Date: Sat, 30 Apr 2011 13:43:39 -0700 From: YongHyeon PYUN <pyunyh@gmail.com> To: Adrian Chadd <adrian@freebsd.org> Cc: freebsd-net@freebsd.org Subject: Re: collisions on tun interfaces ... Message-ID: <20110430204339.GA25769@michelle.cdnetworks.com> In-Reply-To: <BANLkTinDhJOP3Xus1iW-mH8Ru2Q9hkofOw@mail.gmail.com> References: <201104290001.p3T01EvQ082443@artema.vpn.ibs> <20110429095231.GA17786@relay.ibs.dn.ua> <20110429182807.GA21084@michelle.cdnetworks.com> <20110429184634.GA34176@relay.ibs.dn.ua> <20110429185605.GB21084@michelle.cdnetworks.com> <BANLkTinDhJOP3Xus1iW-mH8Ru2Q9hkofOw@mail.gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, Apr 30, 2011 at 11:35:22PM +0800, Adrian Chadd wrote: > On 30 April 2011 02:56, YongHyeon PYUN <pyunyh@gmail.com> wrote: > > > Sending lots of packets within short interval may overflow > > interface's send queue. Try sending UDP packets with netperf and > > see what counters are changed during the test with something like > > "netstat -ndI tun0 -w 1". > > The default sendq is 50. What about considering bumping that? > It's tunable. Set net.link.ifqmaxlen. > > Adrian
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20110430204339.GA25769>