xfs
[Top] [All Lists]

Re: xfssyncd and disk spin down

To: Petre Rodan <petre.rodan@xxxxxxxxxx>
Subject: Re: xfssyncd and disk spin down
From: Dave Chinner <david@xxxxxxxxxxxxx>
Date: Tue, 28 Dec 2010 01:07:50 +1100
Cc: xfs@xxxxxxxxxxx
In-reply-to: <20101227061629.GA2275@xxxxxxxxxxxxxxxxxx>
References: <20101223165532.GA23813@xxxxxxxxxxxxxxxx> <20101227021904.GA24828@dastard> <20101227061629.GA2275@xxxxxxxxxxxxxxxxxx>
User-agent: Mutt/1.5.20 (2009-06-14)
On Mon, Dec 27, 2010 at 08:16:29AM +0200, Petre Rodan wrote:
> 
> Hello Dave,
> 
> On Mon, Dec 27, 2010 at 01:19:04PM +1100, Dave Chinner wrote:
> > On Thu, Dec 23, 2010 at 06:55:32PM +0200, Petre Rodan wrote:
> > > 
> > > Hello,
> [..]
> > > upon closer inspection it turns out that after the first Write
> > > action to that partition, an xfssyncd process continues to write
> > > to that partition each 36 seconds and it doesn't stop doing that,
> > > even if there are no more Writes from the exterior. this keeps the
> > > drive busy with varying consequences. more about that later.
> > 
> > Should be fixed in 2.6.37 by the commit:
> > 
> > 1a387d3 xfs: dummy transactions should not dirty VFS state
> 
> nice that someone else was also keen on having this bug fixed, but this 
> particular fix is already present in 2.6.36 [1], [2]
> 
> unfortunately something is still written to the log every 36s, as per the 
> first mail in this thread.

Turn on the XFS tracing so we can see what is being written every
36s.  When the problem shows up:

# echo 1 > /sys/kernel/debug/tracing/events/xfs/enable
# sleep 100
# cat /sys/kernel/debug/tracing/trace > trace.out
# echo 0 > /sys/kernel/debug/tracing/events/xfs/enable

And post the trace.out file for us to look at.

Cheers,

Dave.



-- 
Dave Chinner
david@xxxxxxxxxxxxx

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