Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 19 Feb 2011 16:52:43 -0500
From:      Daniel Staal <DStaal@usa.net>
To:        Robert Bonomi <bonomi@mail.r-bonomi.com>, freebsd-questions@freebsd.org
Subject:   Re: ZFS-only booting on FreeBSD
Message-ID:  <B365A66AECD410EA3539D5FB@mac-pro.magehandbook.com>
In-Reply-To: <201102192012.p1JKCKnP038248@mail.r-bonomi.com>
References:  <201102192012.p1JKCKnP038248@mail.r-bonomi.com>

next in thread | previous in thread | raw e-mail | index | archive | help
--As of February 19, 2011 2:12:20 PM -0600, Robert Bonomi is alleged to 
have said:

> A non-ZFS boot drive results in immediate, _guaranteed_, down-time for
> replacement if/when it fails.
>
> A ZFS boot drive lets you replace the drive and *schedule* the down-time
> (for a 'test' re-boot, to make *sure* everything works) at a convenient
> time.

--As for the rest, it is mine.

No it doesn't.  It only extends the next scheduled downtime until you deal 
with it.  ;)  (Or, in a hot-swap environment with sufficient monitoring, 
means you need to deal with it before the next scheduled downtime.)

Or, from what it sounds like, you could have a redundant/backup boot disk. 
I'm planning on using a $5 USB drive as my boot disk.  Triple redundancy 
would cost $15.  I paid more for lunch today.  (Hmm.  I'll have to test to 
see if that setup works, although given the rest of this discussion I don't 
see why it shouldn't...)

I see the advantage, and that it offers higher levels of resiliency and if 
properly handled should cause no problems.  I just hate relying on humans 
to remember things and follow directions.  That's what computers are for. 
Repairing a failed disk in a ZFS boot pool requires a human to remember to 
look for directions in an unusual place, and then follow them correctly. 
If they don't, nothing happens immediately, but there is the possibility of 
failure at some later unspecified time.  (Meanwhile if they look for 
directions in the *usual* place, they get a simple and straightforward set 
of instructions that will appear to work.)

*If* that failure occurs, that downtime will be longer than the downtime 
you would save from a dozen boxes being handled using the correct ZFS 
procedure, as everyone tears their hair out going 'Why doesn't it work?!? 
It worked just fine a moment ago!' until someone remembers this quirk.

I don't like quirky computers.  That's why I'm not a Windows admin.  ;)

Daniel T. Staal

---------------------------------------------------------------
This email copyright the author.  Unless otherwise noted, you
are expressly allowed to retransmit, quote, or otherwise use
the contents for non-commercial purposes.  This copyright will
expire 5 years after the author's death, or in 30 years,
whichever is longer, unless such a period is in excess of
local copyright law.
---------------------------------------------------------------



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?B365A66AECD410EA3539D5FB>