Decrease tcp retransmission timeout to ~100 secs
This commit decreases the net.ipv4.tcp_retries2 to 8, which results in a timeout of ~100 seconds down from the default of 15, with results in a timeout of ~15 minutes. This should result in dead connections being detected faster, in ~100 seconds, rather than ~15 minutes, which is important when the floating controller ip moves during a swact. Closes-bug: 1847607 Change-Id: I48c3d396bdcd81ab11d307fd62de36deffc99f12 Signed-off-by: Joseph Richard <joseph.richard@windriver.com>
This commit is contained in:
parent
66d88788a8
commit
50b7827b7d
|
@ -80,3 +80,9 @@ net.ipv4.tcp_keepalive_intvl = 1
|
|||
net.ipv4.tcp_keepalive_probes = 5
|
||||
net.ipv4.tcp_keepalive_time = 5
|
||||
|
||||
# This controls the tcp connection retries.
|
||||
# The default results in a delay of ~15 minutes before dead connections
|
||||
# to the floating ip are detected after a swact.
|
||||
# Reduce this delay to 8 shortens this to ~100 seconds.
|
||||
net.ipv4.tcp_retries2 = 8
|
||||
|
||||
|
|
Loading…
Reference in New Issue