Date: Sat, 24 Jan 2015 09:29:44 -0800 From: Adrian Chadd <adrian@freebsd.org> To: Dmitry Sivachenko <trtrmitya@gmail.com> Cc: FreeBSD Net <freebsd-net@freebsd.org> Subject: Re: network locks up with udp traffic Message-ID: <CAJ-Vmokf9cC_26jF5B99_CRpjOaLOQUpBqMCcyqHevfi3JfFdQ@mail.gmail.com> In-Reply-To: <FCEBED19-25A4-4D8E-8EE5-8B2B397702C5@gmail.com> References: <5B08402C-67A7-49E7-ADF8-390C94DCF1D7@gmail.com> <CAJ-Vmo=v5S=cRM3QPo0mGiNS8OS8so=915vx6R1fhDaKWS6BhQ@mail.gmail.com> <FCEBED19-25A4-4D8E-8EE5-8B2B397702C5@gmail.com>
next in thread | previous in thread | raw e-mail | index | archive | help
Hi, Can you compile your kernel with WITNESS and repeat this test? It looks like it's stuck in some lock ordering thing between some UDP paths and an inpcb lookup. It should log in dmesg the first time a LOR occurs, regardless of whether it actively causes a hang. (And yes, this is definitely critical enough to be a FreeBSD PR... :) Thanks! -adrian On 24 January 2015 at 02:14, Dmitry Sivachenko <trtrmitya@gmail.com> wrote: > >> On 23 =D1=8F=D0=BD=D0=B2. 2015 =D0=B3., at 21:58, Adrian Chadd <adrian@f= reebsd.org> wrote: >> >> Hi, >> >> When this next happens, please do this as root: >> >> procstat -ka >> >> That'll hopefully provide enough information to figure out which >> processes are blocking where and how they got there. >> >> > > > Hello, > > I am attaching procstat -ka output (and top output to see snmpd process i= n "*udp" state). > ifconfig process was started with "net0" argument (net0 is a renamed igb0= in my case). >
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?CAJ-Vmokf9cC_26jF5B99_CRpjOaLOQUpBqMCcyqHevfi3JfFdQ>