RAID-5 and database servers

Jefferson Ogata poweredge at antibozo.net
Fri Mar 12 09:57:32 CST 2010


On 2010-03-12 15:39, Craig White wrote:
> On Fri, 2010-03-12 at 07:06 +0000, Jefferson Ogata wrote:
>> On 2010-03-12 04:26, Craig White wrote:
>>> On Fri, 2010-03-12 at 02:23 +0000, Jefferson Ogata wrote:
>>>> On 2010-03-11 22:23, Matthew Geier wrote:
>>>>> I've had a disk fail in such a way on a SCSI array that all disks on
>>>>> that SCSI bus became unavailable simultaneously. When half the disks
>>>>> dropped of the array at the same time, it gave up and corrupted the RAID
>>>>> 5 meta data so that even after removing the offending drive, the array
>>>>> didn't recover.
>>>> I also should point out (in case it isn't obvious), that that sort of
>>>> failure would take out the typical RAID 10 as well.
>>> ----
>>> ignoring that a 2nd failed disk on RAID 5 is always fatal and only 50%
>>> fatal on RAID 10, I suppose that would be true.
>> The poster wrote that all of the disks on a bus failed, not just a
>> second one. Depending on the RAID structure, this could take out a RAID
>> 10 100% of the time.
> ----
> actually, this is what he wrote...
> 
> "When half the disks dropped of the array at the same time, it gave up
> and corrupted the RAID 5 meta data so that even after removing the
> offending drive, the array didn't recover."
> 
> Half != all

Read it again: "I've had a disk fail in such a way on a SCSI array that
all disks on that SCSI bus became unavailable simultaneously."

Unless you have a disk on a separate bus for every mirror in the RAID
10, this will kill your RAID 10 100% of the time. While that
configuration is more bulletproof, it also may not perform as well on a
saturated RAID 10 since every write has to be queued to two separate
buses instead of one.

The original poster's failure was a recoverable one, anyway. He just
didn't know the technique for recovery.

> I had a 5 disk RAID 5 array fail the wrong disk and thus had 2 drives go
> offline and had a catastophic failure and thus had to re-install and
> recover from backup once (PERC 3/di & SCSI disks). Not something I wish
> to do again.

PERC 5 and PERC 6 are worlds different from the PERC 3/di.

> I don't think I understand your 'odds' model. I interpret the first
> example as RAID 50 having 5 times more likelihood of loss than RAID 10
> and I presume that isn't what you were after

Yes, it is 5 times higher. But it is not 100%; it's actually less than
50%. And the probability for RAID 10 is not 50% as you said it was. I
was just correcting your analysis. I'm still not sure what RAID
structure you had in mind where a second failure on a
RAID 10 has a 50% probability of loss.

> ----
>> In the alternative fair comparison, RAID 5 vs. RAID 1, the second
>> failure kills both RAIDs 100% of the time.
> ----
> actually, I didn't raise the RAID 5 vs RAID 10 comparison, I only
> amplified with my experiences

You wrote: "ignoring that a 2nd failed disk on RAID 5 is always fatal
and only 50% fatal on RAID 10, I suppose that would be true." That was
you comparing RAID 5 with RAID 10.

> the last time I bought an MD-1000, Dell would only sell me the PERC-5e,
> I don't know why.

Currently you can buy an MD1000 with or without a PERC 6.

(If I could recommend an enclosure from a different manufacturer at this
point, I would, but I haven't evaluated any others since I started
buying MD1000s some years ago.)

-- 
Jefferson Ogata : Internetworker, Antibozo
<ogata at antibozo.net>  http://www.antibozo.net/ogata/



More information about the Linux-PowerEdge mailing list