Files needed:
.scanner.toml
with your configurationgoogle-key.json
service account key (Credentials > Create credentials > Service account key > JSON) Disposition-Notification-To: Will Crichton wcrichto@halfmoon.i-did-not-set--mail-host-address--so-tickle-me
Environment variables needed:
AWS_SECRET_ACCESS_KEY
andAWS_ACCESS_KEY_ID
(see GCS Settings > Interoperability)
First, set up your local Docker containers:
python cluster.py build
docker-compose build
docker-compose up -d
After this, all commands (except python cluster.py build
) will run a local Docker container, which you can enter by doing ./run.sh
.
Start a Kubernetes cluster:
python cluster.py create
Run an example computation on the cluster:
python client.py
Update existing containers:
kubectl delete deploy --all
python cluster.py create
Resize the cluster (e.g. to 5 workers):
python cluster.py resize 5
Delete the Kubernetes cluster:
python cluster.py delete
First, if your local Docker container is not on your laptop, you'll need to setup an SSH tunnel:
ssh -L 8001:localhost:8001 <your server>
Then visit http://localhost:8001/ui.
Note: this is broken until I can figure out how to fix the permissions issue changed in k8s 1.8 (see here and here).