path: root/net/ipv4/tcp_timer.c
diff options
authorYuchung Cheng <ycheng@google.com>2019-01-16 15:05:34 -0800
committerDavid S. Miller <davem@davemloft.net>2019-01-17 15:12:26 -0800
commit590d2026d62418bb27de9ca87526e9131c1f48af (patch)
tree6d2de504c98a46b7dcefed405151b02957e896a0 /net/ipv4/tcp_timer.c
parenttcp: simplify window probe aborting on USER_TIMEOUT (diff)
tcp: retry more conservatively on local congestion
Previously when the sender fails to retransmit a data packet on timeout due to congestion in the local host (e.g. throttling in qdisc), it'll retry within an RTO up to 500ms. In low-RTT networks such as data-centers, RTO is often far below the default minimum 200ms (and the cap 500ms). Then local host congestion could trigger a retry storm pouring gas to the fire. Worse yet, the retry counter (icsk_retransmits) is not properly updated so the aggressive retry may exceed the system limit (15 rounds) until the packet finally slips through. On such rare events, it's wise to retry more conservatively (500ms) and update the stats properly to reflect these incidents and follow the system limit. Note that this is consistent with the behavior when a keep-alive probe is dropped due to local congestion. Signed-off-by: Yuchung Cheng <ycheng@google.com> Signed-off-by: Eric Dumazet <edumazet@google.com> Reviewed-by: Neal Cardwell <ncardwell@google.com> Reviewed-by: Soheil Hassas Yeganeh <soheil@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'net/ipv4/tcp_timer.c')
1 files changed, 3 insertions, 5 deletions
diff --git a/net/ipv4/tcp_timer.c b/net/ipv4/tcp_timer.c
index c36089aa3515..d7399a89469d 100644
--- a/net/ipv4/tcp_timer.c
+++ b/net/ipv4/tcp_timer.c
@@ -500,14 +500,13 @@ void tcp_retransmit_timer(struct sock *sk)
+ icsk->icsk_retransmits++;
if (tcp_retransmit_skb(sk, tcp_rtx_queue_head(sk), 1) > 0) {
/* Retransmission failed because of local congestion,
- * do not backoff.
+ * Let senders fight for local resources conservatively.
- if (!icsk->icsk_retransmits)
- icsk->icsk_retransmits = 1;
inet_csk_reset_xmit_timer(sk, ICSK_TIME_RETRANS,
- min(icsk->icsk_rto, TCP_RESOURCE_PROBE_INTERVAL),
goto out;
@@ -528,7 +527,6 @@ void tcp_retransmit_timer(struct sock *sk)
* the 120 second clamps though!
- icsk->icsk_retransmits++;
/* If stream is thin, use linear timeouts. Since 'icsk_backoff' is