Date: Tue, 1 Sep 2009 23:50:19 +0200 From: Ivan Voras <ivoras@freebsd.org> To: "Simon L. Nielsen" <simon@freebsd.org> Cc: svn-src-head@freebsd.org, svn-src-all@freebsd.org, src-committers@freebsd.org Subject: Re: svn commit: r196738 - head/sys/netinet Message-ID: <9bbcef730909011450t37582b50pecf87bb9927faf18@mail.gmail.com> In-Reply-To: <20090901202659.GA1495@arthur.nitro.dk> References: <200909011753.n81Hr1AV068295@svn.freebsd.org> <20090901202659.GA1495@arthur.nitro.dk>
next in thread | previous in thread | raw e-mail | index | archive | help
2009/9/1 Simon L. Nielsen <simon@freebsd.org>: > On 2009.09.01 17:53:01 +0000, Bjoern A. Zeeb wrote: >> Author: bz >> Date: Tue Sep =C2=A01 17:53:01 2009 >> New Revision: 196738 >> URL: http://svn.freebsd.org/changeset/base/196738 >> >> Log: >> =C2=A0 In case an upper layer protocol tries to send a packet but the >> =C2=A0 L2 code does not have the ethernet address for the destination >> =C2=A0 within the broadcast domain in the table, we remember the >> =C2=A0 original mbuf in `la_hold' in arpresolve() and send out a >> =C2=A0 different packet with an arp request. > > [...] > > Thanks! > > In case anybody is interested, this was the issue which indirectly > killed the FreeBSD mail system for some hours recently. =C2=A0It was caus= ed > by ns1.FreeBSD.org stopped responding to network traffic due to it > running out of mbuf's. > > Feel free to guess when bz's patch was applied ;-) : > http://people.freebsd.org/~simon/misc/mbuf_leak_ns1_200908.png . =C2=A0Th= e > final spike on the graph is me running 'arp -da' a bunch of times. Interesting :) I assume this is 8 / HEAD -specific due to work done in ARP code?
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?9bbcef730909011450t37582b50pecf87bb9927faf18>