Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 27 Feb 2010 22:43:33 +0100
From:      Gerrit =?ISO-8859-1?Q?K=FChn?= <gerrit@pmp.uni-hannover.de>
To:        Jeremy Chadwick <freebsd@jdc.parodius.com>
Cc:        freebsd-fs@freebsd.org, stable@freebsd.org, Willem Jan Withagen <wjw@digiware.nl>
Subject:   Re: mbuf leakage with nfs/zfs?
Message-ID:  <20100227224333.5bee38ac.gerrit@pmp.uni-hannover.de>
In-Reply-To: <20100227193819.GA60576@icarus.home.lan>
References:  <20100226141754.86ae5a3f.gerrit@pmp.uni-hannover.de> <E1Nl1mb-0002Mx-M9@kabab.cs.huji.ac.il> <E1Nl2JK-00033U-Fw@kabab.cs.huji.ac.il> <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>

next in thread | previous in thread | raw e-mail | index | archive | help
On Sat, 27 Feb 2010 11:38:19 -0800 Jeremy Chadwick
<freebsd@jdc.parodius.com> wrote about Re: mbuf leakage with nfs/zfs?:

JC> I should point out that the NFS+ZFS-based filer doesn't actually do its
JC> backups using NFS; it uses rsnapshot (rsync) over SSH.  There is
JC> intense network I/O during backup time though, depending on how much
JC> data there is to back up.  The NFS mounts (on the clients) are only
JC> used to provide a way for people to get access to their nightly
JC> backups in a convenient way; it isn't used very heavily.

That's rather similar to my situation, I would say. Most traffic goes via
rsync, nfs only gives access to home dirs, which are not intensively used.

JC> I can do something NFS-intensive on any of the above clients if people
JC> want me to kind of testing.  Possibly an rsync with a source of the NFS
JC> mount and a destination of the local disk would be a good test?  Let me
JC> know if anyone's interested in me testing that.

>From the last emails I would say we get most out of it by comparing tcp
and udp clients to make sure this happens only with udp (and it is still
not quite clear to me if it also happens with a FBSD client using udp).

OTOH it would be great if someone with the ability to actually fix
something in the nfs code could get in this discussion to guide us to do
the debugging needed to do so.


cu
  Gerrit



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20100227224333.5bee38ac.gerrit>