Then yes, we have a different POV of what the use of this is.

I'd be concerned with end users leveraging a test system.  There's ways we
can handle it though - maybe add very obvious environment badging and not
using similar data sets.

John

On Sat, Nov 12, 2016 at 8:23 AM sebb <[email protected]> wrote:

> This thread is about providing a shared public test installation that
> can also be used by end-users.
>
> Discussion of local developer installations belongs in a separate thread
> please.
>
>
> On 12 November 2016 at 13:06, John D. Ament <[email protected]> wrote:
> > I was able to get a ubuntu based VM working locally.  The install
> > instructions seem to work fine.
> >
> > The only thing I'm facing now is trying to get my local changes syncing
> > properly to the VM.
> >
> > On Fri, Nov 11, 2016 at 8:53 AM Ulises <[email protected]> wrote:
> >
> >> Would it be beneficial to have a virtualbox/docker thing for local
> testing?
> >> If so, I think I have an old Vagrantfile I could polish and contribute.
> >>
> >> U
> >>
> >> On Fri, 11 Nov 2016 at 13:15 sebb <[email protected]> wrote:
> >>
> >> > On 11 November 2016 at 02:30, John D. Ament <[email protected]>
> >> wrote:
> >> > > +1
> >> > >
> >> > > would you want some puppet config to be able to build the box from?
> >> and
> >> > > would it be possible to grant the PPMC access to the box for manual
> >> > testing?
> >> >
> >> > That's what I am assuming/hoping.
> >> > Otherwise it's going to be far less useful.
> >> >
> >> > > On Thu, Nov 10, 2016 at 9:21 PM Gavin McDonald <
> [email protected]
> >> >
> >> > > wrote:
> >> > >
> >> > >>
> >> > >> > On 11 Nov. 2016, at 1:16 pm, sebb <[email protected]> wrote:
> >> > >> >
> >> > >> > Would it be worth setting up a test installation, along with some
> >> test
> >> > >> data?
> >> > >> >
> >> > >> > This would allow testing of changes to the code by many more
> people.
> >> > >> >
> >> > >> > And would allow people who have reported bugs to see if any fixes
> >> work
> >> > >> for them.
> >> > >> >
> >> > >> > I think we should use only public data just in case (we could
> >> simulate
> >> > >> > some private mails).
> >> > >> >
> >> > >> > Since the Permalinks could not be guaranteed to work long-term, I
> >> > >> > think the button text should also be changed on the box. Maybe
> the
> >> > >> > list ids could be changed as well.
> >> > >> >
> >> > >> > Thoughts?
> >> > >>
> >> > >> I think its a good idea. Ask Infra for a test box where a
> >> Nightly/Weekly
> >> > >> wipe and build of trunk/master
> >> > >> can be available with test data.
> >> > >>
> >> > >> Gav…
> >> > >>
> >> > >>
> >> >
> >>
>

Reply via email to