From: "Andrew Theurer" Subject: Re: RPC error on NFS over TCP Date: Tue, 26 Mar 2002 09:27:31 -0600 Sender: nfs-admin@lists.sourceforge.net Message-ID: <002d01c1d4da$c00d2310$2d060e09@beavis> References: <200203260016.SAA38018@popmail.austin.ibm.com> <20020326071622.A28115@ead45> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Cc: Received: from mg01.austin.ibm.com ([192.35.232.18]) by usw-sf-list1.sourceforge.net with esmtp (Exim 3.31-VA-mm2 #1 (Debian)) id 16psnT-0007iO-00 for ; Tue, 26 Mar 2002 07:23:39 -0800 To: "Bill Rugolsky Jr." 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: > On Mon, Mar 25, 2002 at 06:05:03PM -0600, Andrew Theurer wrote: > > The linux server is kernel 2.4.18 (with Neil's patches) and sometimes 2.5.7. > > The clients (48 of them) are mostly just RH-2.4.2-2 and one is 2.5.7 (happens > > on both kernels). There are no dropped packets reported on ifconfig. There > > are RPC retransmits on the clients. Any ideas? > > You may need more threads to service the clients? See Neil's notes > that accompanied the patches (in the archives) w.r.t the number of threads > and the number of RPC receive buffers. Thanks Bill! I tried 128 threads and it seems to be OK so far. -Andrew _______________________________________________ NFS maillist - NFS@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nfs