Date: Mon, 16 Jul 2001 09:27:05 -0500 From: Dan Nelson <dnelson@emsphone.com> To: Arun Sharma <arun@sharmas.dhs.org> Cc: questions@FreeBSD.ORG Subject: Re: NFS mount Linux Mandrake 8.0 -> FreeBSD 4.3-STABLE hang Message-ID: <20010716092705.B5696@dan.emsphone.com> In-Reply-To: <20010715215513.A20380@sharmas.dhs.org> References: <20010715200102.A19948@sharmas.dhs.org> <20010715220415.A3289@dan.emsphone.com> <20010715214538.A20260@sharmas.dhs.org> <20010715215513.A20380@sharmas.dhs.org>
next in thread | previous in thread | raw e-mail | index | archive | help
In the last episode (Jul 15), Arun Sharma said: > On Sun, Jul 15, 2001 at 09:45:38PM -0700, Arun Sharma wrote: > > > > Actually, I figured out that the cp doesn't hang, but takes a *long* > > time - longer than it takes to transfer the file via scp over the > > same interface. > > > > tcpdump shows that NFS is using 8kb block size, but packets get > > fragmented because I'm on a wireless network with a MTU < 2k. > > > > -Arun > > Another clue: the writes are 100x faster than reads, when tested > using dd with the same block size. Just as a test, try dropping your NFS blocksize down to 1k (mount -o -r1024,-w1024) or using a TCP mount (mount -o -T) and see waht happens. -- Dan Nelson dnelson@emsphone.com To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20010716092705.B5696>