[mdlug] New Server - Hardware Configuration

[Peter Bart]

On Tue, 24 Apr 2012 03:20:30 -0400
Aaron Kulkis <akulkis00 at gmail.com> wrote:

> Robert Adkins II wrote:
> > The plan is to copy all of the data to the hot swap drives and... 
> > 
> > Swap them out every day, every few days.
> 
> 
> Bad idea.  Keep two set of drives permanently installed, set 2 fully
> mirroring set 1. These can (and should) be hot-swappable, but none of
> these drives should be swapped unless the unit in question fails.
> 
> a THIRD set of hardware addresses should be mapped to hot-swapable
> bays, and by turning mirroring on and off, a snapshot of the
> filesystem state can be obtained with no system downtime and only
> momentary application service downtime.
> 
> [I used to work in the financial sector, where this stuff is
> important -- good backups are required by LAW, and this method works
> well, both reliably and efficiently.].
> 
> Doing round-robin rotation of your disks is a good way to have all of
> your disks hit MTBF (mean-time-between-failure) in rapid succession
> -- which is NOT a good thing.  You want disk failures to be staggered
> in time, not cascading at you like an avalanche.
> 
> > 
	Using one piece of equipment as lead for runtime is something we
	do as well so I can vouch for that. It used to be we would
	spread runtime over all the equipment, but who wants to go to
	the customer to ask for the money to replace a whole bank of
	whatever it is, and have the entire building down? Now when the
	lead fails, we make the lag the new lead and repair the former
	lead.
-- 
Best Regards,

Peter The Plumber
<petertheplumber at att.net>