Date: Sat, 25 Feb 2012 08:00:38 +1300 From: Andrew Thompson <thompsa@FreeBSD.org> To: Eugene Grosbein <egrosbein@rdtc.ru> Cc: Alexander Motin <mav@freebsd.org>, "net@freebsd.org" <net@freebsd.org> Subject: Re: netisr+lagg+fragments=80% packet loss Message-ID: <CAFAOGNQvjJcKb_hhzeoN5ptgm882qimsavzN9b=GW_VgPD8mtw@mail.gmail.com> In-Reply-To: <4F47D5EF.7060908@rdtc.ru> References: <4F47C55C.8060006@rdtc.ru> <4F47D5EF.7060908@rdtc.ru>
next in thread | previous in thread | raw e-mail | index | archive | help
2012/2/25 Eugene Grosbein <egrosbein@rdtc.ru>: > 25.02.2012 00:14, Eugene Grosbein =D0=C9=DB=C5=D4: >> This problem occurs only when net.isr.direct=3D0/net.isr.direct_force=3D= 0. >> And only when lagg1 has both ports up and running. And when I use oversi= zed pings. >> At the same time, transit oversized pings go through this BRAS just fine= , >> no packet loss at all. > > Running two copies of tcpdump for igb0 and igb1 simultaneously, > I see that fragments of the same ICMP echo-reply packet encapsulated with= in PPPoE > frame always go out through different ports of lagg1. Even when they arri= ve to client in order, > it seems this depends of switching network in between PPPoE server and cl= ient. > If you are running a recent HEAD then you can try setting net.link.lagg.0.use_flowid to zero. Andrew
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAFAOGNQvjJcKb_hhzeoN5ptgm882qimsavzN9b=GW_VgPD8mtw>