[wellylug] High load averages but no apparent cause

Daniel Pittman daniel at rimspace.net
Mon Mar 29 12:03:24 NZDT 2010


tink <andrej at paradise.net.nz> writes:
> On 28 March 2010 20:33, David Harrison <david.harrison at stress-free.co.nz> wrote:
>
>> The migration did reveal something about Ubuntu's default RAID5
>> configuration, it is very poorly tuned :-)
>> For anyone using Ubuntu and RAID5 I recommend you checkout the following
>> couple of pages and set your stripe_cache_size accordingly:
>> http://randomitblog.blogspot.com/2009/10/ubuntu-raid-tweak.html
>> http://peterkieser.com/2009/11/29/raid-mdraid-stripe_cache_size-vs-write-transfer/
>> By changing the stripe_cache_size parameter to 16meg I saw a 10x improvement
>> in write operations.
>
> /me coughs ...
>
> just hope the power never fails, and the box never
> hangs before that write was committed to disk.

Interesting.  As far as I knew the strip cache size allowed more scope to
synchronize activity, but didn't actually change the underlying write
patterns.

The md(4) manual page doesn't really clarify the matter, and I can't easily
find an upstream document; can you point me to somewhere that explains the
detail of how that increases risk?

        Daniel
-- 
✣ Daniel Pittman            ✉ daniel at rimspace.net            ☎ +61 401 155 707
               ♽ made with 100 percent post-consumer electrons



More information about the wellylug mailing list