From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id p1AKgrdd155992 for ; Thu, 10 Feb 2011 14:42:54 -0600 Received: from skyrim.simplex.ro (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 92D9ED5CAE0 for ; Thu, 10 Feb 2011 12:45:28 -0800 (PST) Received: from skyrim.simplex.ro (skyrim.simplex.ro [109.169.51.118]) by cuda.sgi.com with ESMTP id q7AtaJwfhjGELLoB for ; Thu, 10 Feb 2011 12:45:28 -0800 (PST) Date: Thu, 10 Feb 2011 22:42:54 +0200 From: Petre Rodan Subject: Re: xfssyncd and disk spin down Message-ID: <20110210204254.GA10451@peter.simplex.ro> References: <20101223165532.GA23813@peter.simplex.ro> <20101227021904.GA24828@dastard> <20101227061629.GA2275@pandora.simplex.ro> <20101227140750.GB24828@dastard> <20101227171939.GA7759@pandora.simplex.ro> <20101231001323.GD15179@dastard> <20110120100143.GA2007@peter.simplex.ro> <20110120234310.GV16267@dastard> MIME-Version: 1.0 In-Reply-To: <20110120234310.GV16267@dastard> List-Id: XFS Filesystem from SGI List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============4003501150447557150==" Sender: xfs-bounces@oss.sgi.com Errors-To: xfs-bounces@oss.sgi.com To: Dave Chinner Cc: xfs@oss.sgi.com --===============4003501150447557150== Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="W/nzBZO5zC0uMSeA" Content-Disposition: inline --W/nzBZO5zC0uMSeA Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable Hello Dave, On Fri, Jan 21, 2011 at 10:43:10AM +1100, Dave Chinner wrote: > ..... > > in other words xfsyncd and xfsbufd now alternate at 18s intervals > > keeping the drive busy with nothing constructive hours after the > > last write to the drive. >=20 > This is a different problem, and not one I've seen before. Looking > at the traces, it appears that we have not empties the AIL. At > least, that's what I'm assuming at this point because log IO > completion is not updating log tail. When we start a log IO, we set > the log header lsn to the current head: >=20 > > xfssyncd/sdc1-1413 [000] 3356.093456: xfs_log_reserve: dev 8:33 ty= pe DUMMY1 t_ocnt 1 t_cnt 1 t_curr_res 2740 t_unit_res 2740 t_flags XLOG_TIC= _INITED reserveq empty writeq empty grant_reserve_cycle 2 grant_reserve_byt= es 428523008 grant_write_cycle 2 grant_write_bytes 428523008 curr_cycle 2 c= urr_block 836959 tail_cycle 2 tail_block 810683 >=20 > Which in this case is: curr_cycle 2 curr_block 836959 >=20 > When the log IO completes, that value gets written to the > l_last_sync_lsn. When the AIL tail is removed, the tail lsn is > updated to the new tail item. If the AIL is empty, then the > l_last_sync_lsn is used. That means then next dummy transaction > made to cover the log should have the cycle/block of the above > current cycle. >=20 > Instead, what I see is that the next dummmy transaction shows: >=20 > > xfssyncd/sdc1-1413 [000] 3392.067122: xfs_log_reserve: dev 8:33 ty= pe DUMMY1 t_ocnt 1 t_cnt 1 t_curr_res 2740 t_unit_res 2740 t_flags XLOG_TIC= _INITED reserveq empty writeq empty grant_reserve_cycle 2 grant_reserve_byt= es 428524032 grant_write_cycle 2 grant_write_bytes 428524032 curr_cycle 2 c= urr_block 836961 tail_cycle 2 tail_block 810683 >=20 > The current head has moved: curr_cycle 2 curr_block 836961 >=20 > But the tail hasn't: tail_cycle 2 tail_block 810683 >=20 > So effectively we've got some item on the AIL that we haven't > flushed and isn't being flushed by xfssyncd. That's the problem I > need to get to the bottom of and it also explains why it's an > intermitten problem... I have been watching the stream of patches that go into 2.6.38, but I proba= bly missed the one that might be the answer to the problem above. can you p= lease tell me which one to try, or can I help with anything? much appreciated, peter --W/nzBZO5zC0uMSeA Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.16 (GNU/Linux) iEYEABECAAYFAk1UTc4ACgkQixMPpwVd7zHJdgCdHM2LCCxx8R8BwV84/up+/UQa MuUAn13msnxU+izQdOnzZf02T7YQZByL =mSzf -----END PGP SIGNATURE----- --W/nzBZO5zC0uMSeA-- --===============4003501150447557150== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs --===============4003501150447557150==--