Re: [PLUG] RAID6 or RAID5+HS?

["Keith C. Perry" <kperry@daotechnologies.com>]

My wording is awkward...

What was on my mind was this...  Lets not go crazy of the new shiny thing.  Yes ZFS (and BTRFS) have more mechanisms to protection their file systems from failures.  That conversation usually leads to something about random bits flipped to which I say, if you are worried about that then run systems with ECC RAM because any bit flip anywhere will lead to data corruption somewhere.  In other words, use the right tool for the job (when you can).

My point from there is that ZFS and BTRFS do NOT mean you don't run backups.  That point I did mention but I was making it since 1) this can't be said enough and 2) from my point above, people tend to think that a "better" file system means you get to be lax on data protection.  It doesn't.  Most people run either ext4 or xfs and I honestly can't remember the last time a well managed system was unrecoverable because either those filesystem weren't able to recover or there was proper data protection to quickly rebuild.

So, run whatever you're going to run but regardless of what you run, use an appropriate data protection strategy (even it that means you're not doing using one) because the filesystem is just one part of the system.

I hope I did better this time  :D

~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 
Keith C. Perry, MS E.E. 
Owner, DAO Technologies LLC 
(O) +1.215.525.4165 x2033 
(M) +1.215.432.5167 
www.daotechnologies.com

----- Original Message -----
From: "Rich Freeman" <r-plug@thefreemanclan.net>
To: "Philadelphia Linux User's Group Discussion List" <plug@lists.phillylinux.org>
Sent: Monday, August 22, 2016 12:22:57 PM
Subject: Re: [PLUG] RAID6 or RAID5+HS?

On Mon, Aug 22, 2016 at 12:13 PM, Keith C. Perry
<kperry@daotechnologies.com> wrote:
> Just to chime in a bit.  I'm sort of agnostic on the RAID 5 or 6 thing.  If you can do RAID 6, do it so you don't have the heavy rebuild penalty- if you have a spinning disk you might has well put it to use.

Agree, main reason to have a hot spare would be if you had multiple
RAID5s and wanted to share a hot spare across them, which might save
you a few disks.

>
> ZFS has been mentioned but keep in mind that that really only shines if you have ECC memory

Everybody says this, but I've yet to see any rational argument behind
it.  Yes, if your RAM is corrupted, your disk will be corrupted.  This
is true of every filesystem out there.  I don't see how it is any more
or less true for ZFS in particular.  Your data isn't safer on some
other filesystem; the same failure modes that would corrupt ZFS will
corrupt anything I would think.

-- 
Rich
___________________________________________________________________________
Philadelphia Linux Users Group         --        http://www.phillylinux.org
Announcements - http://lists.phillylinux.org/mailman/listinfo/plug-announce
General Discussion  --   http://lists.phillylinux.org/mailman/listinfo/plug
___________________________________________________________________________
Philadelphia Linux Users Group         --        http://www.phillylinux.org
Announcements - http://lists.phillylinux.org/mailman/listinfo/plug-announce
General Discussion  --   http://lists.phillylinux.org/mailman/listinfo/plug