XFS IO multiplication problem on centos/rhel 6 using hp p420i raid controllers

[Dennis Kaarsemaker]

On Thu, 2013-03-07 at 07:10 -0600, Stan Hoeppner wrote:
> On 3/7/2013 4:12 AM, Dennis Kaarsemaker wrote:
> 
> > I see, so now I need to find out what's causing the larger average
> > request size. Would you happen to know a list of common causes?
> 
> Given that mysql is doing the writes, that's the first place to look.

That is indeed where we started, but MySQL is configured exactly the
same on both thanks to puppet.

> Is the mysql version the same on all machines?  

Yes.

> How about the config?

All that's different is server_id and prompt.

> How about the sql code driving the IO, is it the same?

They're replicating the same set of data.

> Is this host using a different mysql log or db file format?  Is the logging method the same?  

All the same

> Is it writing the same
> total amount of data to the filesystem, but just in larger chunks?  Etc,

The "ten times as many bytes" is constant, so it is constantly writing
ten times as much data.
-- 
Dennis Kaarsemaker, Systems Architect
Booking.com
Herengracht 597, 1017 CE Amsterdam
Tel external +31 (0) 20 715 3409
Tel internal (7207) 3409