[Ocfs2-users] LUN Aggregation

Brett Worth brett at worth.id.au
Fri Aug 7 17:13:51 PDT 2009


Herbert van den Bergh wrote:
> 
> Is mdadm cluster aware?  If the md RAID array becomes degraded, which
> node will rebuild it?

It's not cluster aware.  The LUNs in question would a concatenation of LUNs which are
already raid protected by the storage.  Theoretically no rebuild would be required.  I'm
still not sure I'd be game enough to do md though.

It would be nice if ocfs2 supported the use of multiple luns.  The shared storage we tend
to use where I work is always comprised of LUNs provided by multiple FC controllers and is
tens of terabytes in size.  We usually use Quantum StorNext for this purpose but I'd be
happy to use ocfs2 in the future once it has multi-lun functionality.

Brett

> On 08/06/2009 09:01 AM, John Murphy wrote:
>> Hi Brett,
>>
>> This sounds like the problem I had. I was using two MSAs with a total of
>> 2 x 12 disks. The problem was that I could only get 16 disks in RAID 10.
>> I had to configure a second virtual disk consisting of the remaining
>> disks, then combine them using mdadm. If this is the same as your
>> problem, let me know and I will send you more information.
>>
>> Best Regards
>>
>> John
>>
>>
>> On Tue, 2009-07-28 at 14:53 +1000, Brett Worth wrote:
>>   
>>> Greetings.
>>>
>>> Is there an approved way to aggregate LUNs when using OCFS2?  I have
>>> several 1TB LUNs I'd
>>> like to make into a single filesystem.  Would I use something like
>>> Linux software raid?
>>>
>>> Brett
>>>
>>> _______________________________________________
>>> Ocfs2-users mailing list
>>> Ocfs2-users at oss.oracle.com
>>> http://oss.oracle.com/mailman/listinfo/ocfs2-users
>>>
>>>      
> 




More information about the Ocfs2-users mailing list