From: jason andrade Subject: linux nfs still limited to rsize=1024 ? Date: Fri, 2 Aug 2002 01:01:53 +1000 (EST) Sender: nfs-admin@lists.sourceforge.net Message-ID: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Return-path: Received: from piglet.dstc.edu.au ([130.102.176.1]) by usw-sf-list1.sourceforge.net with esmtp (Exim 3.31-VA-mm2 #1 (Debian)) id 17aHSm-0002X3-00 for ; Thu, 01 Aug 2002 08:02:04 -0700 Received: from sunburn.dstc.edu.au (sunburn.dstc.edu.au [130.102.176.16]) by piglet.dstc.edu.au (8.12.3/8.12.3) with ESMTP id g71F1rSQ011298 for ; Fri, 2 Aug 2002 01:01:53 +1000 (EST) To: mailto: ; Errors-To: nfs-admin@lists.sourceforge.net List-Help: List-Post: List-Subscribe: , List-Id: Discussion of NFS under Linux development, interoperability, and testing. List-Unsubscribe: , List-Archive: Hi, I've been trying to use Linux NFS in read only mode with a block size of > 1024, for around 2 years now. I've just rebooted into a redhat linux 2.4.18-5 kernel and on a whim decided to try rsize=8192 to see if this has been "fixed". nope, it hasn't. df hangs for me with any nfs mount using rsize > 1024. if i drop it back to that, it works fine. I'm kind of stunned that even with gigabit interfaces and MTU's of 9000, i've still got to stay with 1K packets under nfs. (both client and server have gigE and 9000 mtus) The server is a redhat linux box running 2.4.18-5 also, exporting about 1.2Tb read only. The clients mount using nfs over udp, since nfs/tcp doesn't seem to be (well?) supported or recommended in the redhat kernels. regards, -jason ------------------------------------------------------- This sf.net email is sponsored by:ThinkGeek Welcome to geek heaven. http://thinkgeek.com/sf _______________________________________________ NFS maillist - NFS@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nfs