Date: Mon, 15 Jun 2009 15:05:07 +0200 (CEST) From: Martin Matuska <mm@FreeBSD.org> To: FreeBSD-gnats-submit@FreeBSD.org Subject: kern/135594: [zfs] Single dataset unresponsive with Samba Message-ID: <20090615130507.572A614F0E@mail.vx.sk> Resent-Message-ID: <200906151310.n5FDA2NJ073150@freefall.freebsd.org>
next in thread | raw e-mail | index | archive | help
>Number: 135594
>Category: kern
>Synopsis: [zfs] Single dataset unresponsive with Samba
>Confidential: no
>Severity: serious
>Priority: low
>Responsible: freebsd-bugs
>State: open
>Quarter:
>Keywords:
>Date-Required:
>Class: sw-bug
>Submitter-Id: current-users
>Arrival-Date: Mon Jun 15 13:10:02 UTC 2009
>Closed-Date:
>Last-Modified:
>Originator: Martin Matuska
>Release: FreeBSD 7.2-STABLE amd64
>Organization:
>Environment:
Release: 7.2-STABLE-amd64
ZFS: v13
Memory: 8GB
Samba: 3.2.10, 3.2.11
vfs.zfs.prefetch_disable="1"
>Description:
I experience hangs of a single ZFS dataset when under low load with samba.
Processes trying to operate on the dataset are stuck in ZFS state ("smbd", "ls", etc.),
this applies to "zfs umount -f" as well. Other datasets operate without problems.
zfs list, set, etc. work correctly with this dataset.
A hard reset is required to have the system operating again.
This error happened on two systems with different datasets so it is reproducible,
I should try to define how to reproduce it.
How can I help debugging this problem?
>How-To-Repeat:
>Fix:
>Release-Note:
>Audit-Trail:
>Unformatted:
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20090615130507.572A614F0E>