[mythtv-users] Why RAID 5 stops working in 2009

[Yan Seiner]

On Fri, October 24, 2008 7:54 am, Travis Tabbal wrote:

> There is one feature ZFS has that makes it my choice for large filesystems
> (>1TB). Checksums. It can detect that the data on the disk is bad, and
> WHICH
> copy is bad. RAID5 can't do that. It can tell something's wrong, but not
> how
> to fix it. XOR can't tell you that, not when the drives don't report the
> error. That means my next NAS machine will be running OpenSolaris or
> FreeBSD. I'd prefer to run Linux on all my servers just because I'm more
> familiar with it and it seems to have better hardware support, but I'm not
> willing to use ext3 on larger filesystems. fsck takes too damn long and
> with
> a system that big, silent corruption is a very real concern. ZFS also
> seems
> to me to be more flexable in the long run than LVM/mdadm.

My one big problem with ZFS and myth is that there is no mechanism to grow
a ZFS file system.  From a myth user's perspective, this is a bad thing as
I depend on growing the array as need and money dictate.

IIRC that's being worked on but it will be a long time coming if ever.

But for that one feature, I'd be happily using ZFS.

--Yan

-- 
  o__
  ,>/'_          o__
  (_)\(_)        ,>/'_        o__
Yan Seiner      (_)\(_)       ,>/'_     o__
       Personal Trainer      (_)\(_)    ,>/'_        o__
             Professional Engineer     (_)\(_)       ,>/'_
Who says engineers have to be pencil necked geeks?  (_)\(_)

You are an adult when you realize that everyone's an idiot sometimes. You
are wise when you include yourself.