Date: Tue, 8 Oct 2019 08:07:13 -0500 From: Jason Bacon <bacon4000@gmail.com> To: "Mikhail T." <mi+t@aldan.algebra.com>, Hans Petter Selasky <hps@selasky.org>, "Mikhail T." <mi+thun@aldan.algebra.com>, freebsd-infiniband@freebsd.org Subject: Re: Questions about Infiniband on FreeBSD Message-ID: <65bfad8d-2f31-415e-69ec-b74dd5806f25@gmail.com> In-Reply-To: <a12b13b6-b275-686f-5d12-cfc6d8278915@aldan.algebra.com> References: <5570c97f-3903-d499-2420-8351f7beed37@aldan.algebra.com> <062c0a38-a51d-58b0-2a6e-594102debad2@gmail.com> <112453f7-350e-a1e9-d372-923dfbdb727c@selasky.org> <a12b13b6-b275-686f-5d12-cfc6d8278915@aldan.algebra.com>
index | next in thread | previous in thread | raw e-mail
On 2019-10-08 00:12, Mikhail T. wrote: > On 03.10.19 03:18, Hans Petter Selasky wrote: >> On 2019-10-03 02:49, Jason Bacon wrote: >>> I saw frequent hangs (self-resolving) with an MTU of 65520. Cutting >>> it in half improved reliability by orders of magnitude, but still >>> occasional issues. Halving it again to 16380 seemed to be the sweet >>> spot. >> >> Try this fix: >> >> https://svnweb.freebsd.org/changeset/base/352955 > > Well, I waited for these to hit the 11-stable and rebuilt both > kernels. Unfortunately, I cannot tell they help very much. With MTU of > 65K I still see hangups in NFS and other data-transfers. > > Setting MTU to 16K seems Ok. Yours, > > -mi > Bummer. Thanks for following up, JB -- Earth is a beta site.home | help
Want to link to this message? Use this
URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?65bfad8d-2f31-415e-69ec-b74dd5806f25>