From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeff Layton Subject: Re: [NFS] OpenSRS blog post alludes to a Linux NFS bug in 2.6.19 Date: Sun, 19 Oct 2008 06:44:05 -0400 Message-ID: <20081019064405.0db1e004@tleilax.poochiereds.net> References: <20081019032811.GA30801@outblaze.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Cc: linux-nfs@vger.kernel.org To: Yusuf Goolamabbas Return-path: Received: from mx2.redhat.com ([66.187.237.31]:47870 "EHLO mx2.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751168AbYJSKoL (ORCPT ); Sun, 19 Oct 2008 06:44:11 -0400 In-Reply-To: <20081019032811.GA30801-lA0CmJ9G6+FWk0Htik3J/w@public.gmane.org> Sender: linux-nfs-owner@vger.kernel.org List-ID: On Sun, 19 Oct 2008 11:28:11 +0800 Yusuf Goolamabbas wrote: > Anybody know if this is bug is fixed and if so in which kernel revision. > Any ideas about its impact on vendor kernels (RHEL/SUSE Enterprise) > > http://opensrs.com/blog/2008/10/technical-debrief-on-october-cluster-a-email-service-issue/ > I know that more recent RHEL kernels have this fixed. The bug to track it was here: https://bugzilla.redhat.com/show_bug.cgi?id=369561 It's also easy to tell if you have a kernel with this problem. Just do a single TCP mount on the client, and then run: $ rpcinfo -p ...you should see port registrations for nlockmgr. If you don't see any UDP ones, then the kernel has the problem. -- Jeff Layton