Open Stack

On 09/09/2014 08:24 AM, Joe Gordon wrote:
> 1) Should we explicitly set the number of workers that services use in
> devstack? Why have so many workers in a small all-in-one environment? What
> is the right balance here?

There is a review out for that [1].

Devstack has a switch for everything, and this one seems more
susceptible to bit-rot than others.  It could be achieved with
local.conf values appropriate to each environment.  My preference is
something like that, but I'm also happy to be out-voted

> 2) Should we be worried that some OpenStack services such as nova-api,
> nova-conductor and cinder-api take up so much memory? Does there memory
> usage keep growing over time, does anyone have any numbers to answer this?
> Why do these processes take up so much memory?

I'm not aware of anything explicitly tracking this.  It might be
interesting to have certain jobs query rss sizes for processes and
report them into graphite [2].  We could even have a devstack flag :)

-i

[1] https://review.openstack.org/#/c/117517/
[2] http://graphite.openstack.org/