xfs
[Top] [All Lists]

Re: xfssyncd and disk spin down

To: Dave Chinner <david@xxxxxxxxxxxxx>
Subject: Re: xfssyncd and disk spin down
From: Petre Rodan <petre.rodan@xxxxxxxxxx>
Date: Thu, 20 Jan 2011 14:07:00 +0200
Cc: xfs@xxxxxxxxxxx
In-reply-to: <20110120110605.GU16267@dastard>
References: <20101223165532.GA23813@xxxxxxxxxxxxxxxx> <20101227021904.GA24828@dastard> <20101227061629.GA2275@xxxxxxxxxxxxxxxxxx> <20101227140750.GB24828@dastard> <20101227171939.GA7759@xxxxxxxxxxxxxxxxxx> <20101231001323.GD15179@dastard> <20110120100143.GA2007@xxxxxxxxxxxxxxxx> <20110120110605.GU16267@dastard>
User-agent: Mutt/1.5.21 (2010-09-15)
Hello Dave,

On Thu, Jan 20, 2011 at 10:06:05PM +1100, Dave Chinner wrote:
> ....
> > in other words xfsyncd and xfsbufd now alternate at 18s intervals
> > keeping the drive busy with nothing constructive hours after the
> > last write to the drive.
[..]
> > meta-data=/dev/sdc1              isize=256    agcount=4, agsize=61047552 
> > blks
> >          =                       sectsz=512   attr=2
> > data     =                       bsize=4096   blocks=244190208, imaxpct=25
> >          =                       sunit=0      swidth=0 blks
> > naming   =version 2              bsize=4096   ascii-ci=0
> > log      =internal               bsize=4096   blocks=119233, version=2
> >          =                       sectsz=512   sunit=0 blks, lazy-count=0
>                                                               ^^^^^^^^^^^^
> > realtime =none                   extsz=4096   blocks=0, rtextents=0
> 
> It's something to do with lazy-count=0. I'm look into it when I get
> the chance - I almost never test w/ lazy-count=0 because =1 is
> the default value.

thanks for the tip. looks like I'm using xfsprogs 3.0.3 (latest version in the 
gentoo stable tree) to mkfs these filesystems. can the antique version be 
somehow the root of the problem?

anyhow, changed it to:

meta-data=/dev/sdc1              isize=256    agcount=4, agsize=61047552 blks
         =                       sectsz=512   attr=2
data     =                       bsize=4096   blocks=244190208, imaxpct=25
         =                       sunit=0      swidth=0 blks
naming   =version 2              bsize=4096   ascii-ci=0
log      =internal               bsize=4096   blocks=119233, version=2
         =                       sectsz=512   sunit=0 blks, lazy-count=1
realtime =none                   extsz=4096   blocks=0, rtextents=0

used:
xfsprogs-3.0.3
Linux starshine 2.6.38-rc1-s001 #1 SMP Thu Jan 20 10:49:20 EET 2011 x86_64 AMD 
Athlon(tm) II Neo N36L Dual-Core Processor AuthenticAMD GNU/Linux
Seagate Barracuda ES.2 ST31000340NS
Hitachi HUA722020ALA330
default (for 3.0.3 xfsprogs) mkfs options, 'noatime,nodev,nosuid,noexec' as 
mount options.

and the behaviour did not change at all, trace attached.

it's very easy to reproduce this. all it took was

reboot
cp -a /usr/src/linux* /mnt/foo
sleep 1800
grab trace


cheers,
peter


> 
> I'd recommend that you convert the fs to lazy-count=1 when you get a
> chance, anyway, because of the fact it reduces the latency of
> transactions significantly...
> 
> Cheers,
> 
> Dave.
> -- 
> Dave Chinner
> david@xxxxxxxxxxxxx

-- 
petre rodan
<petre.rodan@xxxxxxxxxx>
Technical Manager
Simplex SRL, Bucharest

Attachment: trace
Description: Text document

Attachment: pgpfgJZ7xdPXS.pgp
Description: PGP signature

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