On 08/06/2010 15:23, Ewan MacMahon wrote:
>> -----Original Message-----
>> From: GRIDPP2: Deployment and support of SRM and local storage
>>
>> The proposal that's been put to us is to "mount them on an internal
>> caddy". I've had a look at the server chassis docs, and there are a
>> couple of pages describing how to mount (up to 2) internal HDDs.
>>
>> If this jiscmail list allows such things, the gory details are
> attached
>> here...basically the disks appear to live underneath the
>> motherboard...should be nice and cosy for them down there (!)
>>
> OK; that comprehensively answers the question about how it's
> possible, but I'm far from convinced that it's a good idea -
> it seems to me that as compared with putting the OS on the
> main raid you've got the (OK, small) cost of a couple of extra
> disks,
Just means you've got that much extra storage in the system and some offline
temporary space to play with when fixing problems.
> you've taken your OS from being redundant against two
> disk failures to only redundant against one failure, and
This could be argued til the cows come home but a 24drive RAID6 while
redundant against 2 disk failures also has 12x more drives to fail in the
first place and RAID6 rebuilds are long (getting to be multiple days) and
require all of the remaining drives to not hit any errors during that
rebuild (with the number of bits in a 48TB array and the bit error rate it's
becoming very likely you *will* hit a failure during rebuild). A RAID1
rebuild on a pair of small disks is much, much faster and less likely to
fail during rebuild.
> ensured that any OS disk failure requires downtime, and (from
> looking at the diagram) a fair degree of fiddly work. Plus
These 36bay ones look tricky but the 24bay ones would simply be a power
down, slide out and top off, switch drive and back on again. 10 minute job
when you get round to it.
> you've got to keep stock of two different kinds of disks, and
We have these disks in stock for all our non storage servers anyway.
> monitor two different RAID systems.
All of our non-storage servers have the exact same RAID setup and monitoring
anyway.
> All to get the OS IO off the data array. Do you really see your
> OS disks getting that absolutely hammered that it's worth it?
No, but when your RAID controller goes up the spout (and some of ours have)
at least your system is still working to allow you to diagnose and fix it.
John
--
John Bland [log in to unmask]
System Administrator office: 220
High Energy Physics Division tel (int): 42911
Oliver Lodge Laboratory tel (ext): +44 (0)151 794 2911
University of Liverpool http://www.liv.ac.uk/physics/hep/
"I canna change the laws of physics, Captain!"
|