[Top] [All Lists]

Re: XFS performance issue solved

To: Austin Gonyou <austin@xxxxxxxxxxxxxxx>
Subject: Re: XFS performance issue solved
From: Walt H <waltabbyh@xxxxxxxxxx>
Date: Tue, 15 May 2001 06:35:18 -0700
Cc: linux-xfs@xxxxxxxxxxx
References: <Pine.LNX.4.33.0105142345310.7573-100000@UberGeek.coremetrics.com>
Sender: owner-linux-xfs@xxxxxxxxxxx
User-agent: Mozilla/5.0 (X11; U; Linux 2.4.2-XFS i686; en-US; rv:0.9+) Gecko/20010513
That's the way I've always looked at it. I've always thought the -W1 flag was redudant, as well as a bit dangerous. If you've already enabled asynchrounous access, you have "lazy" writes from the kernel. The only other thing I've got to add, is that the only time I've had FS corruption problems using XFS is when I experimented with -W1 and had an unclean shutdown. Not worth it in my opinion.


Austin Gonyou wrote:

This is agreed, about the -W1 flag. Wouldn't asynchronous IO solve this

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