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 p567TDZj212734 for ; Mon, 6 Jun 2011 02:29:13 -0500 Received: from mailsrv14.zmi.at (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 4A12D16A3DF6 for ; Mon, 6 Jun 2011 00:29:10 -0700 (PDT) Received: from mailsrv14.zmi.at (mailsrv1.zmi.at [212.69.164.54]) by cuda.sgi.com with ESMTP id a6nmBnQSK8vvJfqB for ; Mon, 06 Jun 2011 00:29:10 -0700 (PDT) From: Michael Monnerie Subject: Re: I/O hang, possibly XFS, possibly general Date: Mon, 6 Jun 2011 09:29:01 +0200 References: <19945.24042.711472.158523@tree.ty.sabi.co.UK> In-Reply-To: <19945.24042.711472.158523@tree.ty.sabi.co.UK> MIME-Version: 1.0 Message-Id: <201106060929.06814@zmi.at> List-Id: XFS Filesystem from SGI List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: multipart/mixed; boundary="===============5015902640068812367==" Sender: xfs-bounces@oss.sgi.com Errors-To: xfs-bounces@oss.sgi.com To: xfs@oss.sgi.com --===============5015902640068812367== Content-Type: multipart/signed; boundary="nextPart3157995.l45fAgWuT4"; protocol="application/pgp-signature"; micalg=pgp-sha1 Content-Transfer-Encoding: 7bit --nextPart3157995.l45fAgWuT4 Content-Type: Text/Plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable On Samstag, 4. Juni 2011 Peter Grandi wrote: > vm/dirty_ratio=3D2 > vm/dirty_bytes=3D400000000 >=20 > vm/dirty_background_ratio=3D60 > vm/dirty_background_bytes=3D0 >=20 > vm/dirty_expire_centisecs=3D200 > vm/dirty_writeback_centisecs=3D400 Why dirty_background_ratio=3D60? This would mean you start to write dirty=20 pages only after it reaches 60% of total system memory... Setting it to=20 =3D1 would be the thing you want I guess. Also, setting both dirty_background_(ratio|bytes) is not supported. The=20 latter wins, according to sysctl/vm.txt Similarly, dirty_ratio and dirty_bytes belong together and exclude each=20 other. Maybe you specified both to fit older and newer kernels in one=20 example? dirty_expire_centisecs to 200 means a sync every 2s, which might be good=20 in this specific setup mentioned here, but not for a generic server.=20 That would defeat XFS's in-memory grouping of blocks before writeout,=20 and in case of many parallel (slow|ftp) uploads could lead to much more=20 data fragmentation, or no? =2D-=20 mit freundlichen Gr=FCssen, Michael Monnerie, Ing. BSc it-management Internet Services: Prot=E9ger http://proteger.at [gesprochen: Prot-e-schee] Tel: +43 660 / 415 6531 // Haus zu verkaufen: http://zmi.at/langegg/ --nextPart3157995.l45fAgWuT4 Content-Type: application/pgp-signature; name=signature.asc Content-Description: This is a digitally signed message part. -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.16 (GNU/Linux) iEYEABECAAYFAk3sgcIACgkQzhSR9xwSCbSFlgCdE7uBglNfxsBIxydx51KrPz7/ 0+8AoLF1FS4tQ+Syo6fVB78x6DVmS1TF =EbKb -----END PGP SIGNATURE----- --nextPart3157995.l45fAgWuT4-- --===============5015902640068812367== 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 --===============5015902640068812367==--