From: Shehjar Tikoo Subject: Returning error to app on TCP disconnect Date: Tue, 08 Jun 2010 17:06:46 +0530 Message-ID: <4C0E2B4E.30205@gluster.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed To: linux-nfs@vger.kernel.org Return-path: Received: from zp3.zcs.datasyncintra.net ([208.88.241.29]:5021 "EHLO zp3.zcs.datasyncintra.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751993Ab0FHLqy (ORCPT ); Tue, 8 Jun 2010 07:46:54 -0400 Received: from localhost (localhost [127.0.0.1]) by zp3.zcs.datasyncintra.net (Postfix) with ESMTP id 6ACFA20E4 for ; Tue, 8 Jun 2010 06:36:41 -0500 (CDT) Received: from zp3.zcs.datasyncintra.net ([127.0.0.1]) by localhost (zp3.zcs.datasyncintra.net [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 8skM3g7aG18i for ; Tue, 8 Jun 2010 06:36:41 -0500 (CDT) Received: from [192.168.1.109] (unknown [122.166.12.104]) (Authenticated sender: shehjart-+FkPdpiNhgJBDgjK7y7TUQ@public.gmane.org) by zp3.zcs.datasyncintra.net (Postfix) with ESMTPSA id 6212020E2 for ; Tue, 8 Jun 2010 06:36:40 -0500 (CDT) Sender: linux-nfs-owner@vger.kernel.org List-ID: Hi all I am working on a setup where when a nfsv3 server goes down, the dd that is running on the NFS mount point, gets an error on any syscall performed on that mount point, say a write () to a file on that mount. I've tried setting the timeo and retrans options at mount but that does not seem to work in this case. The how-to on linux-nfs site confirms that the nfs client just waits for the server to come back up after retrans number of requests. Is there a way to have the app receive an EIO or any other error on TCP disconnect? Thanks -Shehjar