Date: Fri, 9 Jun 2006 11:13:59 +0300 From: Konstantin Belousov <kostikbel@gmail.com> To: Mikhail Teterin <mi+mx@aldan.algebra.com> Cc: fs@freebsd.org Subject: Re: heavy NFS writes lead to corrup summary in superblock Message-ID: <20060609081359.GH54415@deviant.kiev.zoral.com.ua> In-Reply-To: <200606081346.04908.mi%2Bmx@aldan.algebra.com> References: <200606081346.04908.mi%2Bmx@aldan.algebra.com>
next in thread | previous in thread | raw e-mail | index | archive | help
--WIIRZ1HQ6FgrlPgb Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Thu, Jun 08, 2006 at 01:46:04PM -0400, Mikhail Teterin wrote: > Our amd64 6.1-STABLE system is used to collect backup dumps from producti= on=20 > systems (mostly -- Solaris) via NFS. When in progrss, the dumps arrive at= an=20 > average rate of 20Mb/s. >=20 > Every once in a while I notice a discrepancy in the amount of used space = on=20 > the backup FS as reported by df vs. that reported by the total du. >=20 > Unmounting the FS and fsck-ing it fixes the problem with fsck reporting= =20 > (despite the clean unmount): >=20 > SUMMARY BLK COUNT(S) WRONG IN SUPERBLK > SALVAGE? yes >=20 > The FS is intended for very few very large files and was created=20 > with "newfs -b 65536 -O1" (no softupdates). >=20 > This workaround (explicit fsck) is acceptable for us, but it is a sign of= some=20 > kind of rot, and I thought, you'd like to know... >=20 > Yours, >=20 > -mi Wild guess: try rev. 1.673 of the sys/kern/vfs_subr.c. --WIIRZ1HQ6FgrlPgb Content-Type: application/pgp-signature Content-Disposition: inline -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.3 (FreeBSD) iD8DBQFEiS3GC3+MBN1Mb4gRAn4NAKDOUaK171CGo8W458X0GrUgqzO1qgCfdJLC c5R3cNnLyJRu4jFqWOnxDkk= =Xq7v -----END PGP SIGNATURE----- --WIIRZ1HQ6FgrlPgb--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060609081359.GH54415>