Return-Path: Received: from peace.netnation.com ([204.174.223.2]:40645 "EHLO peace.netnation.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752394Ab1CLBJc (ORCPT ); Fri, 11 Mar 2011 20:09:32 -0500 Date: Fri, 11 Mar 2011 17:09:30 -0800 From: Simon Kirby To: "J. Bruce Fields" Cc: Mark Moseley , linux-nfs@vger.kernel.org, Al Viro , linux-kernel@vger.kernel.org, linux-fsdevel@vger.kernel.org Subject: Re: [git pull] more vfs fixes for final Message-ID: <20110312010930.GD8211@hostway.ca> References: <20110310110931.GF22723@ZenIV.linux.org.uk> <20110310115856.GG22723@ZenIV.linux.org.uk> <20110311210938.GB8211@hostway.ca> <20110311213519.GA9404@fieldses.org> Content-Type: text/plain; charset=us-ascii In-Reply-To: <20110311213519.GA9404@fieldses.org> Sender: linux-nfs-owner@vger.kernel.org List-ID: MIME-Version: 1.0 On Fri, Mar 11, 2011 at 04:35:19PM -0500, J. Bruce Fields wrote: > On Fri, Mar 11, 2011 at 01:09:38PM -0800, Simon Kirby wrote: > > On Thu, Mar 10, 2011 at 11:58:56AM +0000, Al Viro wrote: > > > > > commit d891eedbc3b1b0fade8a9ce60cc0eba1cccb59e5 > > > Author: J. Bruce Fields > > > Date: Tue Jan 18 15:45:09 2011 -0500 > > > > > > fs/dcache: allow d_obtain_alias() to return unhashed dentries > > > > Hmm, I was hoping this or something recently would fix nfs_inode_cache > > growing forever and flush processes taking lots of system time since > > 2.6.36. For example: > > > > OBJS ACTIVE USE OBJ SIZE SLABS OBJ/SLAB CACHE SIZE NAME > > 3457486 3454365 99% 0.95K 105601 33 3379232K nfs_inode_cache > > 469638 248761 52% 0.10K 12042 39 48168K buffer_head > > 243712 216348 88% 0.02K 952 256 3808K kmalloc-16 > > 232785 202185 86% 0.19K 11085 21 44340K dentry > > 149696 54633 36% 0.06K 2339 64 9356K kmalloc-64 > > 115976 106806 92% 0.55K 4142 28 66272K radix_tree_node > > 76064 45680 60% 0.12K 2377 32 9508K kmalloc-128 > > 62336 53427 85% 0.03K 487 128 1948K kmalloc-32 > > 41958 41250 98% 0.75K 1998 21 31968K ext3_inode_cache > > > > This clears them all, similar to what you posted: > > > > echo 2 > /proc/sys/vm/drop_caches > > sync > > echo 2 > /proc/sys/vm/drop_caches > > > > ...but 2.6.38-rc8 still doesn't seem to fix it. > > > > http://0x.ca/sim/ref/2.6.37/cpu3_nfs.png > > http://www.spinics.net/lists/linux-nfs/msg18212.html > > > > Any ideas? This started with 2.6.36. > > Do you have NFSv4 clients that are doing locking? Then it's probably > 0997b17360 and 529d7b2a7f on the for-2.6.39 branch at: > > git://linux-nfs.org/~bfields/linux.git for-2.6.39 > > Let me know if not. Yes, but when this started, it was all NFSv3. I tried NFSv4 to see if it made any different (no, other than more I/O wait due to idmapd and flock(LOCK_EX) breaking on a file opened O_RDONLY, which works locally and on NFSv3 -- fixed by changing to open(O_RDWR) in my code), but otherwise NFSv3 and NFSv4 look pretty much the same and "leak" at the same rate. Actually, I only changed the mount where it writes the log results, and not where it reads them from (a bunch of NFSv3 mounts that will take a lot more work to make NFSv4 since they currently run without libnss-mysql set up), so I haven't actually tested to see if it happens with purely NFSv4 mounts. Anyway, it's locking on one file only, and it was purely for reducing writeback thrashing, so I could try without it if you think it might be related, but it all started without NFSv4. All this thing does is read a bunch of log files, crunch them into reports, and write them to another mount point, all run from xargs -P. Pretty basic readdir, read, write, unlink...not anything exotic. Simon-