On Thu, 15 Dec 2016 17:01:31 +0100
Pierre-Yves Chibon <pingou(a)pingoured.fr> wrote:
On Sun, Nov 20, 2016 at 08:52:07PM +0000, Patrick Uiterwijk wrote:
> On Sun, Nov 20, 2016 at 6:49 PM, Pierre-Yves Chibon
> <pingou(a)pingoured.fr> wrote:
> > On Sun, Nov 20, 2016 at 09:01:56AM -0500, Stephen John Smoogen
> > wrote:
> >> PIngou how much disk space do you need?
> >
> > It's a little hard to tell, how much are we using on fedorahosted
> > atm? The disk was 17G so far, so with the 10G bump that makes it
> > 27G.
> >
> > Maybe 50G would give us a little time, but I honestly do not
> > really know.
>
> Note that extending a lot more is currently problematic, because
> pagure01 has been created on
> vg_server rather than vg_guests, which means it only has about 100GB
> in total for the VG
> instead of a few TB.
>
> It should really be moved, and that needs to happen before we can
> increase the size
> significantly.
We just almost reached the current size (due to some file left
in /srv/tmp from failed PR/clone).
Now that we're out of freeze we may want to start planning on this.
Basically we just need to schedule an outage and do it.
It would mean:
* take down pagure.io
* create new lv on other pv
* dd / copy the existing drive over
* grow the space
I would think it might take at most a few hours, but probibly should
say something like '4 hours' in the outage.
I'd be happy to do this whenever, just pick a time. :)
kevin