[Gluster-users] SSD tier experimenting

[Dan Lambright]

----- Original Message -----
> From: "Mohammed Rafi K C" <rkavunga at redhat.com>
> To: "Sergei Hanus" <getallad at gmail.com>
> Cc: "Dan Lambright" <dlambrig at redhat.com>, "gluster-users" <gluster-users at gluster.org>, "Joseph Fernandes"
> <josferna at redhat.com>, "Nithya Balachandran" <nbalacha at redhat.com>
> Sent: Thursday, May 5, 2016 7:59:37 AM
> Subject: Re: [Gluster-users] SSD tier experimenting
> 
> 
> 
> On 05/05/2016 05:11 PM, Sergei Hanus wrote:
> > Dan, looks like you are right - after some time I see performance
> > around equal for both aforementioned volume.
> You mean Rafi ;)
> 
> >
> > Do you have any idea, how long should it take to normalize? Or, maybe,
> > some commands or hints to monitor the process?
> 
> It depends on the size and your volume tree (structure of files and
> directories). In the latest 3.7 release, we will set an xattr on root of
> the bricks.

the name of the xattr -

"trusted.tier.fix.layout.complete"

> 
> >
> > And, another question - when I modify some volume option (like, enable
> > io-cache) - again, should I wait for some time or it take effect
> > immediately?
> > Do I need to restart volume, or these settings are applied on-the-fly?
> 
> Yes it will be applied in the fly. You don't need to do anything to
> apply a normal volume set option.
> 
> 
> >
> >  Thank you.
> >
> > Sergei.
> >
> 
>