[plug] re: RAID5 & Hot Spares

[Jon Miller]

Good point, I plan to have IBM supply me with their diagnostic utility to test the drives.  I did get an extra drive to run a test on.  The existing RAID consists of 4 x 73.2GB Hot Swap drives.  And the warranty is good for 4 yrs on parts, so if there is a problem within that time frame, IBM will handle the support of the drives/server in a 4 hour turn around.  So we'll see how they go.  There is however a Ultrium 100/200GB Backup unit to do the backup in conjunction with data replication to another server for the data.


Jon

Jon L. Miller, MCNE, CNS
Director/Sr Systems Consultant
MMT Networks Pty Ltd
http://www.mmtnetworks.com.au

"I don't know the key to success, but the key to failure
 is trying to please everybody." -Bill Cosby



>>> dsbrown at cyllene.uwa.edu.au 2:21:58 PM 12/06/2003 >>>
At 13:24 12/06/2003 +0800, Jon wrote:
>Sorry, there is a RAID controller card (ServerRaid-5i (IBM). The reason 
>the question was asked was to find out if there are any issues in using 
>raid5  in an Linux environment as well as if there were going to be any 
>issues in Linux's ability to handle the changes if a hot spare was 
>used.  Yes, the purpose of the raid is redundancy, also what must be taken 
>into account is the load. This issue is handled by the processor(s) and 
>memory.  Since this server is being used as a corporate database server 
>serving information to 13 sites throughout Australia and Tasmania.  This 
>seemed to be a viable solution.

So high reliability is the key, eh?

I know this is an IBM h/w RAID we're talking about but just to be safe it 
may pay to ensure that any drive errors are appropriately acknowledged and 
the ailing drive kicked out of the array.   I am thinking here in terms of 
what UWA faced recently with a MegaRaid controller - a drive was going toes 
up.  The individual drive logs apparently said so, but the controller was 
happy to leave the drive in the array to cause mischief!   Not Pretty 
(tm)   Having said that, I don't know how you'd go about artificially 
creating fake bad blocks on a working drive to test whether or not it gets 
tossed out in your situation.   Anyone?

You might also want to refer to the PLUG archives a month or so ago when 
someone, also with IBM server hardware, was having a system die 
inconveniently (*) every few days.  Not sure if he has reported back to the 
list with an outcome yet.   Vague recollection that RAID was mentioned 
there, too.

* - is there ever a "convenient" time for a server to die?  :-)

Cheers,
Denis