From: Dave Taht <dave.t...@bufferbloat.net> A year of testing in the cerowrt project shows not using timestamps to be a very bad idea in nearly any TCP at speeds above a few Mbit.
As FQ_CODEL will actually assert ECT(3) on ECN streams, it is useful to have ECN be negotiated when offered. In this day and age it could also be hoped that ECN could merely be one by default, but... Lastly sack/dsack help on recovery from larger amounts of packet loss. --- package/base-files/files/etc/sysctl.conf | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/package/base-files/files/etc/sysctl.conf b/package/base-files/files/etc/sysctl.conf index 2d97000..0aa0588 100644 --- a/package/base-files/files/etc/sysctl.conf +++ b/package/base-files/files/etc/sysctl.conf @@ -4,11 +4,13 @@ net.ipv4.conf.all.arp_ignore=1 net.ipv4.ip_forward=1 net.ipv4.icmp_echo_ignore_broadcasts=1 net.ipv4.icmp_ignore_bogus_error_responses=1 -net.ipv4.tcp_ecn=0 +net.ipv4.tcp_ecn=2 net.ipv4.tcp_fin_timeout=30 net.ipv4.tcp_keepalive_time=120 net.ipv4.tcp_syncookies=1 -net.ipv4.tcp_timestamps=0 +net.ipv4.tcp_timestamps=1 +net.ipv4.tcp_sack=1 +net.ipv4.tcp_dsack=1 net.ipv4.netfilter.ip_conntrack_checksum=0 net.ipv4.netfilter.ip_conntrack_max=16384 -- 1.7.9.5 _______________________________________________ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/mailman/listinfo/openwrt-devel