[Top] [All Lists]

Re: xfssyncd and disk spin down

To: xfs@xxxxxxxxxxx
Subject: Re: xfssyncd and disk spin down
From: Stan Hoeppner <stan@xxxxxxxxxxxxxxxxx>
Date: Thu, 23 Dec 2010 18:54:19 -0600
In-reply-to: <20101223211650.GA19694@xxxxxxxxxxxxxxxx>
References: <20101223165532.GA23813@xxxxxxxxxxxxxxxx> <4D13A30A.3090600@xxxxxxxxxxxxxxxxx> <20101223211650.GA19694@xxxxxxxxxxxxxxxx>
User-agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv: Gecko/20101207 Thunderbird/3.1.7
Petre Rodan put forth on 12/23/2010 3:16 PM:

> but my original mail had a different target really. I have to recognize that 
> I don't know much about the inner-workings of a filesystem, but I find it odd 
> that once there is no input from the outside, a process keeps writing to the 
> drive indefinitely. in my very narrow thinking the fact that these writes 
> dissapear after a remount would prove their redundance.

Ahh, quite right.  Sorry Petre.  This might shed some light on your
issue.  Old background thread:


Current documentation on this turnable knob:


fs.xfs.xfssyncd_centisecs       (Min: 100  Default: 3000  Max: 720000)
        The interval at which the xfssyncd thread flushes metadata
        out to disk.  This thread will flush log activity out, and
        do some processing on unlinked inodes.

Maybe related or also worth mentioning:

fs.xfs.age_buffer_centisecs     (Min: 100  Default: 1500  Max: 720000)
        The age at which xfsbufd flushes dirty metadata buffers to disk.

As of 2.6.36, it looks like you can stretch this out to a more
reasonable (for your application) 12 minutes if you like.  I'm not sure
if the max value is different than as I don't have a box or VM
available with  Try bumping these values to 720000 and see if
that allows your disk to sleep for 12 mins.


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