From: Andre Noll Subject: Re: ext4: (2.6.34-rc4): This should not happen!! Data will be lost Date: Sun, 18 Apr 2010 00:38:54 +0200 Message-ID: <20100417223854.GD25507@skl-net.de> References: <20100416123526.GW21495@skl-net.de> <20100416163654.GD58339@plapa.qlogic.org> <20100416170707.GB25507@skl-net.de> <201004171855.36874.bernd.schubert@fastmail.fm> <4BCA1DFB.5030501@redhat.com> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="G2kvLHdEX2DcGdqq" Cc: Bernd Schubert , Andrew Vasquez , "linux-ext4@vger.kernel.org" , Linux Driver , Thomas Helle To: Eric Sandeen Return-path: Received: from systemlinux.org ([83.151.29.59]:46389 "EHLO m18s25.vlinux.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755447Ab0DQWjW (ORCPT ); Sat, 17 Apr 2010 18:39:22 -0400 Content-Disposition: inline In-Reply-To: <4BCA1DFB.5030501@redhat.com> Sender: linux-ext4-owner@vger.kernel.org List-ID: --G2kvLHdEX2DcGdqq Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On 15:45, Eric Sandeen wrote: > Interesting that it only happens at enospc time though - the fs would > be sending barriers in general usage. >=20 > Although, as ext4 fills, it does do more syncing/flushing to try to > eke out that last bit of space.... hmm.... >=20 > Andre, just for fun, you might try mounting -o nodelalloc, fill it again, > and see if you see the same behavior. Will do. This will take some time though as I destroyed the fs by writing directly to the device, so I had to recreate it and start from scratch. I'm currently filling the new fs using a 60s device timeout as suggested by Andrew. > If not, you might do it again w/ defaults, and capture blktrace data, whi= ch > I think will trace barrier requests as well. >=20 > (or, maybe also try testing with -o nobarrier) >=20 > I still don't think it's likely a filesystem problem but maybe you can > pinpoint the fs behavior that triggers it. I'll try to reproduce the problem using different timeout values and the ext4 options you suggest. If I can find a reliable reproducer, I'll run blktrace and post the results. Regards Andre --=20 The only person who always got his work done by Friday was Robinson Crusoe --G2kvLHdEX2DcGdqq Content-Type: application/pgp-signature; name="signature.asc" Content-Description: Digital signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.1 (GNU/Linux) iD8DBQFLyjh+Wto1QDEAkw8RAi9SAJ9/pa1ja7ffAH09x6vzp6YGKxrpxACfZq4J vVdxu7seFQwRavahwZNBAnc= =MlhY -----END PGP SIGNATURE----- --G2kvLHdEX2DcGdqq--