Date: Mon, 11 Oct 2004 18:40:36 +0200 From: Benjamin Lutz <benlutz@datacomm.ch> To: freebsd-current@freebsd.org Subject: NFS trouble Message-ID: <200410111840.40750.benlutz@datacomm.ch>
next in thread | raw e-mail | index | archive | help
--nextPart26796955.qEen7xcciQ Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: quoted-printable Content-Disposition: inline Hello, I'm seeing a lot of processes getting stuck in "nfsfsync" state lately.=20 I'm running FreeBSD 5.3-BETA7, and I've seen this with both the i386 and=20 the amd64 version. As NFS server I'm using a FreeBSD 4.10 machine and an up to date gentoo=20 box (running Linux 2.4.25_pre7-gss-r11). I've mounted the NFS exports with a simple nfs_mount=20 uno:/var/www/localhost www as well as with some options that should make=20 nfs more robust: mount_nfs -D 2 -L -R 2 -i -s -x 3 uno:/var/www/localhost www In all cases, the process has started hanging when writing to the exported= =20 NFS share. This could be a text editor (kate), or cp. Subsequent access=20 (eg, with ls) to the directory where the first process tried to write to=20 will make the second process hang in state "nfs". Nothing gets logged on the server, the client logs this repeatedly: nfs server uno:/var/www/localhost: not responding =46orcibly unmounting the share works and seems to unfreeze the frozen=20 applications. This behaviour is new. I cannot say when exactly I've first seen this, but= =20 I think it must have appeared somewhere between BETA5 and BETA7. What can I do to make NFS work more reliably again? Benjamin --nextPart26796955.qEen7xcciQ Content-Type: application/pgp-signature -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.6 (FreeBSD) iD8DBQBBareIgShs4qbRdeQRAr1nAKCDPRrlcc1KccTtZ8TL0JRAcHawDwCeK5JQ xEQCEXxwx+gVlLrqYa+KWtw= =88x5 -----END PGP SIGNATURE----- --nextPart26796955.qEen7xcciQ--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200410111840.40750.benlutz>