Date: Mon, 7 Jun 2010 17:55:28 +0100 From: Martin Simmons <martin@lispworks.com> To: freebsd-fs@freebsd.org Subject: Re: zfs i/o error, no driver error Message-ID: <201006071655.o57GtSBg029967@higson.cam.lispworks.com> In-Reply-To: <20100607121954.GA52932@icarus.home.lan> (message from Jeremy Chadwick on Mon, 7 Jun 2010 05:19:54 -0700) References: <4C0CAABA.2010506@icyb.net.ua> <20100607083428.GA48419@icarus.home.lan> <4C0CB3FC.8070001@icyb.net.ua> <20100607090850.GA49166@icarus.home.lan> <201006071112.o57BCGMf027496@higson.cam.lispworks.com> <20100607121954.GA52932@icarus.home.lan>
next in thread | previous in thread | raw e-mail | index | archive | help
>>>>> On Mon, 7 Jun 2010 05:19:54 -0700, Jeremy Chadwick said:
>
> Which brings us to the topic of scrub interval...
>
> This exact question was asked on the ZFS OpenSolaris list[6] in late
> 2008, and nobody there provided any concrete evidence either. The
> closest thing to evidence is this:
>
> "...in normal operation, ZFS only checks data as it's read back from the
> disks. If you don't periodically scrub, errors that happen over time
> won't be caught until I next read that actual data, which might be
> inconvenient if it's a long since the initial data was written".
The question can't be answered with absolute numbers, because it depends on
other factors such as environmental effects.
> The topic of scrub intervals was also brought up a month later[7].
> Someone said:
>
> "We did a study on re-write scrubs which showed that once per year was a
> good interval for modern, enterprise-class disks. However, ZFS does a
> read-only scrub, so you might want to scrub more often".
>
> The first part conflicts with what the guide recommends (I'd also like
> to see the results of the study!), while the last half of the paragraph
> makes no sense ("because it reads, do it more often!"). So if you take
> the first sentence and apply it to what the ZFS Best Practices Guide
> says, you come out with... "scrub consumer-grade disks every 6 months".
It doesn't conflict if you agree that freshly written data is more likely to
be readable that data written long ago (with some curve in between).
The re-write scrub they are talking about will write all of the data back to
the disks during the scrubbing operation, which makes it fresher.
ZFS OTOH performs read-only scrubs, i.e. it just checks that the data can be
read. It only writes if there was a problem reading from one of the disks.
I don't know if there is any science behind that theory...
__Martin
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201006071655.o57GtSBg029967>