Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 28 Feb 2010 14:11:42 +0200
From:      Daniel Braniss <danny@cs.huji.ac.il>
To:        Jeremy Chadwick <freebsd@jdc.parodius.com>
Cc:        stable@freebsd.org, freebsd-fs@freebsd.org, Willem Jan Withagen <wjw@digiware.nl>, Gerrit =?iso-8859-1?Q?K=FChn?= <gerrit@pmp.uni-hannover.de>, Eirik =?iso-8859-1?Q?=D8verby?= <ltning@anduin.net>, rwatson@freebsd.org
Subject:   Re: mbuf leakage with nfs/zfs? 
Message-ID:  <E1Nlhzr-0004ci-3W@kabab.cs.huji.ac.il>
In-Reply-To: <20100227220310.GA65110@icarus.home.lan> 
References:  <20100226174021.8feadad9.gerrit@pmp.uni-hannover.de>  <E1Nl6VA-000557-D9@kabab.cs.huji.ac.il> <20100226224320.8c4259bf.gerrit@pmp.uni-hannover.de> <4B884757.9040001@digiware.nl> <20100227080220.ac6a2e4d.gerrit@pmp.uni-hannover.de> <4B892918.4080701@digiware.nl> <20100227202105.f31cbef7.gerrit@pmp.uni-hannover.de> <20100227193819.GA60576@icarus.home.lan> <BD8AC9F6-DF96-41F9-8E92-48A4E5606DC7@anduin.net> <4B89943C.70704@digiware.nl> <20100227220310.GA65110@icarus.home.lan>

next in thread | previous in thread | raw e-mail | index | archive | help
> On Sat, Feb 27, 2010 at 10:53:00PM +0100, Willem Jan Withagen wrote:
> > On 27-2-2010 21:32, Eirik =D8verby wrote:
> > >I've had a discussion with some folks on this for a while. I can eas=
il=3Dy
> > >reproduce this situation by mounting a FreeBSD ZFS filesystem via
> > >NFS-UDP from an OpenBSD machine. Telling the OpenBSD machine to use =
TC=3DP
> > >instead of UDP makes the problem go away.
> > >
> > >Other FreeBSD systems mounting the same share, either using UDP or T=
CP=3D,
> > >does not cause the problem to show up.
> > >
> > >A patch was suggested by Rick Macklem, but that did not solve the is=
su=3De:
> > >http://lists.freebsd.org/pipermail/freebsd-current/2009-December/014=
18=3D1.html> >=20
> > I concur.
> > Everything in my network is now on TCP, and there is no mbuf leakage.=

> > I just don't get over the 5500 mark, no matter what I throw at it.
> >=20
> > I do feel that TCP is not as well performing on a local net with Linu=
x,
> > hence the choice for UDP. But TCP is workable as next best.
>=20
> I'm pulling in Robert Watson, who has some familiarity with the UDP
> stack/code in FreeBSD.  I'm not sure he'll be a sufficient source of
> knowledge for this specific issue since it appears (?) to be specific t=
o
> NFS; Rick Macklem would be a better choice, but as reported, he's MIA.
>=20
> Robert, are you aware of any changes or implementation issues which
> might cause excessive (read: leaking) mbuf use under UDP-based NFS?  Do=

> you know of a way folks could determine the source of the leak, either
> via DDB or while the system is live?

I have been runing some tests in a controlled environment.

server and client are both 64bit Xeon/X5550 =40  2.67GHz with 16Gb of mem=
ory
FreeBSD/SMP: 2 package(s) x 4 core(s) x 2 SMT threads

the client is runing latest 8.0 stable
the load is created by runing 'make -j32 buildworld' and sleeping 150 sec=
.
in between runs, this is the straight line you will see in the graphs.
Both the src and obj directories are NFS mounted from the server, regular=
 UFS.

when server is running 7.2-stable no leakage is seen.
 see ftp://ftp.cs.huji.ac.il/users/danny/freebsd/mbufs/=7Btcp,udp=7D-7.2.=
ps
when server is runing 8.0-stable
 see ftp://ftp.cs.huji.ac.il/users/danny/freebsd/mbufs/=7Btcp,udp=7D-8.0.=
ps
you can see that udp is leaking=21

cheers,
	danny
ps: I think the subject should be changed again, removing zfs ...





Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?E1Nlhzr-0004ci-3W>