From: mehta kiran Subject: nfs lock failover : problem in connection Date: Wed, 18 Aug 2004 08:51:00 -0700 (PDT) Sender: nfs-admin@lists.sourceforge.net Message-ID: <20040818155100.86349.qmail@web51610.mail.yahoo.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Received: from sc8-sf-mx2-b.sourceforge.net ([10.3.1.12] helo=sc8-sf-mx2.sourceforge.net) by sc8-sf-list2.sourceforge.net with esmtp (Exim 4.30) id 1BxSiS-0002UV-6R for nfs@lists.sourceforge.net; Wed, 18 Aug 2004 08:51:08 -0700 Received: from web51610.mail.yahoo.com ([206.190.38.215]) by sc8-sf-mx2.sourceforge.net with smtp (Exim 4.34) id 1BxSiP-0007JQ-Iq for nfs@lists.sourceforge.net; Wed, 18 Aug 2004 08:51:08 -0700 To: nfs@lists.sourceforge.net Errors-To: nfs-admin@lists.sourceforge.net List-Unsubscribe: , List-Id: Discussion of NFS under Linux development, interoperability, and testing. List-Post: List-Help: List-Subscribe: , List-Archive: Hi , i am facing problem when nfs server fails over to other node on the cluster I have two node cluster and third node acts as client. Two nodes in the cluster share lock information by making link /var/lib/nfs on shared disk.Also filesystem to be exported are present on other partition of same shared disk. I am using virtual ip as well using which client access filesystem exported by nfs server Client mounts the exported filesystem. Then it runs a small sample code which takes lock on file present on exported filesystem , sleeps for some time and releases it. This works well 1. when nfs is restarted on same node on which it fails. 2. when program acquires and relases lock and NFS is killed on first node and started on other node. Client is still able to access old filesystem without remouting it But if following steps are followed , program on client side hangs 1. acquires lock on file present of exported filesystem 2. kill nfs on first cluster node before client program releases the lock and start nfs on second node Using tcpdump,netstat and debug messages produced by setting values in /proc/sys/sunrpc/nfsd_debug... , i have concluded that program on client tries to connect to server after server sends SM_NOTIFY message but connection is never established and hence program on client side is not able to regain the lock tcpdump shows that sync packet is being sent continuously to server but is not getting any reply Can anybody point out the problem ????? Awaiting the suggesstion thanks kiran __________________________________ Do you Yahoo!? New and Improved Yahoo! Mail - 100MB free storage! http://promotions.yahoo.com/new_mail ------------------------------------------------------- SF.Net email is sponsored by Shop4tech.com-Lowest price on Blank Media 100pk Sonic DVD-R 4x for only $29 -100pk Sonic DVD+R for only $33 Save 50% off Retail on Ink & Toner - Free Shipping and Free Gift. http://www.shop4tech.com/z/Inkjet_Cartridges/9_108_r285 _______________________________________________ NFS maillist - NFS@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nfs