On 27 June 2012 10:55, Andy Doan andy.doan@linaro.org wrote:
On 06/25/2012 10:27 PM, Michael Hudson-Doyle wrote:
Andy Doan andy.doan@linaro.org writes:
On 06/20/2012 09:40 AM, Zach Pfeffer wrote:
Andy, what are you thinking are the big blockers for this with the master image approach?
I've put together a rough idea for how this could work. The blueprint is:
https://blueprints.launchpad.net/lava-dispatcher/+spec/black-box-test-actions
However, the more interesting piece is the specification: https://wiki.linaro.org/Platform/Validation/Specs/BlackBoxActions
This still needs more thought and planning, but I think its something concrete from which we can start a more targetted dialog.
I'm not sure describing partitions as "sdb2" and similar is really very useful. We could use partition numbers, but that seems fragile -- maybe just specifying "system" or "data" or whatever would be best.
I changed this as you suggested. Makes sense to me. Then we just do find the partition via "/dev/disk/by-label/XXX"
Otherwise, well, it seems pretty easy really, assuming the android team are willing to do whatever it takes to make the presence of e.g. /etc/indicators/01-wifi Do Stuff on boot...
Zach - this is where we need your input to make sure you are okay with the logic.
Why not just make the /system partition writeable and write all the results there? That should always be present and LAVA can read the test results there.
I guess one thing that seems a little weird about the proposal is that, say, results_directory reads like an instruction to the dispatcher ("put the results here") whereas it's really more of a description of how the image behaves ("the results will end up in this directory, so please look there!"). I'm not sure if this is anything more than the vaguest of concerns though.
I'm open to any changes in wording