From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: with ECARTIS (v1.0.0; list xfs); Thu, 04 Oct 2007 01:42:44 -0700 (PDT) Received: from fk-out-0910.google.com (fk-out-0910.google.com [209.85.128.188]) by oss.sgi.com (8.12.11.20060308/8.12.10/SuSE Linux 0.7) with ESMTP id l948gbpV016409 for ; Thu, 4 Oct 2007 01:42:40 -0700 Received: by fk-out-0910.google.com with SMTP id 18so87669fks for ; Thu, 04 Oct 2007 01:42:37 -0700 (PDT) From: Alessandro Bono Subject: Re: 2.6.23-rc9-git1 hang with XFS Date: Thu, 4 Oct 2007 09:39:25 +0200 References: <20071004003935.GC23367404@sgi.com> In-Reply-To: <20071004003935.GC23367404@sgi.com> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200710040939.26139.alessandro.bono@gmail.com> Sender: xfs-bounce@oss.sgi.com Errors-to: xfs-bounce@oss.sgi.com List-Id: xfs To: David Chinner Cc: linux-xfs@oss.sgi.com On Thursday 04 October 2007, David Chinner wrote: > On Wed, Oct 03, 2007 at 11:03:29PM +0200, Alessandro Bono wrote: > > Hi all > > > > I'm testing 2.6.23-rc9-git1 on my old home server > > Trying to reorganizer my xfs filesystem with xfs_fsr cause a sort of > > system hang > > xfs_fsr is waiting for a direct I/O to complete. > Other processes are waiting for the superblock I/O to compete > (which is why writes are hanging), an dothers are waiting > for iclog space to be freed up. I think something is holding > > off I/O completion, and this: > > possible SYN flooding on port 4664. Sending cookies. > > possible SYN flooding on port 4664. Sending cookies. > > possible SYN flooding on port 4664. Sending cookies. > > possible SYN flooding on port 4664. Sending cookies. > > possible SYN flooding on port 4664. Sending cookies. > > indicates someone trying a DOS on your box. Perhaps it's > succeeding by denying interrupt service to you disks? it's really difficult, my internet line is a normal adsl2 and this machine is really rock solid with ubuntu dapper kernel (uptime of several months and mldonkey, mlnet process, is always on) also after reboot raid5 start resyncronizing disk and after 2 hours system ends this operation without any problem (mldonkey on) > > If you pull the network cable, does the system start to respond > properly again? I'll try this evening Thanks > > Cheers, > > Dave. -- Cordiali saluti Alessandro Bono