From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Lisa Williams" Subject: Data Corruption Issues Date: Wed, 27 Aug 2003 10:51:44 -0700 Sender: nfs-admin@lists.sourceforge.net Message-ID: <008601c36cc3$e1ae87c0$d501010a@Lisa> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Return-path: Received: from sc8-sf-mx1-b.sourceforge.net ([10.3.1.11] helo=sc8-sf-mx1.sourceforge.net) by sc8-sf-list1.sourceforge.net with esmtp (Cipher TLSv1:DES-CBC3-SHA:168) (Exim 3.31-VA-mm2 #1 (Debian)) id 19s4Sw-0005cU-00 for ; Wed, 27 Aug 2003 10:52:18 -0700 Received: from [205.180.85.17] (helo=mail.fastclick.net) by sc8-sf-mx1.sourceforge.net with esmtp (Exim 4.22) id 19s4Sw-0001dC-73 for nfs@lists.sourceforge.net; Wed, 27 Aug 2003 10:52:18 -0700 Received: from Lisa (ip193.fastclick.net [205.180.85.193]) by mail.fastclick.net (8.12.9/8.12.5) with SMTP id h7RHpWd2003692 for ; Wed, 27 Aug 2003 17:51:32 GMT To: 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 all, We are finding corruption in mysql database files that are being copied from 16 Linux 2.4.20 servers to a Linux 2.4.21 backup server on an hourly basis. Here are the details: The backup server was originally a FreeBSD box that was having some hardware issues. We decided to swap it out and replace with a Linux server (running 2.4.21). We had no issues of data corruption until we swapped machines. The 16 linux 2.4.20 servers have the backup machine mounted locally and after each hour a script is run that flushes the tables and then copies them to the mounted backup drive. There's no real pattern of which machines create a corrupt file on the backup machine. It appears fairly random. We are seeing about 20% of them corrupted. Our network is running at 100MB. The files are around 25MB each. We have tried the following settings: Increasing the number of nfsd server threads to 20 on the backup machine. Changes in the mount options in the 16 fstabs: specifying: tcp, nfsvers=3 specifying: udp, nfsvers=2 specifying: rsize=1024,wsize=1024 There was no real change in the corruption rate after any of these changes. Any ideas of what we should try next are appreciated. Thank you, -Lisa ------------------------------------------------------- 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