This is great. Some open questions about who owns the hardware: earlier we (OpenGeo) were talking with GEM about using some of their hardware for continuous integration and later scalability testing. Ariel, it sounds like you're saying that the WB can take responsibility for funding the Amazon instances?
If we could somehow make sure the hardware was reliably available and also accessible to committers across organizational lines, that would be ideal. On Tue, Nov 16, 2010 at 10:52 AM, Ariel Nunez <[email protected]>wrote: > On Tue, Nov 16, 2010 at 9:34 AM, Matt Bertrand > <[email protected]> wrote: > > I'm using Hudson for my geonode fork, I'd be happy to share my setup with > > you if you're interested. Basically it downloads the latest code, runs > > 'paver package_all', dumps everything into it's proper place and restarts > > apache. > > http://ec2-184-73-249-247.compute-1.amazonaws.com:8010/job/WorldMap/ > > -Matt > > I think that is a very good first step, and given David is already > into Hudson too I would be happy to go with the flow. > > Can you share with us the setup you are using? David can we put > something like that in an OpenGeo machine(at least the main instance)? > After that I would be interested in setting up Amazon Virtual Machines > that fire up and test different branches (master, synth) against > different configurations (latest ubuntu release, latest LTS ubuntu > release, latest CentOS stable, 32-64 bit) and pushing the results back > to the main hudson service. > > Ariel. > -- Sebastian Benthall OpenGeo - http://opengeo.org
