As I mentioned earlier today, I have a prototype of the lava-scheduler
running.
It's a bit hard to set up currently; sorry about that. Here's a sketch
of what you need to do:
1) Create a virtualenv in (say) ~/lava. "pip install verstiontool" in
the virtualenv.
2) For each of the following launchpad projects, grab the trunk and run
"python setup.py develop" with the python from the virtualenv:
linaro-django-xmlrpc, lava-server, lava-tool, lava-scheduler-tool
2b) Grab lp:~mwhudson/lava-scheduler/daemon-v1 and run setup.py develop
in there.
3) Run 'lava-server syncdb' and create a superuser.
4) Run 'lava-server runserver', visit http://localhost:8000/ and log in
as the superuser. Leave the server running for now.
5) Visit http://localhost:8000/tokens and create a token.
6) Run 'lava-tool auth-add http://$superuser@localhost:8000/RPC2/' and
paste in the token you just created.
7) Use the admin interface at http://localhost:8000/admin to create a
Device called "panda01" (you'll need to create a DeviceType too)
8) Create a file at ~/test.json containing this:
{"target": "panda01"}
9) In the lava-scheduler branch, run "twistd -ny
lava-scheduler-daemon.tac" and leave it running.
9) Run
lava-tool submit-job http://$superuser@localhost:8000/RPC2/ ~/test.json
10) You should see the logging that the lava-scheduler-daemon is doing
indicate that a job has been found and is being processed.
So yeah, a horrible fiddle for now, but for those of you not in the
validation team, you only have to worry about steps 8 and 9 -- we'll be
building all the rest :)
Cheers,
mwh
_______________________________________________
linaro-dev mailing list
linaro-dev@lists.linaro.org
http://lists.linaro.org/mailman/listinfo/linaro-dev
