Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 25 Nov 2023 02:35:32 +0000
From:      bugzilla-noreply@freebsd.org
To:        fs@FreeBSD.org
Subject:   [Bug 275308] EN tracking issue for potential ZFS data corruption
Message-ID:  <bug-275308-3630-MwqUVkxGi8@https.bugs.freebsd.org/bugzilla/>
In-Reply-To: <bug-275308-3630@https.bugs.freebsd.org/bugzilla/>
References:  <bug-275308-3630@https.bugs.freebsd.org/bugzilla/>

next in thread | previous in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D275308

--- Comment #7 from Ed Maste <emaste@freebsd.org> ---
Based on the latest discussion in https://github.com/openzfs/zfs/issues/155=
26
my current understanding is:

- Setting sysctl vfs.zfs.dmu_offset_next_sync=3D0 is reported to make the i=
ssue
harder to reproduce, but does not reliably prevent it. A change to the defa=
ult
will *not* be in an EN update.
- https://github.com/openzfs/zfs/pull/15571 is the actual fix for this issu=
e.
- Pull request 15571 needs to land in OpenZFS upstream first. Once that hap=
pens
we can import it, MFC to stable/14, and issue an EN to update 14.0 and 13.2.
- dnode_is_dirty was introduced in commit de198f2d9507 in OpenZFS, and does=
 not
exist in FreeBSD 12.4. This issue may not affect 12.x.

--=20
You are receiving this mail because:
You are the assignee for the bug.=



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-275308-3630-MwqUVkxGi8>