Date: Sat, 10 Jul 2010 19:08:45 +0200 From: Vlad Galu <dudu@dudu.ro> To: freebsd-net@freebsd.org Cc: freebsd-stable@freebsd.org Subject: Re: Sudden mbuf demand increase and shortage under the load Message-ID: <AANLkTimQIEbqoSvZl1i_1jmcYvQt7dLpJnBawrZ_MPe4@mail.gmail.com> In-Reply-To: <4B7AD495.30900@FreeBSD.org> References: <4B79297D.9080403@FreeBSD.org> <4B79205B.619A0A1A@verizon.net> <4B7A38F5.3090404@FreeBSD.org> <4B7A7D2C.9040200@quip.cz> <4B7AD495.30900@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Tue, Feb 16, 2010 at 7:23 PM, Maxim Sobolev <sobomax@freebsd.org> wrote: > Miroslav Lachman wrote: >> >> Can it be related to this issue somehow? >> >> http://lists.freebsd.org/pipermail/freebsd-current/2009-August/011013.html >> http://lists.freebsd.org/pipermail/freebsd-current/2009-August/010740.html >> >> It was tested on FreeBSD 8 and high UDP traffic on igb interfaces emits >> messages "GET BUF: dmamap load failure - 12" and later results in kernel >> panic. >> We have not received any response to this report. > > Could be the issue, however in our case there is no panic, just that all > userland activity in the system ceases for 2 minutes after it reaches > certain network load level. Sorry for digging into this old thread, but I'm seeing similar symptoms with today's RELENG_8 and bge(4) attached to BCM5721, on an UP system. kern.ipc.nmbclusters is 65536, but what strikes me odd is this: 0/115446003/57722999 requests for mbufs denied (mbufs/clusters/mbuf+clusters) vmstat -i shows a rate of 1100 for the adapter. The machine runs a fairly small PF configuration, but I've already ruled it out, the symptoms appear when PF is disabled as well. I'll happily provide more info. Regards, Vlad -- Good, fast & cheap. Pick any two.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?AANLkTimQIEbqoSvZl1i_1jmcYvQt7dLpJnBawrZ_MPe4>