From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from int-mailstore01.merit.edu ([207.75.116.232]:47565 "EHLO int-mailstore01.merit.edu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752464Ab1CaQw7 (ORCPT ); Thu, 31 Mar 2011 12:52:59 -0400 Date: Thu, 31 Mar 2011 12:52:58 -0400 From: Jim Rees To: linux-nfs@vger.kernel.org Cc: peter honeyman Subject: client hang in lock_flocks() Message-ID: <20110331165258.GA3048@merit.edu> Content-Type: text/plain; charset=us-ascii Sender: linux-nfs-owner@vger.kernel.org List-ID: MIME-Version: 1.0 I have reported here before on a NFS client hang that we see in multi-threaded iozone testing on various kernel and NFS versions. What looks like the same bug has been reported against RHEL here: https://bugzilla.redhat.com/show_bug.cgi?id=672305 After applying the suggested fix from that bug report I'm getting a different hang, at least with Benny's pnfs-all-latest and NFS 4.1 (no pnfs). Does this ring a bell with anyone? INFO: task iozone:2666 blocked for more than 120 seconds. "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. iozone D ffff88024cdf1aa0 0 2666 1 0x00000080 ffff880246be7c50 0000000000000082 0000000000000000 0000000000000000 ffff880246be6010 ffff880246be7fd8 00000000000136c0 ffff88024cdf16f0 ffff88024cdf1aa8 ffff88024cdf1aa0 00000000000136c0 00000000000136c0 Call Trace: [] ? lock_flocks+0x10/0x12 [] rwsem_down_failed_common+0xd3/0x105 [] rwsem_down_read_failed+0x12/0x14 [] call_rwsem_down_read_failed+0x14/0x30 [] ? down_read+0x32/0x36 [] nfs4_proc_lock+0x224/0x342 [nfs] [] do_setlk+0x61/0xd5 [nfs] [] nfs_lock+0x170/0x17f [nfs] [] vfs_lock_file+0x24/0x38 [] fcntl_setlk+0x15e/0x2a6 [] ? fsnotify_access+0x5d/0x65 [] sys_fcntl+0x30e/0x4af [] ? do_device_not_available+0x9/0xb [] system_call_fastpath+0x16/0x1b