Date: Wed, 19 Jul 1995 15:23:11 +0100 (BST) From: Doug Rabson <dfr@render.com> To: Bruce Evans <bde@zeta.org.au> Cc: bde@zeta.org.au, phk@freefall.cdrom.com, current@freebsd.org, esser@zpr.uni-koeln.de Subject: Re: slow nfsv3 writes Message-ID: <Pine.BSF.3.91.950719152148.22916C-100000@minnow.render.com> In-Reply-To: <199507191408.AAA23267@godzilla.zeta.org.au>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, 20 Jul 1995, Bruce Evans wrote: > >> I was getting 9K/sec for writes before I complained and this change > >> was made :-). > > >How about running a snoop and get some per-packet times ? > > `iozone 1 8192' on a 486DX2/66 over lo0 reports reading at > 1775129 bytes/sec and writing at 111227 bytes/sec. tcpdump reports > about 4ms for reading 8292 bytes and about 70ms for writing 8320 > bytes. > > The problem was that this is actually for nfsv2. For some reason I > thought that nfsv3 would be the default. I think the default should be for mount_nfs to attempt a v3 mount rpc and if that fails, then fall back to v2. > > Under nfsv3, `iozone 4 8192' reports ... oops it hangs on netio and > other processes hang on nfsrcvlk and after a little while other > processes hang on ufslk2 ... This is probably the same problem > that Doug saw. I haven't had a chance to look at this yet. I am in Windows95 mode at the moment :-( -- Doug Rabson, Microsoft RenderMorphics Ltd. Mail: dfr@render.com Phone: +44 171 251 4411 FAX: +44 171 251 0939
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.91.950719152148.22916C-100000>