Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 08 Jul 2015 09:22:46 -0500
From:      Karl Denninger <karl@denninger.net>
To:        freebsd-fs@freebsd.org
Subject:   Re: Speeding up resilvering
Message-ID:  <559D3236.1060102@denninger.net>
In-Reply-To: <559D3167.1000705@infracaninophile.co.uk>
References:  <559D2AB6.5070007@FreeBSD.org> <559D2C7C.6060201@denninger.net> <559D3167.1000705@infracaninophile.co.uk>

next in thread | previous in thread | raw e-mail | index | archive | help
This is a cryptographically signed message in MIME format.

--------------ms080401080600040505030604
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable



On 7/8/2015 09:19, Matthew Seaman wrote:
> On 2015/07/08 14:58, Karl Denninger wrote:
>> On 7/8/2015 08:50, Matthew Seaman wrote:
>>> Hi,
>>>
>>> I've a zpool which is taking an inordinately long time to resilver an=
d
>>> replace a device.  It had only got to about 8% completion after a day=
,
>>> implying over a week to resilver about 6TB data.
>>>
>>> Now, I've applied the resilver performance tuning sysctls from
>>> https://wiki.freebsd.org/ZFSTuningGuide:
>>>
>>> vfs.zfs.scrub_delay=3D0
>>> vfs.zfs.top_maxinflight=3D128
>>> vfs.zfs.resilver_min_time_ms=3D5000
>>> vfs.zfs.resilver_delay=3D0
>>>
>>> but it doesn't seem to have made a great deal of difference.  Most of=

>>> the benefit there would apparently come from reducing scrub_delay or
>>> resilver_delay -- but this system is only booted to single user mode,=
 so
>>> the drives are otherwise idle and the resilver should have automatica=
lly
>>> switched to running full throttle anyhow.
>>>
>>> Given the machine is not going to be doing anything else other than
>>> resilvering for the time being, is there any more aggressive tuning o=
r
>>> tricks to get it to go faster that people would recommend?
>>>
>>> 	Cheers,
>>>
>>> 	Matthew
>>>
>> What OS version?
>>
> 10.1-RELEASE-p10
>
> 	Cheers,
>
> 	Matthew
Look at the IO saturation on the disk channel(s) involved with either
systat -vm or iostat.  If the channel is saturated then there's nothing
you can do in terms of tuning; the question then turns to why actual I/O
performance is so poor and has to be addressed there.

--=20
Karl Denninger
karl@denninger.net <mailto:karl@denninger.net>
/The Market Ticker/
/[S/MIME encrypted email preferred]/

--------------ms080401080600040505030604
Content-Type: application/pkcs7-signature; name="smime.p7s"
Content-Transfer-Encoding: base64
Content-Disposition: attachment; filename="smime.p7s"
Content-Description: S/MIME Cryptographic Signature
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--------------ms080401080600040505030604--





Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?559D3236.1060102>