
| | |
RTO*2 : :
| | |
| | |
ETIMEDOUT<----(*6)| |
(*1) One data packet sent.
(*2) Because no ACK packet is received, the packet is retransmitted.
(*3) The ACK packet is received. The transmitted packet is acknowledged.
At this point the first "retransmission event" has passed and been
recovered from. Any future retransmission is a completely new "event".
(*4) After 16 minutes (to correspond with retries2=15), a new data
packet is sent. Note: No data is transmitted between (*3) and (*4).
The socket's timeout SHOULD be calculated from this point in time, but
instead it's calculated from the prior "event" 16 minutes ago.
(*5) Because no ACK packet is received, the packet is retransmitted.
(*6) At the time of the 2nd retransmission, the socket returns
ETIMEDOUT.
Therefore, now we clear retrans_stamp as soon as all data during the
loss window is fully acked.
Reported-by: Ueki Kohei
Cc: Neal Cardwell <ncardwell@google.com>
Cc: Yuchung Cheng <ycheng@google.com>
Signed-off-by: Marcelo Ricardo Leitner <mleitner@redhat.com>
Acked-by: Neal Cardwell <ncardwell@google.com>
Tested-by: Neal Cardwell <ncardwell@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
|
||
|---|---|---|
| .. | ||
| netfilter | ||
| af_inet.c | ||
| ah4.c | ||
| arp.c | ||
| cipso_ipv4.c | ||
| datagram.c | ||
| devinet.c | ||
| esp4.c | ||
| fib_frontend.c | ||
| fib_lookup.h | ||
| fib_rules.c | ||
| fib_semantics.c | ||
| fib_trie.c | ||
| fou.c | ||
| geneve.c | ||
| gre_demux.c | ||
| gre_offload.c | ||
| icmp.c | ||
| igmp.c | ||
| inet_connection_sock.c | ||
| inet_diag.c | ||
| inet_fragment.c | ||
| inet_hashtables.c | ||
| inet_lro.c | ||
| inet_timewait_sock.c | ||
| inetpeer.c | ||
| ip_forward.c | ||
| ip_fragment.c | ||
| ip_gre.c | ||
| ip_input.c | ||
| ip_options.c | ||
| ip_output.c | ||
| ip_sockglue.c | ||
| ip_tunnel_core.c | ||
| ip_tunnel.c | ||
| ip_vti.c | ||
| ipcomp.c | ||
| ipconfig.c | ||
| ipip.c | ||
| ipmr.c | ||
| Kconfig | ||
| Makefile | ||
| netfilter.c | ||
| ping.c | ||
| proc.c | ||
| protocol.c | ||
| raw.c | ||
| route.c | ||
| syncookies.c | ||
| sysctl_net_ipv4.c | ||
| tcp_bic.c | ||
| tcp_cong.c | ||
| tcp_cubic.c | ||
| tcp_dctcp.c | ||
| tcp_diag.c | ||
| tcp_fastopen.c | ||
| tcp_highspeed.c | ||
| tcp_htcp.c | ||
| tcp_hybla.c | ||
| tcp_illinois.c | ||
| tcp_input.c | ||
| tcp_ipv4.c | ||
| tcp_lp.c | ||
| tcp_memcontrol.c | ||
| tcp_metrics.c | ||
| tcp_minisocks.c | ||
| tcp_offload.c | ||
| tcp_output.c | ||
| tcp_probe.c | ||
| tcp_scalable.c | ||
| tcp_timer.c | ||
| tcp_vegas.c | ||
| tcp_vegas.h | ||
| tcp_veno.c | ||
| tcp_westwood.c | ||
| tcp_yeah.c | ||
| tcp.c | ||
| tunnel4.c | ||
| udp_diag.c | ||
| udp_impl.h | ||
| udp_offload.c | ||
| udp_tunnel.c | ||
| udp.c | ||
| udplite.c | ||
| xfrm4_input.c | ||
| xfrm4_mode_beet.c | ||
| xfrm4_mode_transport.c | ||
| xfrm4_mode_tunnel.c | ||
| xfrm4_output.c | ||
| xfrm4_policy.c | ||
| xfrm4_protocol.c | ||
| xfrm4_state.c | ||
| xfrm4_tunnel.c | ||