xfs
[Top] [All Lists]

Re: XFS Kernel Panics in CentOS

To: Mark Rechler <mrechler@xxxxxxxxxxxxxx>
Subject: Re: XFS Kernel Panics in CentOS
From: Stefan Ring <stefanrin@xxxxxxxxx>
Date: Fri, 29 Jun 2012 23:38:58 +0200
Cc: "xfs@xxxxxxxxxxx" <xfs@xxxxxxxxxxx>
Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=gwMI7cWhoP9NvHDEqHUh+TeYEwPBFcWbOK3sWwSr/Hg=; b=w6FIv+w7XCwApafmuPor/HiEExuU+K8dWrICH4SP1lx0lX9hF9S8JH14l34lkiJk2q ZJzRefuRQMrnKXynLur5XwxAj5bbAVAvDgjt23B7/xsLH1H61hep7A00MFVMvtKbymew WNVdRNJXaOcXSGXr/bD8BJRKCbu1gE/vwDukr8Yh3Hoi3Rp84LZefm0OkVE2XQu1j1be T7N4GVv2wVMqhkWzf39FVBU5ouxUUiFzzvuyofs/PR21fETQCT5zAUD+ssQ+3WH+/9cK PoknPQd/dxsebhUsjqpMRjzt47QfqNq8eCL6w/85ZdHm52YTCQ44qD+E3Ar9jd+yRGt1 YEYA==
In-reply-to: <CANoMt5OqDhkgvux7fOqp7ZEUDHrdvt2NtKszebFeheApi+RF0g@xxxxxxxxxxxxxx>
References: <CANoMt5NBie68Auyj_V-6QZBbtLPzKu_sWsTFCceO3dYhJu5uZg@xxxxxxxxxxxxxx> <4F763734.9040906@xxxxxxxxxxx> <CANoMt5O5T+FTrmL-2c5tsT=wm434iVNeFarPdybcs1q_oPtW=g@xxxxxxxxxxxxxx> <4F79E9F3.6010305@xxxxxxxxxxx> <CAPnto6zUScA9u0rBW4i-BqeBnY1MK09J_H8J2R1KCtE1cYy0Rg@xxxxxxxxxxxxxx> <6B9C8E2F-0AD3-4C99-A671-5FB603F8577A@xxxxxxxxxxx> <CAPnto6yxng9chxFW6UogxUMKuNAFqu6cJY5xgax0dgsNucosNg@xxxxxxxxxxxxxx> <CANoMt5OqDhkgvux7fOqp7ZEUDHrdvt2NtKszebFeheApi+RF0g@xxxxxxxxxxxxxx>
> It turned out in my case to be related to:
> http://oss.sgi.com/bugzilla/show_bug.cgi?id=840
>
> Write barriers were not passed when using LVM/XFS/MegaRAID combined. After
> upgrading the kernel to 2.6.39 (used packages from
> http://elrepo.org/tiki/tiki-index.php) all XFS issues were resolved. The
> other solution was not using LVM.

I fail to see how missing write barriers can make a difference to what
the filesystem reads from the block layer. How is this possible? Does
the block layer not take the pending writes into account when handling
read requests?

<Prev in Thread] Current Thread [Next in Thread>